Selected Proceedings from the 12th International

Transcription

SELECTED PROCEEDINGS FROM THE 12TH INTERNATIONAL CONGRESS ON PROJECT ENGINEERING ZARAGOZA, JULY 2008 Selected Proceedings from the 12th International Congress
on Project Engineering, AEIPRO, 2009
Edited by AEIPRO (Spanish Association of Project Engineering)
ISBN–13: 978–84–613–3557–2 The works here published are "open access literature". This means that it is freely available on the public internet, permitting any user to read, download, copy, distribute, print, search, or link to the full texts of these articles, or use them for any other lawful purpose, without financial, legal, or technical barriers other than those inseparable from gaining access to the internet itself. The only constraint on reproduction and distribution, and the only role for copyright in this domain, should be to give authors control over the integrity of their work and to ensure that they are properly acknowledged and cited. The Publishers are not responible for any statement made in this publication. Data, discussion, and conclusions developed by authors are for information only and are not intended for use without independent substatianting investigation on the part of potential users. Opinions expressed are those of the Author and are not necessarily those of the AEIPRO (Asociación Española de Ingeniería de Proyectos). 2
INDEX INTRODUCTION..........................................................................................................................7 Creation of the Scientific Committee ........................................................................................8 Evaluation of Submitted Papers ...............................................................................................9 TOPIC AREA: FOOD AND FOREST INDUSTRIES .................................................................13 Values of efficiency in galician dairy exploitations with different productive characteristics
Irimia, S; Resch, C. and Álvarez, C.J. ..............................................................................15 Technical-economic analysis of the pig exploitations in the Autonomous Community of
Galicia (Spain)
Marey Pérez, M.F.; Sobrado Castro, O.; Riveiro Valiño, J.A. and Álvarez López, C.J. ...27 TOPIC AREA: LAND MANAGEMENT, URBANISM, ARCHITECTURE AND CIVIL
ENGINEERING ..........................................................................................................................35 Strategies for sewage treatment in rural areas
Muiño, D.; Cancela, J.J.; Marey, M.F.; Neira, Xan X. and Cuesta, T.S. ..........................37 Automated vectorization of urban areas by using visible and infrared spectrum images
De Cos Juez, F.J.; Prendes Gero, M.B.; García Fernández, R. and Alonso A. ...............47 Model for agricultural production planning
Cardín, M. and Álvarez, C.J. ............................................................................................58 Definition of a methodology for fostering integrated collaboration in small-to-medium
companies of the housing development / construction sector
Santos Fonseca, S. ..........................................................................................................71 Analysis of the factors that condition the expansion of a coastal village through cellular
automata simulation
García, A.M.; Santé, I.; Crecente, R. and López, F.J. ......................................................82 Development of an experimental procedure to anaylse the temperature field in the welded
joint of structural stainless steel
Miguel, V.; Estrems, M.; Martínez, E. and Martínez, A.....................................................93 Validation of a model of typification of agrarian exploitations by technicals of discriminant
analysis
Riveiro, J.A.; Marey, M.F. and Álvarez, C.J....................................................................104 TOPIC AREA: PRODUCT ENGINEERING .............................................................................119 Applicability of the B-Cube model to the services sector
Chulvi, V. and Vidal, R....................................................................................................121 3
Paradigms of exploration and paradigms of exploitation. The artificial disc case
Barberá-Tomás, D.; De los Reyes, E. and Jiménez-Saéz, F. ........................................132 Models for airbag simulation in vehicles adapted for disabled drivers
Masiá, J.; Eixerés, B. and Dols, J...................................................................................142 Cylindrical parallel gear trains design by means of genetic optimization
Sánchez Caballero, V.; Colomer Romero, R.; Pla Ferrando, A.V. and Martínez
Sanz, S. ..........................................................................................................................150 An ontology-based approach to integrating Life Cycle Analysis and Computer Aided
Design
Abad-Kelly, J.; Cebrián, D. and Chulvi, V.......................................................................161 Design process for an electromagnetic power brake with ABS function: Corporate
knowledge management
Villanueva, P.; Martínez, P. and Pérez, A. .....................................................................173 TOPIC AREA: ENVIRONMENT ..............................................................................................185 Analysis of eco-efficiency progress in industrial production
Fernández-Viñé, M.B.; Gómez-Navarro, T. and Capuz-Rizo, S.F. ................................187 Land occupation and transformation by Spanish roads
Garraín, D.; Vidal, R. and Franco, V...............................................................................198 The noise impact category in Life Cycle Assessment
Garraín, D.; Franco, V.; Vidal, R.; Moliner, E. and Casanova, S....................................211 Creation of Ozone maps to register the evolution of this pollutant using geostatistical
techniques
López-Rodríguez, F.; Moral, F.J.; Valiente, P. and Pinilla, E. ........................................222 Spatial multiscale model for the forest areas of planning: An example in the region of
Eastern Marine (Galicia NE)
Díaz Varela, E.R.; Rodríguez Vicente, V. and Marey Pérez, M.F. .................................231 Analysis of packaging waste generation and management in Spain
Paneque, A.; Cháfer, C.; Pacheco, B.; Capuz-Rizo; S.F. and Hortal, M. .......................241 TOPIC AREA: ENERGY EFFICIENCY AND RENEWABLE ENERGIES ...............................253 Urban transport project by hydrogen technologies for international exhibition
EXPOZARAGOZA 2008
Montaner, P. ...................................................................................................................255 An example of bioclimatic construction.The building PETER
López-Rodríguez, F.; Segador, C.; Cuadros, F and Mena, A. .......................................261 4
TOPIC AREA: INFORMATION AND COMMUNICATION TECHNOLOGIES AND
SOFTWARE ENGINEERING...................................................................................................273 Software for determining dripline irrigation pipes using a Pocket PC
Molina Martínez, J.M. and Ruiz Canales, A. ..................................................................275 Ontologies and FBS framework
Cebrián-Tarrasón, D. and Vidal, R. ................................................................................281 Development of project based learning techniques by the implementation of Information
Technologies
Mesa, J.M.; Villanueva, J.; Rodríguez, V. and Rodríguez, F..........................................292 Techniques of remote detection in precision viticulture
Beltrán, P. and Montesinos, S. .......................................................................................299 TOPIC AREA: RURAL DEVELOPMENT AND COOPERATION PROJECTS .......................307 The European Regional Policy, keys for the success: Analysis of the Spanish experience
Regalado, J.; Díaz-Puente, J.M. and Figueroa, B. .........................................................309 Taking into account food and nutrition security in the measurement and assessment of
human development
Afonso Gallegos, A. and Trueba Jainaga, I....................................................................319 Selecting and assessing fair trade suppliers using multicriteria techniques
Cobo Benita; J.R.; Ortíz Marcos, I. and Mataix Aldeanueva, C......................................328 TOPIC AREA: PRODUCTION PROCESS ENGINEERING ....................................................341 Development of a water treatment plant for a rolling-mill train
Martínez, S. and Gaicoechea, I. .....................................................................................343 Workplaces redesign in a food industry to minimize musculoskeletal injuries in workers
Planas, A.E.; Ducun, M.; Tomás, J.A. and Azcona, I. ....................................................356 TOPIC AREA: PROJECT MANAGEMENT.............................................................................365 Analysis of decision-making models for project management
Poveda-Bautista, R.; García-Melón, M. and González-Cruz, M.C. ................................367 R&D&I project management in Spanish universities
Varas, M. and Caamaño, J.............................................................................................378 Construction risk management: Application and education
El-Dash, K. .....................................................................................................................390 TOPIC AREA: TRAINING IN PROJECT ENGINEERING .......................................................405 5
Support to creativity in the semester project using TRIZ
Aguilar-Zambrano, J.; Valencia, M. and González-Cruz, Mª. C. ....................................407 Project management teaching in engineering higher education: A new perspective for
developing competencies
De los Ríos, I.; Ortíz, I. and Díaz-Puente, J.M. ..............................................................418 6
INTRODUCTION
The Spanish Association of Project Engineering is pleased to publish the following selection
of the best papers presented at the 12th International Congress on Project Engineering. After
having organized an annual Congress with an array of universities over the last 15 years —
first at the national and then at the international level— by the end of 2007 the AEIPRO
Directive Board decided to establish a Scientific Committee to evaluate the papers presented
at the Congress. The Scientific Committee has also chosen the works to be published in the
selected proceedings of the Congress. The procedures to establish the Scientific Committee
and evaluate the submitted papers are explained below in detail. It is the hope of the
Committee that the compiled works contribute to the improvement of project engineering
research and improve the transfer of results to the job of Project Engineers.
The Spanish Association of Project Engineering (Asociación Española de Ingeniería de
Proyectos — AEIPRO) is a non-profit organization founded in September 1992. It is an entity
for the professionalization of Project Engineering with the following goals: to facilitate the
association of scientists and professionals within the Project Engineering area; to serve as a
tool for improving communication and cooperation amongst these professionals; to improve
experts’ knowledge in the different fields of Project Engineering; to promote the best
professional practices in these fields; to identify and define the needs that may arise in the
every day development of these activities; and finally, to adopt positions in order to orientate
society when faced with differences within the fields of action. At present it is a National
Association within IPMA (International Project Management Association), an international
association that brings together more than 45,000 members and represents about half a
million professionals from 47 countries.
From its beginnings, AEIPRO has been holding an annual Congress with growing concern
for the improvement of the evaluation process of the presented papers. This consideration
has led to the establishment of a scientific approach capable of adopting the evaluation
criteria of the scientific activity. These criteria, specific to each scientific field, were defined by
7
the Presidency of the CNEAI1 in its Resolution of 11/11/2008. The key issues in the area of
Project Engineering are presented in subfield 6.3. In paragraph 3.b) of the mentioned
subfield, one of the relevant types of contributions is:
“The works published in the proceedings of congresses that are peer reviewed, whenever
these proceedings serve as a vehicle to knowledge diffusion similar to prestigious
international journals”.
Taking the preceding statement into consideration, during the 12th International Congress of
Project Engineering, held in Zaragoza in July of 2008, a scientific procedure was developed
so that the works published in the proceedings of this Congress would meet these
requirements. Tasks to create a Scientific Committee and design the action procedures for
the evaluation of the presented papers at the Congress were developed in order to meet
these goals. The results of these tasks are described in the following sections.
Creation of the Scientific Committee
AEIPRO gathers prestigious Project Engineering professionals as well as university
professors involved in Project Engineering and related areas. The first step in creating the
Scientific Committee was to agree on criteria of excellence for the two groups that are to be
represented in the Committee. For university professors to be considered, they must have at
least one Research Incentive Award (known as a Sexenio) granted by the CNEAI after a
positive evaluation of their research activity over a six-year period.
The criteria for a professional is to have a Certification in Professional Competences level C
according to the criteria established by IPMA for the international Certification system in
Project Management.
The Scientific Committee then had to be formed by professionals and university professors
that meet these requirements of excellence, and are willing to assume the commitments as
members of the Committee. These commitments include the participation in the evaluation of
the presented papers and selection of the works that will be awarded by the Committee each
year. Additionally, the Scientific Committee members are expected to demonstrate
1
Comisión Nacional Evaluadora de la Actividad Investigadora (National Commission for the Evaluation of
Research Activities) in Spain. The Royal Decree 1086/89, of August 28th, concerning the retribution of professors
defines two independent systems to evaluate teaching and research in Spain. The first is defined as an exclusive
competence of the University, while the evaluation of research is the responsibility of the State entrusted to the
CNEAI. The Commission’s role is to carry out the evaluation of the research activities of university professors and
the scientists within the Spanish National Research Council (Consejo Superior de Investigaciones Científicas —
CSIC). Application for a Research Incentive Award (known as a Sexenio), requiring an evaluation of research
activity over a six-year period according to the requirements of the Decree 186/1999 of August 28th and its
following modifications, is voluntary. The purpose of providing benefits for productivity is to improve the quality of
academic research activities, along with the diffusion of the work on a national and international level. CNEAI
applies the advice of scientific community members; and for each evaluation a committee of advisers is created
according to 11 scientific fields.
8
confidentiality during the evaluation process. Taking into account the described criteria and
commitments, the Scientific Committee of the 12th International Congress of Project
Engineering was established as detailed in the following table.
Members of the Scientific Committee
in the 12th Congress of Project Engineering 2008
Role
Name
Institution
President
Ignacio Trueba Jainaga
Universidad Politécnica de Madrid
Carlos José Álvarez López
Universidad de Santiago de Compostela
José Luis Ayuso Muñoz
Universidad de Córdoba
Juan Luis Cano Fernández
Universidad de Zaragoza
Salvador Capuz Rizo
Universidad Politécnica de Valencia
Adolfo Cazorla Montero
Manuel de Cos Castillo
Eliseo Gómez-Senent Martinez
Jesús Guillén Torres
Fernando López Rodriguez
Universidad de Extremadura
Jesús Martínez Almela
SELCO MC. Ingeniería
Ángel Mena Nieto
Universidad de Huelva
Joaquín Ordieres Meré
Universidad de la Rioja
Isabel Ortíz Marcos
Ignacio de los Ríos Carmenado
Rosario Vidal Nadal
Universidad Jaime I
Rubén Rebollar Rubio
Vocals
Secretary
The members of the Scientific Committee were organized into different topic areas within the
Congress depending on their curricula. Each member of the Committee participated in three
areas. The topic areas represented in the AEIPRO annual Congresses are: At-1, Food and
Forest Industries; At-2, Land management, Urbanism, Architecture and Civil Engineering;
At–3, Product Engineering; At–4, Environment; At–5, Energy Efficiency and Renewable
Energies; At–6, Information and Communication Technologies and Software Engineering;
At–7, Rural Development and Cooperation Projects; At–8, Production Process Engineering;
At–9, Project Management; At–10, Training in Project Engineering.
Evaluation of Submitted Papers
A blind peer–review process was established to ensure independence between evaluators
and those evaluated. The complete action protocol is presented in the following chart.
9
Paper received
Corrections made
Format Revised
NO
Correct
Returned with
comments
YES
Paper made anonymous
Sent to evaluators
Evaluators read paper
Evaluation report filled out
Report sent to secretary
Report sent to author
Author reads report
Author makes corrections
YES
Accepted
End of
process
NO
Corrections sent
Substantial
corrections
needed?
NO
Secretary compares with
reports
YES
Evaluators compare with reports
Correct
NO
Notify author
YES
Update database
Send confirmation to author
End of process
10
Implementing this protocol, 100 submissions that scored more than 80 points in the peer–
review process were selected at the end of the 12th Congress of Project Engineering in
Zaragoza. The authors of the selected works were contacted by the AEIPRO Directive Board
and invited to undergo a second peer–review process and —if passing this second review—
to have their works translated into English and published online.
Authors of 74 selected papers expressed interest to participate in the second peer–review. In
order to evaluate these, additional criteria needed to be defined — in addition to the original
parameters used to define the Scientific Committee— to select a Subcommittee for the
subsequent peer–review. The members of this Subcommittee were required to have at least
two Research Incentive Awards (Sexenios) granted by the CNEAI or to hold a level A or B of
the IPMA certification. The selected members of the Subcommittee are presented in the
following table.
Members of the Scientific Subcommittee for the second peer–review
in the 12th Congress of Project Engineering 2008
Role
Vocals
Name
Institution
José Luis Ayuso Muñoz
Universidad de Córdoba
Salvador Capuz Rizo
Adolfo Cazorla Montero
Manuel de Cos Castillo
Eliseo Gómez-Senent Martínez
Jesús Guillén Torres
Fernando López Rodríguez
Jesús Martínez Almela
SELCO MC. Ingeniería
Ignacio Trueba Jainaga
The procedure followed for the second peer–review by the Scientific Subcommittee
coincided with the previous protocol and resulted in the selection of 48 papers. The authors
of the selected works were then requested (and most of them did) to translate their works
into English in order to publish them on line, both by AEIPRO and IPMA. The following
publication is the final result of the entire process and includes 37 selected, peer-reviewed
papers.
11
Area: Food and Forest Industries
TOPIC AREA: FOOD AND FOREST
INDUSTRIES
13
VALUES OF EFFICIENCY IN GALICIAN DAIRY EXPLOITATIONS
WITH DIFFERENT PRODUCTIVE CHARACTERISTICS
Irimia, S.
Resch, C.
Álvarez, C.J.
University of Santiago de Compostela
Abstract
Based on the data from 2007 Analysis of the European Dairy Farmers, the state of labour
productivity (efficiency), measured in kilograms of milk per hour of work, are in Spain at 120,
while the average for the countries of the Old EU Countries a little more than 180. These values
are even worst in the Autonomous Region of Galicia. In this paper we describe the critical
factors that limit efficiency taking into account three differential points with regard to the
transactions carried out by labour on farms, food, milking and cleaning in 11 selected dairies
with different characteristics. It should be noted that regardless of measuring time employees
in the activities it has been determined the adequacy of the same, in order to obtain objective
results. As overall results, we can conclude that food is more efficient when done through
automated means and less efficient farms are those that perform grazing. Milking reaches
higher values in the presence of waiting rooms. While cleaning is more appropriate in the
exploitations with emparrillado beds and rubber.
Keywords: Productive efficient, dairy, Galicia
1. Introduction
Throughout the past decades, the dairy Galician sector, NO of Spain, underwent a process of
rationalization with two main features, first the reduction of the number of exploitations, a result
of the entry in 1986 of Spain in the EU, which led to many exploitations leaving their activity;
and secondly it was increased the number of specialized exploitations in milk production and
consequently increased its size. In Table 1 can be appreciated the magnitude of this structural
change.
15
1993
2003
Ranges
Nº
exploitations
Nº dairy
Nº dairy
cows
Nº
exploitations
1-2
28.200
39.840
1.336
2.270
3-4
12.693
42,960
1.035
3.562
5-9
17.230
114.009
4.605
32.356
10-19
12.383
163.791
6.522
88.750
20-29
3204
74.527
3.490
82.597
30-49
1502
51.142
2.753
99.987
>50
193
15.352
880
61.078
Total
75.405
501.621
20.620
370.600
cows
Table 1: Distribution of dairy cattle exploitations in Galicia in 1993 and 2003 (IGE, 2005).
From the above data it is showed that despite the dramatic reduction in the number of farms,
centered mainly in the smaller dimension, there was an increase in milk production. Also, we
can point out that the number of dairy cattle also fell. These two factors indicate that the dairy
sector which presents a dynamic that clearly looks for an improvement in productive efficiency.
If we define efficiency as the rational use of the means to achieve a predetermined objective.
Efficiency is the requirement to avoid or cancel costs and errors, which will provide us the ability
to achieve the planned objectives and goals with minimal available resources and time, making
their optimization.
In this regard the improvement of the productive efficiency of the bovine exploitations of milk
integrates various factors, quality of food, genetics improvement and productivity of labor,
because this affects the economic benefits (Richards & Jeffrey, 2000). In this work we consider
the improvement of productive efficiency from the increase of productivity of the workforce, as
at present according to previous studies of this team, many of the exploitations in Galicia, are
not profitable, or rather they are profitable when we do not consider the labor of farmer and / or
help family that uses the exploitation and which normally are not counted as production costs
(Richards & Jeffrey, 2000).
During the years 2006 and 2007 it was made a survey to a representative sample of Gallego
Dairy Sector (Riveiro et al., 2008), to determine the status of productive efficiency. It
interviewed 106 dairy exploitations of Galicia, selected for their productive typology, size and
geographic location (Alvarez, et al., 2007), obtaining the values in Table 2.
In the survey was asked to farmers for the time spent on the most significant operation,
particularly in the milking, feeding and cleaning. Subsequently, the values have been compared
with milk production, resulting productive efficiency values expressed in kilograms of milk per
worked hours (kgl / ht).
16
Efficiency (Kgl/h)
Cows
Kgl/cows
Milking
Food
Cleaning
Total
10 a 24
19
5767
102
114
374
38
25 a 39
30
6989
174
332
685
84
40 a 54
45
7422
245
389
1568
119
55 a 69
60
7565
234
474
1424
128
More of 69
78
7073
268
953
2332
157
Average
41
6895
193
405
1116
97
Table 2: Result of efficiency in the tasks of milking, feed and cleaning.
This information has helped to place Galicia in the set of countries within the European Union
(EU), as we see in Table 3.
Country
Effectiveness of work(Kgl/h)
Sweden
190
Denmark
275
Ireland
195
UK
210
Netherlands
230
Belgium
145
Germany
200
France
155
Switzerland
80
Spain
120
Italy
120
Poland
90
Czech
Republic
145
Slovakia
140
Hungary
60
EU (new)
95
EU (former)
190
Table 3: Productive efficiency in different European countries (EDF-European Dairy Farmers, 2007).
17
As we can see the status of the efficiency of Galicia (97 kgl / ht) is very far from the Spanish
average (120 kgl / ht), and even more of the old EU countries (190 kgl / ht).
Different authors have studied possible actions related to the organizational improvement of
production systems of dairy cattle in different areas (Kompas & Che, 2006; Haghiri et al., 2004).
In these works it is indicated the need to consider the relative importance of different used
inputs, the effects of different technology in the exploitation and work procedures.
For over a decade there are multiple systems to determine the effectiveness of a dairy farm and
as to include its result in the characterization of exploitations (Coelli, 1995). Procedures that
allow to identify the exploitations more efficient (Bravoureta and Rieger, 1991), ie, wich show
that a lower cost. Which may or may not be related to the size of the farm or the used
technology (Romain and Lambert, 1995). These systems define procedures to identify the
more efficient farms and to characterize exploitations.
More recently it is affirmed the use of other systems for studying the efficiency of dairy farms,
information systems (Grossi & Freitas, 2002) to monitor exploitations in search of productive
efficiency. Or the use of "Data Envelopment Analysis” (DEA) for this purpose.
Finally highlight the work of this team, in an attempt to obtain information on the results of the
106 surveyed exploitations, and due to the limited correlation of results given the wide
variability of typologies and size ranges, sought an approach applying neural networks that
related variables of the productive system with the productive efficiency of the exploitation
(Alvarez et al., 2008).
In this article, within the overall objective of improving the efficiency of dairy cattle farms in
Galicia, we seek the genuine determination, that is measured on the farms, of the efficiency of
the workforce.
It is necessary to consider the value of 97 kgl / ht, wich is provided previosly, and was based on
survey of 106 representative farms of Galicia, where the values were based on estimates of the
farmers. By contrast, in this case the values will be determined performing the direct
measurement of times of the different operations in their own farms.
The work also includes the designed specifications to train the system of search of information,
the definition of the number of repeated measurements, the consideration of operations to
perform and finally some preliminary attempts of determination of the key variables of the
productive system and its relationship with efficiency.
2. Methodology
2.1. Choice of sample
To make the study randomly were selected 11 dairy farms in the sample of 116 defined above,
in proportion to the number of exploitations in each of the defined ranges.
Our intention was to obtain a representative sample of different typologies of farms, covering
the different characteristics of productive employees systems.
18
2.2. Interview to the farmer through a survey
Our intention was to take into account the one hand, three different aspects regarding the
operations carried out by the workforce on dairy farms: the tasks of cleaning, feed and milking,
assessing the cleanliness of each of those areas, and secondly, how long do they lead.
The exploitations were subjected to an initial visit where a preliminary survey was developed
with the aim of characterization. It was sought information regarding the number of cows in
production, and total milk production and per cow, the cleaning used system (slatted, by
waterjet or automated with earth scoop), the milking system (in place or room) and the feeding
system (manual or truck mixer), as well as the costs and benefits involved in the productive
process. It also will provide information on the constructive parameters of the exploitation
shape, dimensions and materials used in the different units.
The interviews were conducted directly to the workers of the exploitation (owners), which
introduced a number of drawbacks: the limited available information on the actual costs of
operation (not being independent from the family home), and on the other hand, the mistrust of
many of them to provide actual data (for fear of penalties for excess production and to maintain
a disrepute between them).
2.3. Data capture
This task was in the direct measurement of time, used by farmers in the different tasks, which
was followed the following routine for that information was comparable.
The measure of time of milking was considered since the moment in which was introduced the
first cow in the waiting room, until the milking room was ready for the next milking of the day.
For the measure the spent time in the food was taken from the time to preparate the food until it
was finished the distribution of food for livestock.
Regarding the cleanliness of the exercise yard was seen with ordinary character that wich was
done manually; if the cleaning was done by mechanical means it was considered that the spent
time on cleaning was zero.
In the cleaning of the beds was considered the spent time to clean or accommodate the beds
The spent time to clean the feeders and drinkers it was considered the spent time in these
operations.
The above measures were repeated three times, on each farm, ignoring the first half due to the
influence that caused "novelty" of taking measurements in the productive routine and it was
obtained the average of the other two.
It is important that all operations were performed in days of the week than either Friday or
Monday, after to check that there were no distinct or specific routines of work in the rest of the
weekly calendar.
2.4. Desk work
The obtained information was standardized, because the measurements and observations
troop point values, and our desire was to reflect the values of each year. For this, we consider
the productive system in the area, with concentration of births in spring and lactations of 305
days, taking into account changes in the work of milking and food.
19
3. Results
The studied sample had exploitations from 12 to 119 cows in production, where only 5 had a
waiting room for milking, of which, less one square, the rest had a rectangular shape. There
were 4 milking system in circuit and different types of milking room 1 in parallel, another in
tandem and 5 in thorn, of which only two had automatic withdrawal of teatcups.
Cleaning systems are specified in 6 slatted, 4 earth scoops and 1 exploitataion that cleaned
with water by hand. Regarding the assessment of the state of cleanliness changed between 2
and 4, in a range of 0 to 5, emphasizing the highest level of the cleaning the earth scoops and
manual cleaning with water. As for the cleaning of bed was performed in all cases by hand,
emphasizing its cleaning (value of 4) which used rubber, straw and sawdust us material of rest
of livestock.
As for the food systems, 8 used self cars and 3 performed the operation manually.
It is noteworthy that there was no significant correlation between the different systems and the
level of production or in the quality of this.
Regarding to efficiency was obtained as standardized average results the values that are
shown in Table 4.
EFFICIENCIES (kgl/ht)
nº cows
average 43
Milk
(kgs)
Milk/cow
Milking
Cleaning Food
Total
361627
8249
315
399
113
361
Table 4: Average values of the studied sample.
Comparing these data with estimates of Table 3, we note that the average number of cows of
the exploitations are practically equal in value, while in terms of milk production per cow, we are
significantly better, which shows the goodness of the selected sample in the quality of the
tested exploitations.
In terms of values of overall efficiency we are somewhat below the Spanish average of 120 to
113 kgl / ht. As for the reference value of Galicia, in Table 2, we are just over approximately
16%, which meaning that the estimates are exaggerated by the farmers in this ratio compared
with measurements. Regarding the differentiated values of efficiency according to tasks, we
have that the milking shed actual measurements in the estimates of farmers, at 63%; while that
in the cleaning values are much higher than the estimates (over 300%) and in food values are
similar values between the measures and estimates.
The previous consideration indicates that estimates of the farmers are pessimistic regarding of
milking and overly optimistic for cleaning, but given the impact of these tasks on the total, the
values of overall efficiency are quite similar.
The operative obtained procedure to determine the overall efficiency is based on measurement
of the milking values which are considered three activities, preparation, milking and cleaning of
milking room; in the cleaning was considered whatever the cleaning of coutyards, beds,
feeders and drinkers, and in terms of food was determined operations of preparation of silos
and forage, preparation, and preparation of the ration and tapping.
20
In following Table 5 there are the overwall data of studied exploitations including the different
efficiencies.
Efficiencies(Kgl/h)
nº
cows
MIlk (kg)
MIlk/cow
Milkink
Cleaning Food
total
ABADÍN
12
110126
9177
114
91
114
35
GRUEIRA
28
200000
7143
179
274
82
47
LUDRIO
28
230000
8214
238
252
503
98
MONDRIZ
35
200000
5714
94
103
109
34
SAA
43
365000
8488
395
750
280
135
PICATELOS 48
503000
10479
495
435
550
163
FRAIALDE
58
400000
6897
328
263
200
84
LEA. POL
58
500000
8621
656
1174
1093
304
BAILLE
63
657000
10429
446
1286
1292
263
LOMBAS
67
456000
6806
309
625
598
154
VILAFRIO
119
1241000
10429
1163
785
1017
321
MÁXIMUM
119
1241000
10479
1163
1286
1292
321
MÍNIMUM
12
110126
5714
94
91
82
34
Table 5: Values of efficiency measured in the considered exploitations.
Facing the values of total efficiency and in number of cows (Figure 1), as was expected, we
obtain a regression line with a very limited correlation. But if we divide it in two portions, we
can say that it adjusts of more suitable way for exploitations with fewer than 50 cows, which
may be representative, while for older exploitations is totally useless.
3.1. Efficiency in the milking
If we consider of independent manner the activities we have that in the milking, it is observed
that the more efficient exploitation is wich has the largest number of cows, although in the
range of 50 cows, there are exploitations that are above of the average and others not, this is
because in this range there are exploitations with circuit and other rooms, and logically the
milking rooms are the most efficiency.
Preparation: We emphasize that the devoted time to the preparation of milking is around 1020%, and the exploitations that devote less time to this task are those that have waiting room,
and within these are more efficient those who have pushed, because it prevents the farmer
leaves the room to introduce to lots of milking.
21
Milking: It is the activity which devotes more time, around 60-70%. In this case there is clear
evidence that the spent time on this task is directly proportional to the number of cows.
Although we must stress the exploitations that milk few cows, do not perform this task in a
much less time, due to the milking system is not properly designed. Logically it was noted that
the exploitations with milking rooms are more efficient than those that use circuit.
Cleaning of the milking rooms: The spent time on this task is between 10-30%, although we
have not clear evidence of why more time is devoted in some exploitations than other,
however it is stressed the efficiency of exploitations that have a larger number of cows. We can
stress that are a little more efficient those with milking room compared to those that use circuit
3.2. Efficiency of cleaning
Regarding to cleaning it is noted that the range of more efficient exploitations are those that are
around 60 cows, this is because the task of cleaning of major spent time is the cleaning of
cubicles and in these exploitations, the system of used bed is bed of rubber which is the more
efficient system.
Courtyard of exercise: Spent time at this stage is between 30-40% of the spent time in cleaning.
Although there are exceptions in exploitations with slatted and have not performed cleaning,
considering sufficient the autocleaning that it being done with slatted, neither is considered the
spent time to this task, when is performanced with earth scoop if this is automatic.
Cubicles: Within the cleaning is the stage to which is devoted more time between 30-60%.
Cubicles in general have an acceptable cleaning, and exploitations have greater efficiency are
those in which the bed is made of rubber because they are cleaned more quickly. It is also
noted that efficiency is inversely proportional to the number of cubicles, this task is done
manually because it is not possible to mechanize.
22
Feeding troughs: This task deals between 30-40% of the spent time on cleaning. Here we do
not establish a clear evidence of the cleaning state with the spent time to it, because it should
be proportional to the devoted time to this task with the number of cattle and this is not the
case.
Drinking troughs: The cleanliness of the drinking troughs is a relative parameter. In some
exploitations are cleaned only when they are dirty, i.e., there is not a routine cleaning; in others
have a very efficiency because the cleaning consists in to empty the drinking troughs using a
lever. We also note that are more efficient the linear drinking troughs than of cup, this is due
because a linear drinking trough does the same function than three of cup.
3.3. Efficiency of food
With regard to food the more efficient exploitations are those that have more cows, this is
because the larger one exploitation less time devoted to this task and this is the result that
these farms use mixer truck to prepare the ration and provide it for livestock, thus saving much
time.
On this stage, we have very different efficiencies due to the exploitations where the food is
done by hand, to provide green fodder includes the time it takes to go get it. Among the
exploitations where food is provided by truck mixer, the low efficiency is because the silos are
removed from the farm.
Tapping: On this point we have very little data, because the majority of exploitations do not
apply an independent tapping of the ration.
3.4. Relationship of the total efficiency with characteristics of the exploitation
In an attempt to find the relationship of the total efficiency with the exploitation together in this
paragraph we use the total surface and the shape indicators (Alvarez, et al., 2007).
Per square meter and cow: If we calculate the total area of the exploitation as the surface of the
exercise yards, cubicles, feed yard and milking room; and set the ratio of m2/cow we can
compare the values of gained efficiencies.
In Figure 2 we see that the greatest efficiencies, are in exploitations with 20 square feet per
cow, to higher or lower surfaces the efficiency decreases.
Total efficiency per index of shape of the exploitation: The index of shape is defined as: The
occupied area by the exploitation divided between the perimeter of the square. The results are
in Figure 2.
23
m2/cow
Index of shape
Graphic 2. Relation of the Efficiency.
We note that the highest efficiencies are in the exploitations where the index of shape is around
0.04, and in general with higher and lower values the efficiency decreases.
However, noting the figure 2 we see, for the two previous cases, there is no clear evidence
between these rates and the efficiency, so we cannot set conclusions.
4. Conclusions
In relation to the presented data and keeping in mind the number of analyzed exploitations, and
differences between them in terms of number of heads and management, we will proceed to
see the connections between the various parameters:
Milking: The preparation of milking is reduced when we have a waiting room, and within them
with the available data, we can say that it is a little more efficient the rectangular waiting room
than the square room, if both have pushers. In the milking, we can say that is the stage more
efficient in those exploitations which have milking room compared with those with circuits, and
regarding to type of room, we do not have enough data to know which is more efficient. In the
cleaning of milking systems, we are unable to differentiate clearly between the different milking
systems, although there is more efficiency in the farms that have milking room against those
that have circuit.
Cleaning: Regarding to the exercise yard, we see that this task becomes more efficient when it
is mechanized a little. In the exploitations the cubicles that having a higher efficiency are those
that have a bed made of rubber, and lower efficiency are those with bed made of straw. Rubber
beds despite being the most efficient are also those with a good assessment of the state of
cleaning. As for feeders we cannot establish a clear relationship between efficiency and the
value of the state of the feeding troughs. In the drinking troughs we note that the linear drinking
troughs have more efficiency than the cup drinking troughs.
Food: In the preparation of the silos and fodder we note that are more efficient the exploitations
with ration compared with those that have not ration and within these, the exploitations with
major efficiency are those where the silos are in the vicinity of the exploitation and on the other
the farms that provide green fodder have very low efficiency because the spent time on
collecting.
24
Preparation for the ration: The most efficient exploitations are those that have this mechanized
task facing those which do it manually.
Tapping: In this case we cannot establish the more efficient exploitations, because not all them
apply tapping, some of the Vilafrío have the freely disposition to the animal, in other is applied
with feed and most of them do not apply it.
Finally to stress the importance for the future to continue with these studies given the large
impact that the results of efficiency, both total and each operation have regarding to productivity
of dairy farms in Galicia; and in therefore the potential increase of output if it is made a
productive management.
References
Álvarez, C.J.; Cuesta, T. S.; Iglesias, G.; Marco, J. L. and Resch, C. J., 2008. Neural Network
Analysis of Dairy Farm Efficiency in Galicia. Summited Transation of ASABE
Álvarez, C. J.; Riveiro, J. A. and. Marey, M. F., 2008. Typology, Classification and
Characterization of Farms for Agricultural Production Planning. Spanish Journal of Agricultural
Research. 6 (1), pp. 125-136
Álvarez, C.J.; Marey, M.F.; Amiama C., 2007. Criteria for assessing dairy farm types Recursos
Rurais 1(3), pp 51-59, Lugo.
Bravoureta, B. E. And Rieger, L., 1991. Dairy farm efficiency measurement using stochastic
frontiers and neoclassical duality. American of Agricultural Economics 73(2): 421-428.
Coelli, T. J., 1995. Recent developments in frontier modelling and efficiency measurement.
Australian Journal of Agricultural and Resource Economics 39(3): 219-245.
EDF-European Dairy Farmers, Analysis 2007. Cost of Production Comparison (2007). Institute
of Farm Economics and Rural Studies, Federal Agricultural Research Centre (FAL), Germany
Grossi, S. D. and Freitas, M. A. R. 2002. Reproductive and productive efficiency in commercial
dairy herds monitored by information system. Revista Brasileira de Zootecnia-Brazilian Journal
of Animal Science 31(3): 1362-1366.
Haghiri, M., Nolan, J. F. and Tran, K. C., 2004. Assessing the impact of economic liberalization
across countries: a comparison of dairy industry efficiency in Canada and the USA. Applied
Economics 36(11): 1233-1243.
IGE., 2005. Encuesta de explotaciones de vacuno en Galicia. Santiago de Compostela, Spain:
Instituto Gallego de Estadística.
Kompas, T. and Che, T. N., 2006. Technology choice and efficiency on Australian dairy farms.
Australian Journal of Agricultural and Resource Economics 50(1): 65-83.
Maseda, F., Díaz, F. and. Álvarez, C. J., 2004. Family Dairy Farms in Galicia (N. W. Spain):
Classification by Some Family and Farm Factors Relevant to Quality of Life. Biosyst Eng. 87(4):
509-521.
Richards, T. J. and Jeffrey S. R., 2000. Efficiency and economic performance: An application of
the MIMIC model. Journal of Agricultural and Resource Economics 25(1): 232-251.
25
Riveiro,J.A.; Marey, M.F.; Marco, J.L.; Álvarez, C.J., 2008. Procedure for classification and
characterization of farms for agricultural production planning. Aplication in the Nortwest of
Spain Computer and Electronics in Agriculture, 61(2) pp. 169-178.
Romain, R.; Lambert, R., 1995. Technical efficiency and costs of production in the dairy sector
of Quebec and Ontario. Canadian Journal of Agricultural Economics-Revue Canadienne
D´economie Rurale, 43(1) 37-55.
Acknowledgments
The authors are grateful to the General Secretariat R & D of the Galician regional government,
for funding this research through the project "Improving the efficiency of dairy farms in Galicia,
with reference: PGIDIT06RAG29101PR.
Correspondence (for further information, please contact):
Sonia Irimia Fernández – GI Projects and Planning
Polytechnic University. University of Santiago de Compostela. Departamento de Agricultural
Engineering Department
Campus Universitario s/n, 27002, Lugo, Spain.
Phone: +34 982 285900 ext 23262
E-mail: [email protected]
26
TECHNICAL-ECONOMIC ANALYSIS OF THE PIG EXPLOITATIONS IN
THE AUTONOMOUS COMMUNITY OF GALICIA (SPAIN)
Marey Pérez, M.F.
Sobrado Castro, O.
Riveiro Valiño, J.A.
Álvarez López, C.J.
Universidad de Santiago de Compostela
Abstract
Since the mid-50's of the past century in Galicia has evolved subsector pig until today where an
average 1,220,000 slaughtered pigs with a production of 94,400 tonnes per year originate
mainly from industrial exploitations 581 breeding and the 745 existing feedlots. The sector is in
a complex process of adaptation to new market conditions characterized by an increase in
production costs. This paper shows the results of the analysis of the production process of 33
pig exploitations bait selected randomly in 4 provinces of Galicia. The results are organized so
that from the weight (Kg) and the average price per (€/Kg) of the animal priming and marketed
analyzes all elements of the productive process (acquisition of piglet, food, animal health costs,
costs of legal proceedings production facilities, costs of hand work) in terms of their influence
on the performance end of it. It shows the different strategies exploitations classified according
to their efficiency.
Keywords: Subsector pigs, productive process, efficiency analysis
1. Introduction
Spain is the second country of the European Union and the fourth country in the world in terms
of pork production. In recent years, pork production has grown strongly and continuously in
Spain. The last census reported that there are a total of 23,5 million heads, of which 2,5 million
is in line with Iberian pigs, reaching a production of 3.122.577 kg of meat. Most of the pig hut is
distributed in Catalonia, with 26% of the total, followed by 18% of Aragon and Castile and Leon
14%. (I.N.E.,1999). The distribution of holdings reaches 31.000 pig exploitations.
Throughout the past decades occurred processes of modernization and verticalization of
production chains. The merger is also very sharp and there are very little independent
exploitations. In most cases they are cooperative societies or associations with the feed, or
even to slaughter their own meat industries. The production of pig meat cannot be absorbed by
internal demand, so that exports are an essential resource for the survival of the sector. Exports
came to 617.300 tons or representing an increase of 16,8%. The main client of this foreign
trade is France with 32,1% of the total, followed by Portugal with 19,4%. According to industry
27
experts, by 2010 production will continue growing strongly in Spain (12%) while it will slow
down in most of the European Union.
When making an overview of the pig Galician compared with Spanish noted the significant
weight loss within the national context in the past thirty years. While the total Spanish census
has tripled in three decades, Galician remained constant. The pig subsector is one of the most
important subsectors of the industrial Galician livestock. In 1997, the pigs received a total of
114,26 € million which represented 10,5% of the final Galician production livestock (PFG).
In Galicia in 1998 had a total of 841 exploitations with a total production of 82.779 places. Most
of the exploitations, 74% had a capacity of 99 units or less, while 70% of the places were
concentrated in exploitations bait was 385, 89% of such exploitations had room for 200 or more
concentrated where 98% of places bait. In 1998 there were 206.789 places in fattening
exploitations. In Galicia in 1998, the average price of animal feed was 0,83 € / kg pig and of
piglet was 29,33 €/UD (I.G.E., 2007).
The prices of feed compounds showed a sharp rise over the past months, which affected all
livestock sectors, but especially for pig farmers. This increase is mainly due to poor grain
harvest. (Ministry of the Presidency, 2000)
Royal Decree 324/2000 (Ministry of the Presidency, 2000) and R.D. 3483/2000 (Ministry of the
Presidency, 2000) amending the previous, established the basic standards of management of
pig exploitations. Its main objectives were: 1) Be a new regulatory framework in the
management. 2) Adjust the harmonious growth of the sector. 3) Reduce the maximum
dissemination of diseases. 4) To preserve natural resources and protect the environment.
1.1 SWOT Analysis of the pig subsector in Galicia
A SWOT synthetic analysis of the pig subsector in Galicia produces the following results
Weaknesses
-
Most of the pig exploitations do not have the relevant management plan slurry.
-
The health status of Spanish pig hut poses certain limits on exports. One of the main
problems is Aujeszky's disease.
-
The urban pressure put up the price of rural land that makes it difficult causes install
new exploitations, especially in coastal areas.
Strengths
-
The favorable reception by the Royal Decree 324/2000 of the pig sector, mainly in
its zootechnical classification and productive capacity.
-
The protection of the environment increases the maximum capacity defined pig
exploitations.
-
The establishment of the minimum distance promotes a proper health management
of pig exploitations.
-
The government and industry can anticipate the emergence of potential epizootics.
-
The production systems "throughout the entire inside out" as a measure of
biosecurity on bait exploitations and transaction of piglets, can improve sanitary
conditions.
28
-
Implementation of the farmers by the rules in respect of sanitary infrastructure, and
measures of hygiene.
-
Proper maintenance of the Registry Book of the exploitation and the correct
identification of the animals, which ensures both the control on the movement of
animals and the traceability of the final product.
-
Trend to the capacity increase of exploitations, and that is causing a lowering of
production costs.
-
The application of new production systems reflect the sector's interest in being in the
forefront.
Threats
-
Environmental management of slurry and manure, disposal of pig carcasses can
cause sanitary risks and environmental damages.
-
The competition between tourism and the pig is causing the closure of many
exploitations, especially in coastal areas.
Opportunities
-
Ensure that the distribution of pig exploitations in all territory results uniform and
rational with available resources.
-
Control and eradication over the medium term of certain diseases.
-
The significant support that will ensure the start of a one-stop shop to facilitate
administrative tasks to the farmers.
-
The use of genetic of quality will allow for greater adaptation to market demand,
while minimizing costs and offering different products.
-
The guarantee of traceability will allow increasing the consumer confidence in pork
products.
2. Object
Development and validation of an analytical methodology to identify in the current conditions in
the near future, the current situation and the potential production in Galicia for the production of
the pig subsector by learning from the current situation. Specifically this study aimed to explore
the technical and economic situation of the exploitations selected.
3. Materials and methods
3.1 Area of study
The work covered the entire region of Galicia.
3.2 Preparation of census exploitations the pig subsector
It is made a request to the Rural Counseling requesting information on the pig sector in
Galicia. From the response is established a census classified into three groups:
29
1
Industrial breeding exploitations: 581.
2
Small exploitations with breeding: 2.749
3
Feedlot: 745.
3.3. Selection and distribution of the sample
30 exploitations were selected randomly from groups 1 and 3 (industrials with breeding and
feedlot).
They were located geographically and established a program of visits for the completion of the
questionnaire, consisting of telephone contact with the owner of the farm and arrange an
appointment for the survey. In this first contact was made in explaining the objectives of the
survey and were guaranteed anonymity in all phases of it.
3.4. Elaboration and implementation of the questionnaire
The questionnaire was divided into three parts; 1) location of the farm. 2) specific purpose of
this work, related to the production process in which they ask for race and working method as
well as the costs of this process, 3) refers to the facilities available to the operating state that
the data were not provided by respondents, because in most cases unknown. The interviewer
made the measurement field for each section.
Is shown as an example to analyze the questionnaire conducted at an exploitation.
Survey nº1
1
The situation of exploitation - Santiago of Compostela ( Coruña)
2
Race - Pietrain
Use system - Intensive
Number-300
Percentage of lows exploitation - 2%
Reset - No
Bait end units - 294
Hours/year dedicated workforce - 792
Desired weight of pigs fed - 105 kg
Price € / kg fattening pig – 1,15
Kg. of feed they eat to be fattened - 220
Kg. feed prices - 0,22 € / kg
Piglet price - 40 € / u
Zoosanitary expenditure – 178,5 € / year
Cost of electricity every 2 months - 45 €
Veterinary expenditure - 75 € / year
Labor costs € / year – 5.544
3
m² of accommodation - 240
m² store - 24
m³ of liquid manure pit - 200
Cost of accommodation - € 54.000
Cost of storage – 1.200 €
30
Cost of the slurry pit - € 11.000
3.5. Analysis of the questionnaire
When the data collection finished, its organization and subsequent analysis was carried out
using the program SPSSTM 15.1. The analysis were of descriptive type and correlation among
which we have
Central tendency:
Media. Sum of all values divided by its number.
Median. Value below which is 50% of cases
Fashion. Value that is repeated over
Sum. Sum of all values
Scattering:
Standard deviation. It measures the degree to which the values a variable that takes
away from the average. It is obtained as the square root of the variance.
Variance. Measure of dispersion is obtained by dividing by n-1 the sum of the squares of
the differences between each value and the average.
Amplitude. Difference between the maximum and minimum value.
Minimum. Smaller value.
Maximum. Larger value.
Form of Distribution.
Asymmetry. Index that expresses the degree of asymmetry of the distribution. Positive
asymmetry indicates that more extreme values are above average. Negative symmetry
values indicate that the most extreme are below average. Rates close to zero indicate
symmetry.
Kurtosis. Index that expresses the degree, to which a distribution accumulates values in
its queues, compared with the values accumulated queues a normal distribution. The
positive kurtosis indicates that in the queues of distribution there are accumulated more
cases than in the queues of a normal distribution. The kurtosis indexes close to zero
indicate similarity with the normal curve.
Correlation Analysis
The standard measure of correlation is the correlation coefficients, which are numerical indexes
that have the important property to quantify the degree of linear relationship between two
quantitative variables. It is usually selected the following:
Correlation coefficient of Pearson (Equation 1). The most commonly used to study the
degree of relationship between two quantitative variables. It is obtained typifying the
average of the products of differential ratings in each case (deviations from the average)
in the two correlated variables:
31
ρ xy =
Cov(x, y)
Var(x)Var(y)
(1)
Cov Covarianza
Var
Varianza
4. Results
Race - All feedlot use the same race of pigs: the Pietrain, which comes from Belgium, a
town called Brabant. It originated from pigs Normans. It has a high-quality channel, but low
growth rate, conversion index and reproduction (requires a high level of food to increase its
weight and reproduction). By contrast is the highest percentage of noble pieces, but with quite
fat intramuscularly. It is one of the races that which shows PSE (white meat, soft and
exudative). In Spain is considered one of the most muscular of the world, besides having best
qualities as a reproductive and production of ham, very well adapted to the different means of
exploitation.
Use system - In all cases the use is intensive. The characteristics of the installations,
discussed in another work presented at this conference and facilities of management and
increase profitability, incites that this is the most appropriate method for exploitations breeding
pigs. Changes in environmental regulation and management of animals for meat production,
are promoting the need for soften allowing the introduction of more extensive productive
processes. A significant number of exploitations surveyed were raised that possibility in the
future.
Number of pigs - The average number stands at 967 pigs per exploitation and year, the
median is 840 and the fashion 800. The standard deviation of this variable reaches a value of
613, reaching the rank of the same in 2.250, which states very clearly the different sizes of
exploitations. Orense province is presenting the largest installations with 1.230 pigs per farm on
average. The largest exploitation of the surveys was located in the province of Lugo with 2.400
pigs. On the other side smaller exploitations are located in Pontevedra, which averaged 716
pigs are like the smallest of all exploitations with 150. As expected the number of pigs and
positively correlates well with all the economic variables. In the opposite direction is found the
percentage the lows per death which decreases, suggesting that larger installations are more
efficient in the early stages of rearing which are often the most crucial, and this is because it is
most modern installations.
Percentage of casualties in the exploitation - the medium area is 2,52%, the median
stands at 2,45% and the fashion is 2%. The standard deviation of this variable reaches a value
of 0,008, reaching the rank of the same at 4%. Province Lugo is in this case with a 2,59% which
is the highest value and at the other end is Orense with only 1,76%. The correlation analysis
shows interesting results. As had been discussed the size of exploitations is related to the
decrease in the number of casualties, which is the same in the work on the exploitation, so the
devoted hours to the management are correlated negatively in -0,352. On the contrary seems
to have no relation to the hours for disinfection with at -0,01 not being significant Pearson's
32
correlation. The reason behind this may be in the fact that a greater number of hours is due to
installations and cleaning systems are older.
Hours / days of labor - the medium work in the Galician pig exploitations is 2,68 hours/
day, the median and the fashion is at 2 hours. The standard deviation of this variable reaches a
value of 1,28, reaching the rank of the same in 2. It is in Orense, where the average is at 4
hours and Lugo at 2,44 to the contrary. The correlation analysis are especially significant
regarding to the size of exploitation 0,55. As a result of this variable shows the dedication to the
Galician pig farm is of partial character, in many cases be linked to other activities such as the
exploitation of veal.
Costs of labor € / year – the average of this cost, of great importance to the profitability
of these exploitations are located in the 7880 €, the median is 5.880 € and fashion have the
same value. The standard deviation of this variable is worth 3.325 €, with the same range of
13.075 €. The province that spent more on labor per exploitation is Orense with 10.882 € on
media and the least Pontevedra is just of 6.205 € of media. The correlation analysis, shows as
expected, high correlations with all dimensional variables and significant negative correlations
with variables related to the bad management of exploitations such as the percentage of lows.
Price € / kg fattened pig - The result is always the same for this variable and has a
value of 1,15 € / kg. That price has not increased in recent years, but it has got down.
Kg of feed consumed per pig – Considering that the average weight of sale of the pigs
in Galicia is at 110 kg, it is necessary to achieve an average of 277,9 kg / pig, with a mode of
300 Kg. / pig and a range of 100 kg. By province are pigs of Orense with 316 kg while in
Coruña on average is 256 kg. The feed consumption is highly correlated, values 0,596 and
0,564, with variables such as hours dedicated of the exploitation. These results seem to
indicate a lower efficiency of the work, which may be associated with older installations that
require more work and represent worst conditions for animals that have an impact on declines
in production.
Price of Kg. of feed - The medium price stands at 0,26 € / kg, median and mode at
0.30. The standard deviation of this variable reaches a value of 0,03, reaching 0,08 in the
range. By provinces the prices do not seem to vary and do not seem to depend directly of the
given consumption which is not correlated with consumption.
5. Conclusions
The pig subsector in Galicia has gone through a phase of recession in recent decades. At
present, as a result of difficulties in the markets and rising prices of raw materials need to enter
in a phase of change that will suppose a new process of recession in the number of
exploitations.
It should be noted that currently there are two types of productive models in the pig installations
of Galicia. On one hand we find with exploitations with a greater number of years and with more
traditional forms to work and other side with more modern exploitations applying a more
exhaustive management in its activity. The same happens for provinces, the phenomenon of
the integration in COREN (Cooperatives of Orense) more widespread in the province of Orense
incites that there are the biggest and most modern installations.
A crucial factor for the future of these exploitations is in the fact of not having a sufficient size to
save costs and adapt to new market circumstances. In an economic model of product prices to
33
the low and cost of raw materials and models of production increasing, the solutions have to
pass by the diversification and specialization or by increasing the size. However, obtained
results indicate that none of these circumstances is occurring.
References
I.G.E. Prices of pigs 2007 (Online). In: Internet http://www.ige.eu/dba/Controlador. Consultation
in May 2007.
I.N.E. Agricultural Census 1999. (Online). In: Internet <http//www.ine.es> (Consultation,
May,2007).
Ministry of the Presidency, 2000. Royal Decree 324/2000, of March 3, laying down basic rules
for management of pig exploitations"B.O E., (58). Pp 9505 – 9511
Ministry of the Presidency, 2001. Royal Decree 3483/2000 of December 29, amending the
Royal Decree 324/2000 of march 3, laying down basic rules for management of pig
exploitations", BOE (11). Pp 1434-1435
Manuel Francisco Marey Pérez
University of Santiago of Compostela. Agricultural Engineering Department.
Polytechnic University. University Campus. 27002 Lugo. Spain.
Phone: +34 982252231 Fax: +34 982241835
34
Area: Land Management, Urbanism, Architecture and Civil Engineering
TOPIC AREA: LAND MANAGEMENT,
URBANISM, ARCHITECTURE AND
CIVIL ENGINEERING
35
STRATEGIES FOR SEWAGE TREATMENT
IN RURAL AREAS
Muiño, D.
Cancela, J.J.
Marey, M. F.
Neira, Xan X.
Cuesta, T. S.
Abstract
The application of the Water Framework Directive and the Directive 91/271/CEE conditions
the objectives of the Plans of Wastewater management approved in different Autonomous
Communities in Spain. In all the lines of performance the treatment of the towns with more
than 2,000 e-i has been prioritized. It is foreseeable that in one second phase the smallest
towns will study. In this case, the collective, individual or decentralized systems will be
valued of wastewater management. This paper analyzes the present situation of the
wastewater management in the rural scope of the province of Lugo by means of a survey in
the 67 city councils. The high-priority lines of performance and the methodology for the
selection of the suitable technology are defined from the collected data.
Keywords: Wastewater management, rural planning
1. Introduction
During recent years in Spain is making a major financial effort to adapt the infrastructure for
sanitation and sewage treatment. This is so due the growing awareness about the need to
rationalize the use of a limited natural resource as water and the obligation with existing
legislation.
The explicit requirement to purify the waste urban water is collected in the European
Directive 91/271/EEC (European Commission, 1991). With the aim of achieving the
compliance with this Directive approved the National Plan of Reorganization and Purification
1995-2005 (Council of Ministers, 1995) and the Royal Decree Law 11/1995 (B.O.E.
312/1995) transposing EU legislation in our legal framework.
In the case of cores with particular entity, due to the volume and variability of the composition
of the treated water, are conditioned to the use of so-called conventional systems. In small
cores the problem is completely different because the volume of waste is not as important
37
and, moreover, is made almost entirely of organic components. This allows great flexibility to
select the best treatment for each core or group of cores considered (Muiño et al., 2005).
In Spain, the way to assimilate this European normative within the provided period, varies
depending on the populational organization of the territory (Garcia et al., 2001). In the case
of Galicia, where there are the 50% of all entities of the population of Spain, exists with a
homogeneously populated territory (INE, 2004). This particular demographic distribution
explains the fact that almost 89% of institutions of habited population have a census of less
than 100 inhabitants, as illustrated in Table 1.
Cores
Population
Inhabitant
Number
%
Number
%
<100
25.625
88,79
664.213
24,15
101-500
2.838
9,83
539.694
19,62
501-2.000
300
1,04
271914
9,88
>2.001-10.000
98
0,34
1.275.0054
46,35
Total
28.861
100,00
2.750.875
100,00
Table 1: Singular entities of population in Galicia according to inhabitants of right 2003 (INE, 2004).
This geographical and demographical reality conditions the wastewater treatment (Alvarez et
al., 2006). The irregular topography of the region, with a great orographical variability, but
without high altitudes, makes it difficult connecting of next cores by collectors.
In Galicia, for the purpose of transposing the European and national legislation to our legal
framework were adopted the Law 8 / 2001 (Presidency of the Xunta, 2001) and the drafting
of the Plan of Sanitation of Galicia 2.000-2.015 (Presidency of the Xunta, 2001).
The Plan of Sanitation of Galicia is focused almost exclusively on the cores of population
older than 2.000 h-e believing in the fulfilling of the Directive 91/271/EEC. It must be
remembered that in Galicia are no less than 28.763 cores minor than 2.000 inhabitants which
agroup almost 54% of the population (INE, 2004). In this sense, it is proposed to connect of
the cores of more than 100 people. In the case of cores with less than 100 inhabitants, is
proposed its connection to an urban agglomeration or their considering as isolated core with
individual sanitation.
Although in the Plan of Sanitation of Galicia does not contain the pollution load caused by the
livestock, considering the same of diffuse character, the management of wastewater
generated by this sector is an issue of paramount importance. The census of livestock of
Galicia is close to 900.000 head of cattle, one million head of pigs and about 300.000 head
of sheep. However, in rural areas, the influence of this activity in the management of
wastewater depends on the type of exploitation and its location (Cuesta et al., 2006).
We start from a current situation where the availability of existing infrastructures is very
uneven in the set of the territory (Cuesta et al., 2006). In this reality we have to unite the
growing social and legislatitive demand, of new actions that are financed and implemented
38
by different agents (local entities, private initiatives, Deputations, Counseling or the
Administration of State).
This situation is compounded by the wide range of technical solutions, adapted to different
situations, that now the technology offers us (Cajaraville et al., 2007). Each of the systems of
available treatment is associated with different levels of demand in the purification and the
necessary investment and, what is particularly important, in maintenance costs.
From the above description of the problem of wastewater management in rural areas seems
necessary to promote the coordination of social involved agents. It is therefore important to
establish basic standards or recommendations on the use of different technologies and to
define the main lines of action for a proper planning for wastewater management in the rural
areas.
2. Preliminary approach
Considering the Council as the unit of action, and watershed as planning unit, is proposed an
open system that allows us to simulate, considering the reality, different alternatives from a
technical, economic and social approach. The variables that must be considered are:
• Demands and needs of purification.
• Economic and technical capacity for the proper maintenance and functioning of
installations.
• Ecological factors of the receiving environment and social factors of acceptance.
To identify the possible locations of the WWTP and the most appropriate treatment systems,
it needs to start with the knowledge of the characteristics of wastewater, the legal conditions
and the desired quality in the effluent (Cuesta et al., 2006). It is always essential to seek the
balance between the cost of installation and performance while minimizing the production of
waste and try to obtain profits when this is possible. Furthermore, we must link the different
alternatives in the purification with the alternatives in the design of sanitation networks.
In this way, at different levels or stages, it is necessary to assimilate the current situation and
generate alternatives to the raised problem. Thus, in the definition the initial data relating to a
council, it is obligatory to consider the following aspects: (Muiño et al., 2005)
•
Points of current discharge: geographical location, and characteristics of the
discharge and characteristics of the receiving environment.
•
Typification of the basic characteristics of different discharge depending on the flow
and seasonality, population (number, type and forecasts), organic load and possible
individual products.
•
Existing infrastructure of sanitary and purification.
•
Elements of demographic and economic character.
•
Environmental and social characteristics.
Based on existing data, the next step is the location and dimensioning of the potential points
the purification and establishment of potential clusters. For this, we must assimilate a number
of sections such as:
39
•
Defining of treatments of purification more suited for each core or grouping.
•
Location of possible points of discharge and requirements of the receiving
environment.
•
Basic definition of potential networks of sanitation.
•
Elaboration of alternatives and economic evaluation of the implementation and
maintenance.
An open design of the system should allow us to pose a problem from different perspectives.
In this way we can compare different solutions to a precise problem, analyze the reality of a
local entity or watershed (Carballo et al., 2007). This will allow us the selection of alternatives
in the design of sanitary networks and support to other planning entities.
The proposed methodology is developed in three different stages and is applied to a pilot
basin for the purpose of validating the results for the design of a model of general
application. Currently we are working to get results in three small watersheds, particularly in
the following rivers: Estanco (Antas of Ulla), Madalena (Vilalba), Ferreira (Guntín) and Mao
(Monforte de Lemos), all in the province of Lugo (Carballo et al., 2007).
The methodology described suggests that, once defined the prevailing levels of quality in the
receiving environment, through its characterization, we must, in accordance with the
characteristics of the generated wastewater, to determine the level of required treatment in
each case and, then, to select the most appropriate technology of treatment. In a final phase
alternatives of dimensioning and location will be generated and the technical and economic
evaluation of them.
3. Characterization of the receiving environment
If is taken as work unit the watershed, then will be characterizated the receiving environment
under the Water Framework Directive (European Commission, 2000). This methodology,
applied to the definition of ecological status and other characterizations in the pilot basin is
based on a complete tour of the river. The roadmap is divided into three phases: information
gathering, river tour, sampling and analysis of the results (Muiño et al., 2007).
The phase of gathering information aims to locate, analyze and synthesize all the necessary
documentation to know the characteristics, ecological quality and uses of water from the
basin. The information collected before of integral tour will allow doing a previous tranches
and planning the field work. This step of doing tranches could be amended subsequently, so
that they can integrate of logical form all the indicators at the time of the definition of the state
of the different tranches of study and the receiving capacity of them.
This phase includes the following tasks:
•
Mapping analysis to 1:5000 scale of the basin. Calculation of average heights and
slopes.
•
Geological and hydrogeological analysis of basins. Analysis the climate of the area.
•
Analysis of land use: crop, forest areas, livestock, activities of tourism, socioeconomy.
40
•
Inventory of hydraulic existing and future infrastructure for supply and discharges.
•
Study of water demand through the interpretation of features practices.
•
Hydrological analysis by tranches using methods of simulation (Témez) and data of
the foronómica network.
•
Analysis of the physico-chemical quality of water from the collected samples in the
integral tour of the river.
•
Analysis of the biological quality of the river (aquatic vegetation, macroinvertebrates,
ichthyofauna).
•
Analysis of the natural environment, especially the characteristics of the riparian
forest, wildlife associated to riparian environments and areas of natural interest.
•
Analysis of different hydraulic factory in the area.
•
Study of the artistic and cultural heritage.
•
Georeferencing of all elements in a GIS.
In the second phase is observed the integral tour of the entire river, and also the rest of the
basin. Throughout this phase should be made the remarks in continuous, sampling and
differentiation of the river in tranches (description and qualitative assessment).
The comments in continuous allow recording of all those anthropic condition referring to
physicochemical and hydromorphological indicators. Sampling embraces the measure of
flows of the municipal supply catchments in order to determine the availability of water,
sampling of benthic macroinvertebrates by surber network and its subsequent identification
in the labarotary and water sampling for analysis physico-chemical.
As for the differentiation of the basin in tranches our interest is in to describe the
hydromorphological characteristics (variation in depth, width of the channel, type of
substrate, the presence of fast, backwaters and weirs), the physical-chemical characteristics
(analysis of samples and visual quality) and biological characteristics (ichthyological and
benthic wildlife of invertebrates). These elements will allow us to make the qualitative
valuation by hydromorphological, physical-chemical and biological parameters. WFD
specifically mentions three types of indicators:
•
Hydromorphological indicators: regime of flowrate, continuity, variation in depth and
width, velocity of water, structure and substrate of the bed, structure of the riparian
zone.
•
Physicochemical indicators: visual quality of the water (transparency, colour,
presence of foam, smell), water quality (analysis in laboratory).
•
Biological indicators: spatial and temporal variation in the capture of salmonidae
species (trout) and cyprinid, taxonomic composition and abundance
through indexes TBI and BMWP '. (Jaimez et al., 2006)
The ultimate objective of the phase of Analysis of the results is the definition of ecological
status and determining the ability of the river as a receiving environment of wastewater.
41
4. Definition of the needs in the management of wastewater
Once that the objectives of current quality are known for the receiving environment, it is
necessary to determine the pollutant load of issued wastewater. With these data it could
establish the necessary level of treatment and technology available alternatives.
4.1. Definition and location of the discharge
In the GIS is necessary to have an inventory of existing discharges and with possible
discharges of defined groupings previously. These singular points must characterized and
categorized for future generalization.
4.2. Determination of the level of treatment
To determine the level of necessary treatment of wastewater of a specific core it is necessary
to consider the local conditions and the objective of defined quality for the receiving
environment and the maximum and minimum required limits of the purification.
The guidelines in determining the level of wastewater treatment in the case of small cores in
rural areas have that adapted to the previously defined reality assuming limitations on the
financial capacity and management.
To set the level of performance of elimination of DBO5 and MES contemplate us:
•
The ability of the receiving environment, taking into account the expected uses of
water (maximum concentration of BOD5 in mgO2/l and maximum concentration of
MES measured in mg / l).
•
The pollutant discharged load (total number of inhabitants equivalent of 60 g
DBO5/day).
•
Equivalent connected population to systems of secondary treatment.
•
Equivalent population that still has not systems of secondary treatment.
•
The flow of drought in the water course of the receiving environment with a return
period of 5 years.
4.3. Alternative in the purification
Once determined the profits it is proposed alternatives in the different treatments. The
primary treatment is reserved for populations who poured into fragile water channels, with a
significant dilution and having little severe drought. The secondary treatments are necessary
in the rest of cores. They may consist in a primary treatment (optional), followed by a natural
treatment or, if necessary, of a conventional treatment.
It is necessary to make a special evaluation in the following cases:
•
When the discharge takes place in an Area of Natural Interest or in a particular
volume of water susceptible to eutrophication.
•
When the discharge takes place in an area with an aquatic life of special interest.
•
When the discharge takes place in a volume of water that is used as a source of
supply.
42
•
When the receiving channel presents a very strong seasonality.
In the first three cases, is necessary to consider the need to eliminate the ammonia nitrogen.
In the third case, also it will be necessary to achieve nitrate concentrations below 10 mg N03N per liter. When this limit is exceeded, the system will be provided of a process of
denitration.
In the latter case, when exists circumstances of low dilution and vulnerability of the receiving
environment, will be a priority the elimination of discharges during the critical period. This can
be achieved with an effluent storage or a direct application on the ground.
5. Conclusions
As a final conclusion of this study is proposed the definition of a Support System for decision
making which must meet a number of conditions on the geometry of the system and the
mathematical structure.
A Spatial Decision Support System in the management of wastewater is a practical
application of Spatial Decision Support System (Spatial Decision Support System SDSS).
These tools are based on an integrated set of software, which allows helping in determining
the optimal location of various equipments. The necessary system must facilitate:
•
The exploration of the treated problem, with the aim of formulating hypotheses that
get it resolved.
•
The generation of alternative abundant and varied solutions.
•
A precise assessment (if possible quantitative) of the merits and disadvantages of
different solutions.
To meet these requirements are necessary different components: a graphics generator, a
report generator, a system of database management, a system of management of
mathematical models capable of generating alternatives and solutions and to evaluate them.
As for the geometry of the system, the spatial allocation of infrastructure and services in rural
areas is an important engineering problem, with significant economic, social and
environmental impacts. In this way, the inadequate situation of a filter system causes
unnecessary expenses, lack of consideration of the environmental characteristics produce
excess of costs and important nuisance to the population in addition to damaging the natural
environment.
It is necessary to determine the characteristics of the WWTP (EDAR), of the emissaries and
the sewage network, for its proper digital representation in GIS:
•
Spatial distribution of WWTP: Concerning a specific point (or centroid of each of the
spatial considered areas). One of the essential elements of the procedures that are
discussed later is the obtaining of the location and the consistent distances between
offer and demand.
•
Spatial distribution of the emissaries: In principle, the offer is always linked to specific
points in space and that is due to homes or rural cores.
•
Relationship between points WWTP / Emissaries and the sewerage network: To
connect points (of offer and demand) and sanitation system are added new linear
43
elements (tranches), so that each point of reported of offer / demand (which is not
located in a network node) with the nodes of the network surrounding it and near that
point.
As to the mathematical required structure we should describe the requirements of the model
that constitutes it. The objective of this model is to maximize the amount of products of the
population, the demand for each item and the distance to the nearest WWTP, ie, the
minimum distances between each point with the population and the installation (Lopez-Rubio
et al., 2002).
It aims to achieve the maximum spatial efficiency from the viewpoint of the affected
population, as the separation between unwanted installations and the set of the population
that is the highest possible. The purpose of the model is the maximizing of the minimum
distance between installations and the centres of the closer core (Woods & Franco, 1995).
This model raises the concept of maximum spatial efficiency of different way than the
previous ones. The objective is to ensure that the shortest distance between WWPT and
some of the cores result the possible largest (and thus the nuisances are becomed in the
minimum possible), given the spatial distribution of the populated places and
raised
candidates points.
For the evaluation of the raised solutions is intended to use, as Multiapproach technique, the
linear weighted sum of the scores of the various generated solutions, it is assumed the
procedure of the linear weighted sum of the established values for each solution on different
criteria for to be used for the evaluation. Thus, the solution to get the highest score will be the
best solution (Pomerol et al, 2000).
References
Alvarez C.J., Cuesta T.S., Cancela J.J. and Marey MF, 2006. Management of wastewater in
rural areas in Galicia, Spain (European Union). Information Technology, Vol 17 (3), pp.87-96.
B.O.E. 312/1995. Royal Decree Law 11/1995 of December 28, which establishes rules
governing the treatment of urban wastewater. Head of State.
Cajaraville, A., Cancela J.J., Cuesta T.S., Fandiño M. and Marey M.F., 2007.Current status
of wastewater treatment in rural communities in the province of Lugo. Water Technology 284:
48-52.
Carballo, R. Cancela J.J., Cuesta T.S., Churches G. and Neira X.X., 2007. Integrated
management of water in a small agricultural basin: the purification of water in small villages.
XI International Congress on Project Engineering, Lugo, Spain.
Carballo, R. Cancela J.J., Cuesta T.S., Iglesias G., Martinez E.M. and Neira X.X., 2007.
Definition of the ecological status of a small agricultural basin in northwestern Spain under
the Water Framework Directive. IV National Congress and Congress Iberian Agroingeniería,
Albacete, Spain.
Council of Ministers, 1995. Resolution of 28 April 1995, which provides for the publication of
The Council of Ministers of 17 February 1995, by approving the National Plan of
Reorganization and sewage treatment. Ministry of Environment and Housing.
44
Cuesta TS, Alvarez CJ, Cancela JJ, Miranda D. and Neira XX, 2006. Wastewater
management evaluation by using the opinion survey in Galicia, Spain. Water Resources
Management, Vol 20 (3), pp.104-115.
Cuesta T.S., Cancela J.J., Cajaraville A., Fandino M., Marey M.F., Muiño D. and Neira X.X.,
2006. Diagnosis and planning systems, wastewater treatment in rural communities in the
province of Lugo (Spain). X International Congress on Project Engineering, Valencia, Spain.
Cuesta T.S., Carballo R., Soto M. and Neira X.X., 2006. The wastewater in small rural
communities: introduction of cleaner low-cost systems. V Iberian Congress on Water
Management and Planning, Faro, Portugal.
D.O.G. 104/2001, 2001. Resolution of 22 May 2001, which gives publicity to the approval of
the Plan of Reorganization of Galicia 2000-2015. President of the Xunta de Galicia.
D.O.G. 155/2001. Law 8 / 2001 of August 2, Protection of water quality of the Galician Rías
and Public Management of urban wastewater purification. Presidency of the Xunta.
European Commission, 1991. European Directive 91/271/EEC of 21 May 1991 on the
treatment of urban wastewater. European Commission. Brussels, Belgium.
European Commission, 2000. Water Framework Directive, Directive 2000/60/EEC of 22
December 2000 establishing a framework for community action in the field of water policy,
Off. J. Eur Comm., Vol 327, pp.1-72.
Garcia J., Mujeriego R., Obis and Bou J., 2011. Wastewater treatment for small communities
in Catalonia (Mediterranean region), Water Policy, Vol III (4), 2001, pp.341-350.
INE, 2004. “List 2003". National Institute of Statistics. Ministry of Economy and Finance,
Madrid.
Jaimez P, Palomino JA, Luzon J and Alba J, 2006. Comparison of methodologies for
evaluating the ecological status of watercourses. Water Technology 278, 42-57.
Lopez-Rubio, E., Munoz-Perez, J., Gomez-Ruiz, J.A., 2002, Self-Organizing Dynamic
Graphs, Neural Processing Letters 16 (2) 2002 93-109.
Muiño D., Cuesta T.S., Iglesias G. and Neira XX, 2005. Alternatives to the sewage treatment
in small rural communities: the case of the Autonomous Community of Galicia. Proceedings
of the IX International Congress on Project Engineering, Malaga, Spain, pp.115-116.
Muiño D., Cuesta T.S., Cancela J.J., Carballo R., Marin A. and Neira X.X., 2007. Definition of
ecological status of river Estanco (Antas of Ulla, Lugo) under the Water Framework Directive.
Water Technology 290: 58-63.
Pomerol, J.-Ch. and Barba-Romero, S., 2000. Multicriterion decision in management,
principles and practice, Kluwer, Dordrecht.
Woods, J., Franco, S., 1995. Location-allocation models and Multicriteria Evaluation for the
Location of unwanted facilities. Geographical Series 05, University of Alcalá. Publishing
Service
45
Thomas S. García Cuesta
Department of Agricultural Engineering, University of Santiago de Compostela
Polytechnic School, Campus s / n 27002 Lugo, Spain.
Phone: +34 982 22 39 96
Fax: +34 982 28 59 26
URL: http://www.usc.es/agrofor
46
AUTOMATED VECTORIZATION OF URBAN AREAS BY USING
VISIBLE AND INFRARED SPECTRUM IMAGES
De Cos Juez, F.J.
Prendes Gero, M.B.
García Fernández, R.
Universidad de Oviedo
Alonso, A.
Seresco S.A.
Abstract
A GIS is a computer system capable of capturing, storing, analyzing, and displaying
geographically referenced information; that is, data identified according to location. The
power of a GIS comes from the ability to relate different information in a spatial context and
to reach a conclusion about this relationship. The extension of their use is limited by the
possibility to incorporate information. Most of the times it must be done manually through the
digitalization of the images. Unfortunately this is a complex and tedious task and it
represents the maximum contribution to the cost and source of mistakes. The present work
shows a software application destined to be used as support for the soil uses planning in the
concrete case of urban environments. Present application, throw the combined analysis of
images in visible and infrared spectrum, is capable of providing vectorized information so
much of the buildings and streets as of the location of green spaces without human
intervention.
Keywords: GIS; Urban design; Infrared imaging, automatic vectorisation
1. Introduction
The constant evolution of technology based on capturing aerial or satellite images, as well as
the more sophisticated computer resources, have managed to act as a substitute for
fieldwork related to the extraction of geographic characteristics. This means a considerable
cost-saving, resulting in a globalization of resources and systems a short time ago reserved
only to a few organizations. Nevertheless, these processes are still far from being fully
automated (Clapp 1997); therefore it is necessary to do some research on the development
of intelligent software in order to deal with certain processes which, although they may be
seen as simple for human understanding, they are far from being so for a machine
(Pijanowski 2002).
47
This particular case is located into the digital cartography field. The purpose of this work is to
automate as much as possible the making of a territory/ field map from high resolution
images. It is proposed the development of an automated tool capable of providing
information on street axes and borders displayed in those images, as well as green areas.
The starting point of every Geographic Information System is the capture of aerial images,
either taken at low altitude or by satellite. In any case, prior to the beginning of the image
acquisition process, it is advisable to establish the requirements of our system. The highest
resolution is not always necessary, since a great amount of information could have the
opposite effect if what is required is a roughly field analysis.
Generally speaking, it is convenient to take several exact images of the same position. If we
capture images from a satellite, two pictures taken at two different times may be used to
identify and discard shades and objects in motion which could be easily mistaken. Provided
the images are captured at low altitude, they must be taken simultaneously and avoid retakes.
In order to solve this problem, the best option considered is the combination of images with
complimentary information, as the ones taken with thermographic and digital cameras.
Thermographic cameras allow us to highlight vegetation areas where RGB images have
colour values close to black; or else detecting objects at first glance hidden in shades,
though its infrared radiation is enough to identify them in a second RGB image.
2. Methodology
To achieve the proposed aims and objectives, a series of characteristics must be fixed as to
indicate whether something should be considered as green or urban area.
In connection with colours, all greyish areas should be considered as urban areas (streets).
This is due to the fact that the asphalt colour varies depending on its type and age, being
thus impossible to determine its exact tonality accurately. In addition, given the amount of
shades shown in the images, black areas should also be taken into account. Generally
speaking, colours with similar components are taken into consideration, except those closer
to white. On the other hand, any green coloured area must be analysed in order to identify
vegetation areas, as such areas vary their colour tone depending on their type, typology and
preservation status.
In relation to form, different perspectives are used depending on the required typology. In the
case of streets, their specific width limits are essential to leak those areas which do not meet
those limits. Likewise, another key feature to be taken into account is the streets
interconnection. Therefore, isolated segments are immediately discarded from filtering.
The following diagram represents the general process:
48
+
RGB
Image
IRG
Image
Graphic
Editing
Program
Difference
RGB- IRG
(all
channels)
Street
Detection
Diferencia
RGB- IRG
(red
channel)
Green Areas
Detection
2.1 Previous steps.
Prior to the implementation of this process comes the combination of IRG and RGB images
to create input images. Two different versions of the same image (IRG and RGB) are merged
by a graphic editing program.
Applying the “minus” arithmetic operation between IRG and RGB images (in this particular
order) and covering all colours (red, green and blue), we will get an image where green
areas are displayed in red, while urban areas are highlighted in green. Unfortunately, grey
areas cannot be accurately filtered. Streets and grey areas will remain in black. Illustration 2
displays the input images as well as the results of this training process.
49
Illustration 1: Input and output images alter merging IRG and RGB images for the process of street detection.
The previous arithmetic operation was applied among images, in this case covering red
colour only. In this particular case, the output is a black and white image with green and
urban areas separated from the rest, as it can be seen in Illustration 3:
Illustration 2: Output image alter merging IRG and RGB images for the process of detection of green areas.
2.2 Street detection
The following diagram shows the street detection process (Illustration 4):
50
Street
Binary Image
Generation 1
Binary Image
Generation 2
Binary
Binary
Classification
Result
Illustration 3: Diagram for the street detection process.
2.3 Generating binary image 1
To generate the binary image from the input images, it was first covered the IRG image pixel
by pixel, assigning a false value to every with RGB average components higher than 180; or
else, when the difference between the maximum and minimum of their components is below
a certain threshold, being 40 for this case. This method allows us, on the one hand, to
discard most of the pedestrian areas as well as white areas, and on the other hand, it helps
also to discard some set of pixels which cannot be properly evaluated by the generating
binary image 2. Illustration 5 shows the results of the implementation of the aforementioned
techniques on the input images.
Illustration 4: Result of generating binary image 1.
51
To generate the binary image 2, an algorithm is used to classify the pixels of the input image
depending on the values of its R, G and B components. The resulting image is scanned pixel
by pixel, separating those pixels with a difference between its red and green components is
higher than a certain threshold of the rest.
When the algorithm is finished, it will show a binary image where green areas are
represented by 1, and the rest of the image is represented by 0. A. A dilatation algorithm will
be applied to the output image to reduce noise.
Once green areas are apart, it was performed a similar process for buildings, which will be
green coloured. It was necessary to scan the image once again pixel by pixel, classifying
those pixels depending on the one hand on their different green and blue components, and
on the other hand on their green and red ones.
At this point, a binary image is obtained where buildings are represented by 1 and the rest by
0. However, contrary to what occurs with green areas, it is not yet possible to reach a
conclusive result. As one may see taking a closer look, it can be noticed not only backyards
and roof windows, but also shades produced by the difference in height levels, as not
belonging to the same domain. On the contrary, small street areas were classified as
buildings - red coloured cars – Therefore, it was necessary to scan once again over those
areas.
It is essential to pay further attention to those areas which are not considered as “buildings”
due to their small size. In order to do that, a flood fill is applied to the image, so that different
parts of the image are coloured differently, as can be seen in Illustration 6:
Illustration 5: Images resulting form binary image application.
2.5 Classification
It was first implemented on those areas with not enough width to be considered as
streets.These areas are frequently found in parks with asphalted paths or areas with
adjoined buildings at different heights, projecting shadows which lead to misleading
classifications. The Chamfer Distance System (Echeverria 2008; Yeh 2002), is used to
measure distances up to the edge of every pixel.
52
Finally, the line formed by pixels with a distance value higher than those corresponding to
opposite pixels in any possible directions in window 3x3. It must be added those pixels
whose distance value is equal or higher than that of their neighbours. Once the dividing
borderline is calculated, its pixels are covered, searching for those with a distance value is
lower than the minimum threshold considered for a street; then they are rejected.
Coming to this point, it is possible to yet consider certain areas as part of the domain (Calijuri
2002). These areas are not considered as streets, however they have not been erased in
previous stages by the cleaning algorithm, sue to their size. The elimination of these areas
was made according to width and continuity criteria. Given the interconnection among
streets, those areas with no connection with borderlines must be rejected as discontinuous.
Using the condition of appearing in 2 of the 4 borderlines and not in all of them is made to
reduce loss of information. Finally, the image is once again covered calculating the minimum
width for each continuous border area (Choi 2000). If that width value is surpassed by a
tolerance value determined by a percentage of the size image, it is then considered too wide
to be a street and is rejected from the domain. This is used to eliminate rivers or asphalted
areas, like courts or fields in corners.
It is at this point when we get the definite image. The results must be highlighted to be
finished, as pictured in illustration 7:
Illustration 6: Final result of automated identification algorithm for streets in urban areas. Plaza del Pilar.
2.6 Green areas detection
As with the previous case, to detect green areas a scheme as represented in Illustration 8
was followed:
53
Green Areas
Binary Image
Generation 1
Binay Image 1
Green Areas Isolation
Intermediate
Result
Classification
Result
Illustration 7: Diagram of detection of green areas process.
In this case, the image resulting from the previous stage is already a binary image.
Therefore, it is applied an enhancement process similar to the one used in other binary
images, in order to reduce the noise.
2.8 Green areas isolation
This is the process in charge of classifying elements within a green area. First, the grass
area must be separated from the rest. One of the characteristics observed when working with
grass areas is the high level of uniformity, compared to urban tree areas. Therefore, the
green area is covered pixel by pixel, searching those whose colour dissimilarity is lower than
a certain threshold (Phua 2005).
Although the homogeneity level is higher than in other areas, certain level of heterogeneity is
also found in adjoined pixels. A low number of discordant elements is allowed among
neighbouring pixels, adjusting thus the thresholds on the grounds of acquired experience
when applying the algorithm.
In any case, windows wider than 3x3 should be used, as the number of pixels to be
compared with each other must be wide enough to discard urban tree areas. The proper size
of windows will be determined by image resolution.
The corresponding widening algorithm will be applied the same as in other processes. The
orphan pixels will be cleaned before a conclusive result.
54
Illustration 9 shows the result of the application of the proposed algorithm to an image detail.
It is also included the original infrared and RGB details.
Illustration 9: Results of application of grass detection algorithm.
The last stage of the 3 process consists in separating different species within the same urban
trees area. For this purpose, we concentrate on their different colour tones in the RGB as
well as infrared images. In this particular case, the attention is paid mainly on tress with dark
green foliage. When it comes to classifying, the heterogeneity level of adjoining pixels must
be taken into account, as it is higher in urban tree areas (Mathey 2008). As a consequence,
although it may be seen as uniformity from the distance, it may not be quite so. This makes
us allow more pixels which do not meet the colour requirements, widening thus the
conditions and tolerance of the set. Therefore, the algorithm to be used would be similar to
the one used for separating trees from grass, allowing for RGB values more than for the
uniformity of the set.
55
As the presence of little shades is constant in all areas throughout urban trees, it must be
taken into account the frequency of darker pixels we are bound to find. When a slightly
darker pixel from those around is found, it will be checked if any of its “neighbours” belong to
the seeking set (Wong 2008). Supposing these pixels are not yet scanned, it must be then
calculated which set they belong to, t is then incorporated to the domain if a significant
number of their neighbours belong to the same set. The same will be applied to pixels with
some shades around it, which will be included in the domain along with their adjoining
neighbours.
The process finally finishes with the application of the above mentioned clean and
enhancement algorithms, proceeding afterwards with the merging of the input and output
images for the final result, as can be seen in Illustration 10:
Illustration 10: Result of the application of tree classifying algorithm.
3. Conclusions
As a conclusion, a code implementation allows for a 75% automatic street detection and 85%
green area detection within an urban area. This represents an important step forward in what
such type of tools are capable to perform.
The generated tool enables to store new content in the geographic information systems for
the Local Administration, thus improving their own management and making them more
efficient in customer service.
Finally, regarding future steps on this project, it is expected to maintain and update the
generated code with new generated data (input images from new locations). This will not
only enhance its usefulness, but also serve of great hep to contrast and filter the algorithms
expressly designed for this application.
56
Likewise, it lays the foundations for future species classification for the defined green areas.
References
Bhering, E.; Calijuri, M.L., Marques, E., Meira, E. and Maia, F., 2000. Digital land-use
cartography—the example of Viçosa, Brazil, Engineering Geology, Volume 63, Issues 1-2,
pp1-16
Brown, G.; Manik; G.; Pijanowski, B. Shellito, B., 2002. Using neural networks and GIS to
forecast land use changes: a Land Transformation Model, Computers, Environment and
Urban Systems, Volume 26, Issue 6, pp. 553-575
Clapp, J.M., Rodriguez, M. and Thrall, G.; 1997.How GIS Can Put Urban Economic Analysis
on the Map; Journal of Housing Economics, Volume 6, Issue 4, pp. 368-386
Coomes, D., Echeverria C.; Hall, M. and Newton A.C. Spatially explicit models to analyze
forest loss and fragmentation between 1976 and 2020 in southern Chile; Ecological
Modelling, Volume 212, Issues 3-4, 10, pp 439-449
Keechoo Choi and Wonjae Jang, 2000. Development of a transit network from a street map
database
with
spatial
analysis
and
dynamic
segmentation.
Transportation Research Part C: Emerging Technologies, Volume 8, Issues 1-6, FebruaryDecember 2000, Pages 129-146
Krcmar, E.; Mathey, A.; Dragicevic, S. and Vertinsky, I.; 2008. An object-oriented cellular
automata model for forest planning problems, Ecological Modelling, Volume 212, Issues 3-4,
pp. 359-371
Minowa, M. and Phua, M.; 2005. A GIS-based multi-criteria decision making approach to
forest conservation planning at a landscape scale: a case study in the Kinabalu Area, Sabah,
Malaysia, Landscape and Urban Planning, Volume 71, Issues 2-4
Sumić, Z., Venkata, S. and Yeh, E., 2002 Intelligent Tools and Their Applications in
Geographic Information Systems, Expert Systems, pp 489-551
Wong, N. and Jusuf, S.; 2008, GIS-based greenery evaluation on campus master plan.
Landscape and Urban Planning, Volume 84, Issue 2, 6, pp. 166-182
Francisco Javier De Cos Juez.
Área de Proyectos de Ingeniería Universidad de Oviedo.
C/ Independencia 13, 33002 Oviedo (España).
Phone: +34 985 10 42 72
Fax: + 34 985 10 42 56
URL: http://www.api.uniovi.es
57
MODEL FOR AGRICULTURAL PRODUCTION PLANNING
Cardín, M.
Álvarez, C.J.
Abstract
In this paper is described one of our most successful projects, which was the establishment
of a Model for Agricultural Production Planning in 21 rural areas (“comarcas”) of Galicia. The
first part of our research consisted in obtaining field information about these areas. Fieldwork
was carried out between 2002 and 2004, and consisted of 4.348 surveys made to farmers
from these 21 “comarcas” of Galicia, more than 350 interviews to experts, and of course
bibliographic, statistical and cartographic research. The second part consisted in the
generation of a Model for Agricultural Production Planning (including agricultural and
livestock farming, and forestry), which could be used for decision-making assessment in the
application of policies, programs and plans at this “comarca” scale. It was created to be an
instrument to plan agricultural uses of land, to rationalize and optimize the sustainable
exploitation of rural soils, and to achieve higher levels of rural development The core of this
model was the establishment of 44 indicators of sustainability (social, environmental and
economical indicators), and the integration of them in a computer application.
Keywords: Agricultural Production Planning, indicators of sustainability, indicators of rural
development.
1. Introduction
By the end of 2001, the University of Santiago de Compostela, under a form of joint venture
(Unión Temporal de Empresas or UTE) with EIDO GALICIA, S.L. Consultants, successfully
tendered for a contract to provide technical assistance on the project ‘Development of
Agricultural Production Planning Surveys in 22 comarcas of the Autonomous Community of
Galicia, 2001-2002’, funded by the Galician Administration. (UTE EIDO-USC, 2004)
The aim of our Agricultural Production Planning Studies was to generate an objective Model
of Agricultural Production Planning (comprising agriculture, farming and forestry) to support
and dynamize production sectors, which could be used as a decision-making tool to
implement policies, schemes and plans aimed at rural comarcas (Andersen et al, 2007)The
model was used as an instrument to allocate agricultural land uses, to rationalize and
optimize the sustainable use of rural land and to promote rural development (Álvarez et al.,
2008; Riveiro et al., 2008).
58
The Objective Agricultural Planning Model for Galician Comarcas synthesizes the information
pertaining to the distinctive features of the current situation and of the parameters that
govern the evolution of the situation based on the analysis of a number of elements that
characterize the agricultural subsystem of an area (natural environment, socioeconomic
conditions, infrastructure and legal framework). By using this model, the potential situation of
agricultural production in the area considered can be delineated. (De Wit & Van Keulen,
1988; Riveiro et al. 2005).
Agricultural Production Planning was considered as a process for the spatial organization of
agricultural and forestry products that allocated specific land uses to priority land areas at the
comarca level. The aim of such a spatial organization process was to achieve sustainable
development by optimizing agricultural production systems according to structural and
socioeconomic conditions and by considering environmental concerns (Riveiro et al., 2005).
The Comarcas included in the Agricultural Production Planning research were:
•
Province of A Coruña: Arzúa, Bergantiños, Noia, Ordes, Terra de Melide and Ortegal.
•
Province of Lugo: A Fonsagrada, Os Ancares, Terra Chá, A Ulloa, A Mariña Occidental
and Terra de Lemos.
•
Province of Pontevedra: O Baixo Miño, Tabeirós-Terra de Montes, Caldas, O Salnés
and A Paradanta.
•
Province of Ourense: Terra de Celanova, Terra de Trives, O Ribeiro, O Carballiño and
A Limia
Figure 1: Location of the 22 comarcas studied in the Autonomous Community of Galicia, NW Spain.
2. Methods
2.1. Characterization of Comarcas
The characterization phase involved the systematic and comprehensive collection of data
pertaining to aspects that could be used to describe the current situation of the structures
and productive sectors that form the agricultural structure of the comarca
The collected data were structured into three levels:
59
1) ‘Objective information’, obtained by reviewing all the documentary sources available
(literature, maps, statistics, internet information…).
2) ‘Field information’, obtained from a survey on the agricultural sector that was conducted
through direct and personal interviews with experts who were well acquainted with the
situation of the comarca (more than 350 experts over the 22 comarcas studied).
3) ‘Individual information’, obtained from a field survey among 4384 farm owners of all the
parishes and productive sectors in the different comarcas. The farmers survey was a
valuable source of information because it checked a variety of aspects related to the attitude
and competence of producers, which is particularly relevant in the assessment of the
response of affected farmers to the implementation of specific measures and actions.
2.2. Zoning of Comarcas
Although the contract was regional in scope, the research team searched for homogeneous
units within the regions or comarcas in order to increase the definition of the Objective
Agricultural Planning Model and to correct the internal heterogeneity of comarcas. The
defined homogeneous units were composed of groups of parishes with similar
characteristics. The research scope changed then from a municipal to a parish range; that
meant an additional effort to obtain the pertinent information.
Based on parish units, the regional territory was subdivided into a number of units with a
given level of homogeneity that could be useful in crop planning. To delineate such units, the
environmental, structural and socioeconomic characteristics of the different parishes were
analyzed. The 88 units resulting from the subdivision of comarcas were termed Ecological
and Economic Units (EEUs) and became the basic units of analysis and production of
results.
The method used to perform such a subdivision was based on multivariate cluster analysis.
Cluster analysis is aimed at solving problems in the classification of variables into
homogeneous groups. Consequently, cluster analysis defined parish groups (EEUs) or
clusters, so that the variance between variables of the same group was minimized and the
variance between variables of different groups or EEUs was maximized.
2.3. Data management
After the main characteristics of the initial situation were known and every comarca was
divided into homogeneous units, the following step was to compare the suitability of every
crop or land use for a given Ecological and Economic Unit (EEU). To that end, the so-called
Suitability Matrix was constructed to contrast the information obtained from the
characterization of comarcas for every EEU and every crop and land use considered.
2.4. Estimator design
To construct the matrix, a set of ‘Sustainability Indicators’ or ‘Estimators’ was defined.
Sustainability indicators provide information about the social, environmental or economic
constraints that determine the feasibility of farms for a given agricultural or forest land use in
every EEU. These estimators used information from the 1999 Agricultural Census, and also
from the conducted interviews and surveys.
60
By constructing estimators (one for each answer -or group of answers- to the survey, one for
each specific aspect -or group of data- from the Census), the total score obtained (number of
positive responses or specific value), which was interpreted in absolute terms, was
transformed into a relative value that was used to compare different EEUs and Comarcas. By
using relative values, the deviations of the analyzed values (e.g. responses to survey
questions or census data) from the mean of the entire set of comarcas can be known.
The general construction of the estimators is described below. There are three types of
estimators, classified according to the operational procedure used: simple, complex and
composite estimators. The method developed to obtain a Simple Estimator can be best
described by way of an example. Let us consider the percentage of positive responses to a
question in the field survey, named question K. The construction of the corresponding simple
estimator, EK, consists of the following steps:
1. Estimation of the percentage of positive responses to question K (or aspect K) in Unit
X, denoted by PK,X.
2. Estimation of the percentage of positive responses to question K (or mean value of
aspect K) for the whole set of comarcas, denoted by PK,T. This value, which coincided
with the mean value of positive responses, was assigned a Simple Estimator EK value
of 0.5.
3. Search for the Unit with the lowest value of positive responses to question K (or
minimum value of that aspect) from among the whole set of comarcas. Such a value
was denoted as MINK, and was assigned a Simple Estimator EK value of 0.
4. Search for the Unit with the highest value of positive responses to question K (or
maximum value of that aspect) from among the whole set of comarcas. Such a value
was denoted as MAXK, and was assigned a Simple Estimator EK value of 1.
5. The value of the Simple Estimator EK for Unit X was obtained by interpolation: if the
value of PK,X was lower than the mean for the Comarcas, the EK value for Unit X was
obtained by interpolation between MINK and PK,T. If the value of PK,X was higher than
the mean, the EK value for Unit X was obtained by interpolation between PK,T and
MAXK. Figure 2 shows the construction method.
Highest % of positive responses
% of positive responses for unit X
Average % of positive responses
Lowest % of positive responses
0,0
0,5
Value of the
estimator for
unit X
1,0
Figure 2: Conversion of total scores into homogeneous units.
61
Simple estimators were built such that low values of the estimator (minimum = 0)
represented frequencies of positive responses below the mean for the comarcas, and high
values of the estimator (maximum = 1) represented frequencies of positive responses above
the mean. The mean value for the comarcas was set at 0.5.
Such a method was valid when assessing responses to simple questions. However, the
response could sometimes take more than two values (positive/negative) or values lying in
different ranges. In such cases, another type of estimator had to be defined: Complex
estimators.
The methods used to construct Simple or Complex estimators can be used to construct
virtually any numerical value based on field survey values (% positive or negative responses
to over 60 questions) or census values (% area allocated to a specific land use, as compared
to the average for Galicia). The problem comes down to simplifying the information available,
which is too rich and sometimes redundant, and to applying it to a suitability matrix.
For simplification purposes, Composite estimators were constructed. Composite estimators
were used to group (as a mean or weighted mean) the responses to two or several related
questions, such that the value of each estimator provided considerably more information than
the mere survey results or census data.
2.5. Suitability Matrix
The Suitability Matrix included two sets of elements, grouped into ‘Rows’, in which up to 50
different crops or livestock productions were represented, and ‘Columns’, in which up to 44
sustainability indicators were identified and tabulated.
A suitability matrix was constructed for each EEU, such that a total of 88 suitability matrices
were obtained. Each matrix contained data pertaining to the estimators for the relevant area
and assessed such information for every product.
The suitability of the crop or land use ‘n’ was equivalent to the following value:
44
PS n = ∑ WFi xTEi ,n
(1)
i =1
Where:
PSn: Total value of the suitability of crop or livestock product ‘n’ for an EEU, between a
minimum value of 0 and a maximum value of 1000. The matrix construction method allows
for the comparison of the PSn value obtained with the values obtained for other EEUs, such
that the suitability of the different comarcas for producing a crop or land use can be
compared.
WFi: Weighting factor for Constraint i. Weighting factors are used to adjust the relative
weights of the different constraints or limiting factors. Weighting factors are constant for all
the EEU and crops, but can be modified by the user.
62
TEi,n : Transformed Estimator of Weighting Factor i for crop or land use n. The transformed
estimator stands for the value of the sustainability indicator defined to characterize
constraint/limiting factor i, weighted for each specific crop or land use. The weighting factors
defined above are unique because each estimator is weighted for each crop.
By calculating Suitability PSn, a numerical value between 0 and 1000 is obtained that
includes the weighted sum of the different estimators by crop and EEU, and represents the
potential carrying capacity for a crop or land use in an EEU.
This value presents two remarkable characteristics:
- It represents a relative comparison value among the 50 crops/land uses considered. This
allows to set a hierarchy for these productions in the EEU, and so to establish the most
suitable crops/land uses for that EEU, which will undergo the following analysis.
- Suitability Values are obtained through a homogeneous method for all the EEUs, so it is
possible to recognize for which Units a crop/land use is more suitable, thus the suitability
values for a crop or land use can be organized by EEU into a hierarchy.
Each factor was assessed by using the sustainability criteria defined in the above section,
and could be transformed into numerical values that varied for every EEU and crop or land
use.
2.6. Calculation of Transformed Estimators
A key element must be considered in the calculation of transformed estimators: the
sensitivity of every crop or livestock product to each of the issues included in the numerical
values of the estimators. For example, one can obtain a numerical value for the average
slope of a given geographical area, and this value will be included numerically in the Slope
Estimator. However, the effect of slope on the suitability of a crop is dependent not only on
the average slope value, but also on the suitability of the slope value obtained for crop
productivity. Thus, slope values exceeding 10% may impede the introduction of a given crop
in an area, while the same slope range may be suitable for other crops.
Accordingly, the value of each estimator must be adjusted to the sensitivity of each crop or
livestock product. For that purpose, various transform functions were applied. These
transform functions were different for each estimator and for each crop or livestock product
considered within each estimator. By applying the specific transform function for a crop to an
estimator, a new value was obtained: the Transformed Estimator.
63
Table 1: List of Estimators (I).
64
Table 1.List of Estimators (II)
Table 1: Shows the 44 estimators used in the model and the weights assigned to each estimator.
3. Results
To illustrate the potentials of the model, an example is provided below. The example
discusses the results obtained for Baixo Miño, a comarca located in Southwest Galicia; more
specifically, for its EEU-1.
The five EEUs obtained from the spatial zoning process were distributed as shown in Map 1.
65
Figure 3: Division of Baixo Miño comarca into EEUs.
Table 2 includes the results of the suitability matrix for EEU-1, and table 3 the joint analysis
of the matrix for EEU and crop (Mixed classification).
66
Table 2: Suitability matrix for EEU-1
67
Area: Product Engineering
Table 3: Mixed matrix for EEU-1 /crops.
As shown in Table 2, the most suitable crops are horticultural crops, followed by a group of
crops composed of cut flowers, vineyard, mirabelle plum (native variety of plum, Prunus
domestica L. var. syriaca) and ornamental plants. The analysis of the overall assessment of the
results sorted by order (relative classification within an EEU, obtained from the suitability
matrix) and magnitude (overall classification of a crop or land use for all the comarcas
considered as a whole) and summarized in Table 3, coincides with the situation shown in the
previous table, only with some differences in the relative position of crops, which do not affect
the election of the suitabilities assessed. Mirabelle plum ranks first because of the current
weight of the crop, which is limited to the Baixo Miño comarca.
68
The main factors that maximize the suitability of these crops and uses are:
-
Horticultural crops: The EEU presents good weather conditions for this product.
Availability of established commercial channels and a processing industry in the region.
High membership of farmers to associations or groups. The comarca is near seasonal
consumption areas. Availability of hired labour. High potential for these crops.
-
Cut flowers under cover: Proximity to large seasonal consumption areas. Availability
of hired labour. High current weight of the crop.
-
Vineyard: Good suitability of the EEU for this product. Good integration of the crop in
the agricultural landscape of the EEU. Good marketing conditions, favoured by the
existence of a good commercial channel, the strong presence of processing industries
in the region, and the availability of hired labour. PDO Certificate.
-
Mirabel: Good marketing conditions, favoured by the existence of a good commercial
channel, and presence of processing industries in the region. Proximity to large
seasonal consumption areas, and high potentiality of internal markets. Availability of
hired labour. Production of the crop is almost exclusive of the region. High potentiality of
the crop.
-
Ornamental plants: Good integration of the crop in the EEU, as the climate is very
suitable for this product. Good marketing conditions, favoured by the existence of a
good commercial channel and a high potentiality of the foreign markets. Availability of
hired labour. High current weight and potentiality of the crop.
4. Conclusions
Our first consideration for the drawing of conclusions is based on the premise that no model,
regardless of its quality, can replace the work of a technician; we expect it to be an efficient
help.
The model presented in this paper has proved useful in agricultural production planning and
has enabled the Galician Administration, Xunta de Galicia, to define operational measures and
action policies based on the results of the model.
Moreover, comparative assessment methods, which are typical of project engineering, have a
high potential for prioritizing objectives and for using all the information available.
This model can be further developed by introducing more (comprising more aspects) and better
(more detailed) information. The expansion of the model to the whole Autonomous Community
of Galicia would produce more efficient assessments. Furthermore, in the decision process new
indicators can be introduced, or existing ones improved.
Obviously the model must be more deeply tested, and is widely open to new improvements,
particularly in aspects as the weighting coefficients for the indicators or the automatic use of
transformed indicators to obtain an immediate assessment system. The limitations of the model
derive from two main aspects: 1) the need to analyze the weighting of the indicators, i.e. the
relative weight of each indicator in the final result and 2) the use of transformations that
represent the suitability of environmental factors for the characteristics of each crop or land use.
69
References
Álvarez C.J., Riveiro J.A., Marey M.F., 2008. Typology classification and characterization of
farms for agricultural production planning. Spanish Journal of Agricultural Research. Vol. 6 (1),
pp. 125-136.
Andersen E., Elbersen B., Godeschalkc F., Verhoog D., 2007. Farm management indicators
and farm typologies as a basis for assessments in a changing policy environment. Journal of
Environmental Management Vol. 82 (3), pp. 353-362.
De Wit C.T., Van Keulen H., 1988. Application of interactive multiple goal programming
techniques for analysis and planning of regional agricultural development. Agricultural Systems.
Vol. 26, pp. 211-230.
Riveiro J.A., Álvarez C.J., Miranda D., Pereira J.M., 2005. Profitability and production
requirements for land use allocation of farming and forestry land. Biosystems Engineering. Vol.
90 (4), pp. 477-484.
Riveiro J.A., Álvarez C.J., Marey M.F., Marco J.L., Alvarez C.J., 2008. Procedure for the
classification and characterization of farms for agricultural production planning: Application in
the Northwest of Spain. Computers and Electronics in Agriculture Vol. 6 (1), pp. 169–178.
UTE EIDO-USC., 2004. Estudios Comarcales de Ordenación Productiva Agraria de 21
comarcas, Xunta de Galicia. Consellería de Política Agroalimentaria e Desenvolvemento Rural.
Santiago de Compostela.
Acknowledgements
This research has been funded by the Spanish Ministry of Education and Science, General
Research Directorate, CICYT, within the framework of Research Project ‘Integration of
information into a methodological model applied to decision making in the management of
agricultural production planning processes, AGL2006-04789/AGR” (X-2006 to IX-2009)
(National Programme). Research group: GI 1716-Projects and Planning-USC.
Also contributed the FPU Program of the Spanish Ministry of Education and Science.
Marta Cardín Pedrosa
G.I. 1716 - Proyectos y Planificación
Escuela Politécnica Superior. Universidad de Santiago de Compostela.
Departamento de Ingeniería Agroforestal.
Campus Universitario s/n, 27002, Lugo, Spain.
Phone: +34 982 285900 ext 23323
Fax: +34 982 285926
70
DEFINITION OF A METHODOLOGY FOR FOSTERING INTEGRATED
COLLABORATION IN SMALL-TO-MEDIUM COMPANIES OF THE
HOUSING DEVELOPMENT / CONSTRUCTION SECTOR
Santos Fonseca, S.
Encauza; G.I.C.C. Alanta, SL
Abstract
The project aims to create a methodology for managing construction projects, in a way that
area, quality, information technology and legislative factors are taken into account. The
methodology has been developed to facilitate both awareness and definition of the legal
requirements in the area and for each company. It also aims to manage all the documents that
fulfill these requirements through Microsoft Sharepoint Server Services in conjunction with
Small Business Server 2003 and Terminal Server licenses, while also working along with other
software such as PRESTO, CONTAPLUS and Microsoft OFFICE. As a result, the company has
become better prepared to understand its role in society and the importance of INTEGRATED
COLLABORATION for bringing together information that was previously dispersed within the
company. In this way, it integrates different processes of a real estate enterprise, from
choosing the adequate plot for its execution up to its concrete realization, including the
definition of the ground plan.
Keywords: Project Management, Technology of the Information, Systems of Management,
Tools of Management, Quality, Real Estate Promotion and Construction.
1. Introduction
This works aims at developing a methodology for integrating all the distinct processes involved
in the construction sector. It is supposed to be a tool for helping the coordination of different
professionals in each stage of the project, the management of all the knowledge produced
within the company during the execution of a project and the operational and strategic
communication of the guidelines and goals of each different project.
The small to medium sized company would benefit most from this methodology in the area of
development and construction. Its directive body should be clearly interested in enhancing the
deployment of resources in the performance of tasks, be interested in getting to know the gap
between real versus planned for all the activities encompassed in the project and be interested
in a type of structured information, in real time, for facilitating decision-making.
ALANTA, a small-to-medium developer and constructor in need to develop its organizational
capacity through the implementation of information technology tools, has decided to invest in
the creation of a methodology to guide the updating process of its managerial tools. The
71
managerial deficits identified by ALANTA are common to many other companies of its sector
and size. Many companies which have achieved the implementation of a certified system of
management, such as UNE-ISO 9001/2000 and 14001/2004 by AENOR in the case of
ALANTA, still notice that they need to integrate their managerial systems. The fact that
information technology might provide the tool for helping decision-making has prompted in
many companies the willingness to reform their organizational structures in relation to data flow,
collaboration, coordination and adaptation to an ever changing legal environment.
2. The Development / Construction Sector
The Spanish Development/Construction sector has been extensively regulated. There is, in this
sense, an opportunity for keeping legislation as a guideline for designing the data flow of a
company, as each company has the need to organize its internal processes in relation to the
legal requirements that regulate a real estate enterprise.
Undoubtedly, the existence of a specific legislation for this sector demonstrates the importance
of the real estate activity within the Spanish economy. Real estate purchase is normally an
important step in the customer's life, and they, for making such a decision, usually require a
great deal of warranties, state-of-art technological improvements and increasingly a selfsustainable relationship with the environment. All these factors impact upon the legislation
regulating the sector and it becomes necessary for the company to adopt a strong managerial
attitude as to cope with both new and old legal requirements.
Among the extensive legislation in the sector, some can illustrate what has been above said:
− Law 57/1968, regulating the anticipated amounts in construction and sale of housing units
(BOE #181, July, 29th 1968).
− RD 515/1989 about the protection of consumers in relation to information access in the
purchase and rent of housing.
− Law 38/99, Ordination of Edification, November, 5th, published in the BOE# 266, November,
6th 1999 - pages 38925 to 38934.
− Royal Decree 314/2006, March, 17th, bring into force the Technical Code of Edification.
− Ley 32/2006, October, 18th, regulates subcontracts in the Construction Sector.
− Resolution from August, 1st 2007, from the Dirección General de Trabajo, in which it is
incribed in the record and published the IV Convenio Colectivo General del Sector de la
Construcción.
− Royal Decree 105/2008, February, 1st, which regulates the production and management of
debris and refuse from construction and demolition.
Among the examples, its is possible to notice that the legislation refers to the relationships of
development/construction companies with its clients, workers, providers, public administration,
environment and the execution of the work itself. It reinforces, therefore, all the legislation that
is normally applicable to any other company in a different sector.
For better comprehending the operation of the real estate sector, with the help of the Law 38/99
– Regulation of Construction, it is possible to identify all the different agents in the construction
process: the developer, the project designer, the construction site director, the execution
director, institutions and labs of quality control for construction the providers and the
72
customer. Through this list, it is possible to understand that the coordination of collaboration is
one of the key elements for the success of construction project.
In order to develop a methodology for the implementation of an integrated collaboration
platform, one has to follow the usual structures of coordination in construction projects, in
accordance with market practices. Thus, it is possible to distinguish:

Coordination centered around Optional Direction (Construction Director and Execution
Director);
In order to form the Optional Direction, the most traditional way of conducting construction
projects, developers, either guided by their experience or by a preliminary viability research,
hires a Project designer. This designer normally ends up as the Construction Director,
normally indicating a trustful Execution Director, who jointly are the Optional Direction.
The Optional Direction helps the Developer to formalize a construction contract with the
main contractor (constructor) for effectively executing the project. The main contractor
becomes responsible for the execution of the construction and for hiring all necessary
subcontractors. The Optional Direction remains with the technical and economic
coordination of the construction.
Even though the Optional Direction is present in some processes in the project, even
helping the developer to articulate the project and select the constructor, its participation
should not be characterized by a integrated coordination, as it does not take part in all
stages of the project and does not stay close to the decision-making processes.

Coordination centered in the Main Contractor;
In another version, the coordination of the project is through the developer, which becomes
the Main Contractor, managing the effective implementation of the Execution Project and
becoming responsible for contracting all the specialized subcontractors. For designing the
project and forming the Optional Direction, it is usual to follow the same above mentioned
procedure, but the Execution Direction is normally indicated by the Developer or becomes
part of its responsibilities.
In this arrangement, Developers are better structured for carrying out the execution of the
construction. However, the coordination of the project is not yet fully integrated, as there
might be a gap between the technical coordination, the projector and future Construction
Director, and the coordination of the remaining areas of the project, in the hands of the Main
Contractor.

Coordination centered in the Delegated Developer;
In the case there is a Delegated Developer, it is assumed that the entrepreneurs, generally
from outside the sector, are going to take the economic responsibility and seek for support
from someone who can coordinate the project in its entirety. This one would be responsible
for the preliminary assessments, the contraction of a projector for designing a project, the
formation of an Optional Direction, contracting several specialized subcontractors and the
sale (if that is the purpose) up to the sale follow-up.
There is a very professional relationship between proprietor and delegated developer, as
the latter has to be well prepared for providing periodical information about the situation of a
project. In this sense, it would be easier for the Delegated Developer to perform its tasks if
the coordination better integrates all the participants that cooperate in the project and
knows, in real time, the condition of each activity in the project as to report all the
information as readily and accurately as possible.
The development/construction sector is extremely complex and there are many aspect which
need to be further developed. In this work, it has been attempted to provide to small-to-medium
companies a methodology in which theuy can rationalize the use of their resources.
73
3. Objective
Analyse the peculiarities in the real estate sector, mostly identifying its key managerial aspects
and its context.
Develop a methodology which can fit the singularities of the real estate sector, in order to offer
a tool easy to implement and direct application for the directing body of a company in
development and construction.
Implement the methodology in a small-to-large company in development and construction.
4. Methodology and Case Study
4.1 ALANTA
ALANTA is a small-to-medium developer and constructor which has decided to support and
finance an academic research within its working environment as to qualitatively improve its
organizational structure for facing new challenges in a more competitive, complex and
uncertain market. Its CEO, Luis Gracia Adrián, used to say at that moment: “The real estate
market is increasingly more developed, its customers increasingly more demanding, and there
is no room for non-professionalized companies”. In January 2005:

It achieved its credentials from Aenor in both UNE-ISO 9001/2000 and UNE-ISO
14001/2004;

It prepared itself for facing an imminent crisis in the real estate market and attempted to
adjust its internal organizational structure, its internal processes, its operational
communications and enhance its projects in a digital culture, as to create a more
competitive portfolio and maximize its resources;

It kept all its documentation in paper-based archives, even though electronically
generated through spreadsheet and word processors (Microsoft Office);

It was not fully integrated, since all the tasks that were not necessary for the certification
processes were not within the system. This situation appeared to rely on two distinct
and informal systems: one for “quality” and the other for management;

It used to communicate its values through the Management Handbook, whose
hardcopies were sent to every collaborator. Besides costly and environmentally
inappropriate, it required an excessive workload to keep it updated;

It had a technological base with computers connected through a simple network without
client/server protocol. The files were organized within Windows Explorer with a few rules
for the creation of folders. The interpretation of users, however, resulted in a hardly
controllable process, in which documents produced by a third party were not easily
accessed;

It used CONTAPLUS for accounting, PRESTO for project control and Microsoft
OFFICE.
4.2. Definition of a Methodology for implementing a Platform for Integrated Collaboration
In order to face the challenge of developing and implementing a Platform of Integrated
Colaboration, three main goals were set:
74

Check / adapt the continuously compliance with all the requirements that are imposed to
the company and the projects, by verifying the records generated in the execution of
different processes in order to avoid duplication of compliance;

Guide the collaborators of the company through a Map of Processes, which allows for
the inclusion of a new contribution, at any time, of an action within the scope of the
project and the company. On the top of it, the agents should be capable of
understanding the requirements satisfied with each record and the responsibility
originated by the generation of each record;

Shift from a fragmented management, paper-based, to an integrated management,
based on computerized tools, with the possibility of remote collaboration (as the
construction projects are geographically dispersed) in a standardized and organized
setting for facilitating the consult of documents and information of the projects and the
company;
4.3 Reduction by Adherence Layers
The reduction by adherence layers has been planned to facilitate a systematic verification if all
the records in the management system satisfy all the requirements in ALANTA, but also for
enabling an ongoing adjustment to new requirements that come to be enacted later.
Each layer of reference will mirror at least the normative guideline of reference (it is
considered to be an internal normative guideline when the requirements are created within the
company itself), the source of reference and the class of application (Construction,
Management, Environment, Development, Health and Safety, Urbanism) and any other
requirements to be observed.
Requirement
(6)
Requirement
(7)
ren
ce
La
ye
r(
3)
Once identified the reference layer, the definition of requirements will take place in relation to
their application character (legal, those required by the legislation for the sector, and
strategic, those that the company deems necessary to be implemented according to a
strategic normative guideline); the application activity, the execution unity (construction
unities, a stage in the process, or a sub-process), the execution process, the record that
proves its compliance (the document that shows the fulfillment of a requirement); the
responsible for the requirement and the requirement client (the one who demands its
compliance).
ren
ce
La
ye
r(
2)
Re
fe
Requirement
(8)
Layer of Adherence
Requirement
(4)
Registry - 1
(1 y 5)
Registru - 3
(3 y 7)
Registry -2
(2, 4 y 6)
Registry - 4
(8)
Re
fe
ren
ce
La
ye
r(
1)
Re
fe
Requirement
(5)
Reduction by
Adherence Layers
Requirement
(1)
Requirement
(2)
Requirement
(3)
Total Requirements = 7
There is
ADHERENCE
only when they match
Registry, Responsible
of requirement and
Requirement client
Total Registries = 4
75
4.4 The Processes Map
The Processes Map ought to offer any participant the opportunity to place him or her, while
developing a task, at least in relation to:

his/her needs: which documents are necessary for developing the task;

his/her understanding: what has been fulfilling while developing the task;

his/her importance: which client will deploy the result of the task;
However, the Processes Map should work as a “road map” in which at any time the participant
might understand all the possible ways for a project to be borne, grow and turn into a database
(documental flow). It also has to allow for finding out all the criteria that the company needs to
fulfill for assessing its viability (profit margins indexes and strategic necessity) and the criteria
for decision-making that needs to be embodied in the project.
The previous Reduction by adherence layers facilitates drawing the Processes Map. However,
for facilitating the interaction of processes and documents, given the complexity of some
management systems, there should be reminded that the Processes Map might be unfolded
through auxiliary documents.
The work of coordinating constructing projects is very complex as it depends on bringing
together distinct areas of knowledge. It can be from technical areas, as architecture and
engineering, but it may also involve salesman, administrator, accountants and computational
technical support. On the top of it, the level of acquaintance and understanding of those diverse
collaborators varies greatly within a project or company. Therefore, it requires an extra effort
form the company for achieving a Processes Map which can coordinate all these collective
contributions.
4.5 Information Technology
The definition of a software tool which can be used as a Collaboration Platform is not easy to
achieve, as the selection parameters are so variegated and full of technicalities, making it not
user friendly. Hence, following the methodology of Reduction by adherence layers and
designing a good Processes Map enables the company to be in a better position for knowing its
information flow. It makes possible to reduce the information to a minimally necessary flow and
to define those who need it and when it is needed. This has helped to develop a standard for
sensibly evaluating complex tools:
I. In relation to the type of solutions available in the market:
o Classify solutions: Mix of functional tools (integration solution that encompasses
tools for different branches of knowledge) or tools of single database, the so-called
ERP (Enterprise Resources Planning – integration solution for distinct branches of
knowledge in a single tool which defines, internally, the data flow);
o Sub-classify the tools of single database among the so-called developed ERPs
(solutions developed by technology companies from their own needs) and vertical
(solutions developed for a sector from an open source standardized platform).
II. In relation to the level of compatibility between the resources of the application and the
needs of the company (software):
o Establish the processes which should be covered by the software;
o Define the level of compatibility by determining to which extent each software fulfills
the minimum processes of the company
76
III. In relation to the need of resources for the implementation:
o Define the typology of the network and the configuration of the equipment for running
each of the options (hardware);
o Define the amount of time and human resources needed for set-up, implementation
and loading of historical database for getting each solution operative;
o Define the cost of implementation.
IV. Select the most appropriated solution for the company
o Classifying Solutions – assign higher scores for tools of a single database;
o Sub-classifying tools of a single database – assign higher scores for the vertical
and modular ones;
o Level of compatibility – the higher the level of compatibility the the higher the score
and discard solutions whose level the compatibility is below 70%;
o Resources for implementation – higher score for the lower costs;
Results in ALANTA
ALANTA has followed the steps of the methodology in a way that:
■ Application of Reduction by Adherence Layers:
○ It has been obtained the minimum records for complying with all the requirements of
implementation in the company;
○ it has been identified about 200 records to comply with all the requirements, remaining
near 75 records for construction projects, 70 records for development projects and 55 records
for the management of the requirements of a project called Administration. However, the
number of necessary records may change based on some characteristics of development or
construction projects;
○ It has been proved easy to adequate records of the system by changing both the Layers of
Reference and the requirements of application, as there were regulatory changes over the
implementation period;
■ Map of Processes: It has been noticed that, when adjusting the Map of Processes to
premises of the methodology, it has been transformed into an efficient browser for situating
collaborator in relation to the task that is going to be performed. The collaborator has all
information, in a structured way, to carry out the task, generating greater fluidity in
operational communication with the recipient of the work.
the
the
the
the
■ Information Technology:
○ it has been defined the process of documentation management as the most important to
encourage a more integrated collaboration in ALANTA;
77
o
it has been taken into account that ALANTA already worked with PRESTO,
CONTAPLUS, MS Project, as well as Microsoft Office;
o
few options provided a solution with a Level of Compatibility above 70% (adopt a
solution with a low Level of Compatibility means either to relinquish some features
already planned or incur a higher cost for adaptation);
o
the solutions were still very focused on the accounting aspects of the projects (economic
and financial management) and deficient in the management of flows and documents;
o
in general the applications did not encompass both the management of development
and of construction, or they were very strong in one area at the expense of the other;
o
some solutions for the management of development projects proposed the integration
for the management of construction with the use of PRESTO;
o
the need to invest in new hardware and operating systems would cost between 10,000
and € 20,000 €;
o
the cost of the solutions in information technology was too broad, starting from
30,000.00 € up 240.000,00 €;
o
the solutions that showed a Compatibility degree of more than 70% were above of
80,000 €;
o
the time of implantation and burden of historical data for the implementation of a new
solution was not less than 90 days, but it could reach up to 180 days;
4.7. Platform of Integrated Collaboration in Sharepoint Server Services
It was not possible to find tool that could adjust itself 100% to ALANTA expectations in relation
to cost/quality among the 11 softwares analysed. Hence, it has adopted a strategy of
introducing small changes for later facilitating the impementation of a management tool, giving
time to the market for providing a more complete and less expensive management tools in the
meantime. These small changes required the company to:

Base the internal network in a client/server protocol with a powerful server and a good
system of backup;

Renew the hardware as to allow everyone to work with the same operational system;

Enable the remote connection to ALANTA's server through a Virtual Private Network
(VPN) and Terminal Server;

Integrate PRESTO and Contaplus, while acquiring more licenses from PRESTO;

Develop an intranet for the document management on Sharepoint Serveices Server,
which comes without charge within Microsoft Server 2003.
In order to take the results of the Reduction by adherence layers and the Map of Processes into
a structure for document management, it has been necessary to:
-
Create lists for Reference Layers, Application Requirements, Processes, records,
Projects and Users;
-
Create a list for filing all documents created or received in ALANTA in terms of the
projects where a document belongs, the type of record it satisfies and the stage in which
it takes place;
-
Define windows for Project management, Process execution and management of each
agent records;
78
-
Set up Warnings for telling each participant of new entries, changes or elimination of
documents.
In Figure 2 it is possible to see a diagram of the relationship between lists by their attributes
and a suggestion for the minimally necessary attributes which should be defined for each type
of list.
Layer of Adherence
Requirements of Application
Regulation
Character of Application
Source
Activity
Class of application
Execution unity
Requirements
Records
Process
Records
Process
Type
Class
Responsible
Execution unity
Process client
Projects
Type
Status
Filing
User
Project
Name
Record
Position
Document
Status
Sharepoint’s Lists
Relation between Lists
Sharepoint’s Alerts
Figure 2: Relationships between lists and its attributes.
Figure 3: Screen of Management of Documents in a Project.
In Figures 3 and 4 are reproduced the computer screens generated for the project
managements and execution of processes. For the project management, all the documents of
each project can be observed by processes and records, as they can also be filtered by
79
projects and by a tab in which the abstract of processes can be accessed. In the project
execution screen, there are a number of windows for browsing the different stages of the
process, including entry registers, exit registers and also the description of each stage in the
process in interactive way that offers access to systems forms.
Figure 4: Screen of Process execution.
4. Conclusions
The challenge faced in this work has brought many advantages, as it has been possible to
implement, besides what had been originally planned, other functions which have
complemented the Integrated Management System in ALANTA. There is no doubt, though, that
it has only been possible to manage the reluctance to changes with the dedicated support from
the company direction throughout the process of implementation.
However, it is important to bear in mind that this methodology is a cheap alternative for smallto-medium companies aiming at digital management. The SSS has two important
disadvantages in relation to ERPs, as it is not a tool with a single basis and the SSS lists are
not guided in a relational way. Nowadays there are a lot of management tools, more
encompassing and with lower costs of implementation, than at the time this study was
conducted. It is important to keep in mind, though, that even with the increase of open source
tools, also for ERPs, the lessons of this work can also be applied in these new scenarios.
In relation to the implemented tool, ALANTA has already managed two construction projects.
The first is in Barbastro (Huesca), which is in its concluding stage, and the other is in Labuerda
(Huesca), already concluded. It has also been deployed in two projects of development and
construction, one in Arròs (Lérida), concluded, and the other in Laspuña (Huesca), under
execution. It has also managed a number of preliminary analysis of construction and
development/construction projects which have been turned down or under negotiation. Besides
these constructing projects, ALANTA manages all its administrative documents as just another
project, named Administration. All these projects have taken place concomitantly with the
development of this tool.
80
The experience we had up to now offers some important conclusions:
9 It is increasingly imperative that companies have more than simply spreadsheets and
word processors in their range of computational tools, incorporating also standardized
platforms of open source. “The Sharepoint Services Server” has proved to be easy to
implement and a great potential for developing business;
9 It has been noticed that ALANTA has become much more flexible in adapting itself to
changes in normative and market frameworks, rendering the company much more
competitive. The easiness in which these adaptations took place was certainly due to
the readiness of the tool for incorporating such shifts in its structure (recent legal
changes, real estate crisis, new requirements for public contracts);
9 In ALANTA, the management system has been capable to integrate corporative
language and improved operational communication;
9 Each new project in ALANTA has improved in terms of quality, as it became easier to
check the flow of documents of past projects and provide readily information.
Finally, it has been possible to implement in ALANTA a tool which enables INTEGRATED
COLABORATION in construction projects. This tool presents itself as an option for other
organizations, as these companies in construction and real estate will always have an
enhanced position if they are formally structured to follow the current legislation, to produce
high quality items and to keep the balance with the environment in which they operate.
References
Buyens, J., (Traducción: Eloy Pineda Rojas); 2006. Windows SharePoint Services. Translated
from the first English edition of Windows sharepoint services inside out Mexico: McGraw-Hill
Interamenricana.
Instituto Nacional de Estadística Español (INE) www.ine.es
Kaplan, M. and Norton, D. (Traducción: Adelaida Santapau); 1997. Cuadro de Mando Integral –
The Balanced Scoread. Barcelona: edición Gestiones 2000, S.A.
Lock, D. 2003. Fundamentos de la gestión de proyectos. Spain: AENOR.
Scasso, R.H. 1999. Dirección Integrada de Proyecto – DIP - Project Management. Madrid:
Publicaciones de la E.T.S. de Ingenieros industriales de la Universidad Politécnica de Madrid.
Vargas, R.V.; 2000. Gerenciamento de Projetos. Rio de Janeiro – Brazil: Brasport.
Salazar Santos Fonseca
Ph.D. Candidate in the Inter-University Doctoral Program in Project Direction at Navarra
University and Knowledge and Management Consultant at Encauza.
Plaza Fernando El Católico, 2 - bajos, 22300 Barbastro - Huesca (España).
Phone: (+34) 974 31 42 12
81
ANALYSIS OF THE FACTORS THAT CONDITION THE EXPANSION
OF A COASTAL VILLAGE THROUGH CELLULAR AUTOMATA
SIMULATION
García, A.M.
Santé, I.
Crecente, R.
López, F.J.
Abstract
Cellular automata, conceived by the mathematician Stanislaw Ulam and the physicist John von
Neumann, have proven to be good tools for implementing models to predict and analyze land
use change and urban growth dynamics. The implementation of such tools requires knowing
the singularities of the processes to be simulated. Most of the models implemented so far have
been concerned with the expansion of big cities but little has been done in relation to small
urban areas. In this paper, we analyze the variables that govern urban growth in Ribadeo, a
town of about 9000 people in northeastern Galicia, Spain, using logistic regressions and a
simple cellular automata model that will allow us to study the factors that must be considered
when performing simulations with these tools in small urban areas.
Keywords: Cellular automata, urban growth, variables.
1. Introduction
1.1. Urban simulation background
In the late 1950s and early 1960s, a stage of sociopolitical development arose, which resulted
in the growth of big cities. Such urban growth led to the appearance of new problems in urban
areas, mainly related with transport. The need to scientifically analyze these new problems led
to the development of the first models for the study of traffic. Later, land use allocation models
were developed that were eventually coupled with the former, constituting integrated models.
In the 1970s, new growth dynamics emerged due to the rapid development of urban areas.
These new processes were more complex than the processes that inspired the creation of
integrated models. To make the study of such a complexity more accessible, the models were
divided into subsystems, avoiding the establishment of too many relations between them. This
led to a reduction in the capability of the models to simulate responses to changes in the
82
dynamics. Models became complex, they needed a great amount of input data and provided a
very general view that was not really useful to planners. All this led to the need to use new
modeling methods that could overcome such limitations.
With the development of computer science in the 1970s and the consequent increase in the
computing power of computers, using new methods became possible. In 1970, Waldo Tobler
developed a new population growth simulation model based on cellular automata (CA). With
the advances in computers in the 1980s, the interest in the study of cellular automata increased
and Helen Couclelis established the theoretical basis for the application of such paradigm to the
simulation of urban growth (Berling-Wolff and Wu, 2004).
1.2. Cellular automata
Cellular automata were developed by the mathematician Stanislaw Ulam and the physicist John
von Neumann to study self-reproduction and the modeling of biological processes. In the
1970s, the English mathematician John Hotton Conway created “The game of life”, the first
application of a cellular automaton aimed at reproducing the growth of an animal population.
The model was composed of a grid in which each cell could adopt two states; “alive” or “dead”,
according to some transition rules that determined that:
•
If a grid cell was dead and surrounded by 3 living cells, the cell was born.
•
If a live cell was surrounded by 2 or 3 live cells, the cell kept living.
•
If a live cell was surrounded by less than 2 live cells, the cell died of loneliness and if it was
surrounded by more than 3 live cells it died of competence.
These rules were applied at discrete time intervals, such that at each iteration of the model all
the cells were updated simultaneously.
The game of life is a formal cellular automaton. Therefore, CA are composed of:
•
A grid or raster space.
•
A group of states that define the cells of the grid.
•
A definition of the neighborhood of each cell.
•
A series of transition rules that define the state of each cell based on the states of the
neighboring cells.
If the game of life is iterated several times, complex patterns emerge. This complexity arising
from underlying simplicity makes CA a powerful tool for the simulation of urban growth because
of their emergent behavior, given that CA can exhibit complex global dynamics arising from
simple rules on a local scale.
Cellular automata are able to reproduce growth patterns with fractal geometry, which can be
found in most cities. Because CA work on a grid, they facilitate the use of raster data obtained
with a geographical information system (GIS) or remote sensing techniques.
The simplicity of cellular automata and their ability to simulate urban growth allow us to easily
analyze the processes simulated by them. For this reason, CA are useful tools for research. A
number of authors (Sui and Zeng, 2001; Wu and Martin, 2002; Cheng and Masser, 2004;
Aguilera, 2006) have studied the processes that control the growth of large cities using cellular
automata. Almost all the existing studies tackle the growth of large urban areas, but only a few
models have been developed to study the dynamics of small towns. In this work, CA are used
83
to analyze urban growth in Ribadeo, a coastal municipality on the border between Galicia and
Asturias, Northwest Spain.
Earlier in this section, we defined a formal CA using the Conway’s game of life and its basic
rules as an example. Formal cellular automata do not have the abilities required to correctly
simulate urban growth processes. Consequently, some relaxations to the formal rule should be
adopted in order to obtain more accurate results (Verburg et al., 2004; White, 1997). To design
a model that can produce accurate results, the variables and processes that influence urban
growth processes and interactions must be known.
In this paper, we have performed an analysis based on the research by Sui and Zeng (2001),
Wu and Martin (2002), Cheng and Masser (2004) and Aguilera (2006) in order to identify the
variables that may condition urban development using logistic regression techniques. Such an
analysis allowed us to select the most important variables to include in a CA based model and
to determine its capacity to simulate urban growth. Results were analyzed and some
conclusions were drawn from the results.
1.3. Methods
Ribadeo is a municipality of 9619 people located in the province of Lugo, Spain, on the border
between Galicia and Asturias. The urban core of Ribadeo constitutes the administrative and
service center of the area.
In 1987, the Puente de los Santos Bridge was built over river Eo estuary, communicating the
north of the municipality with Asturias. This infrastructure brought added value to Ribadeo and
the village became the service center of the Asturian municipalities on the other shore of the
estuary. The construction of the bridge fostered the growth of population and commerce. New
infrastructures like the Transcantabrian motorway will attract more growth to the area.
To perform the present analysis, photointerpretation techniques were used to obtain land use
maps of Ribadeo and its four surrounding parishes for the years 1995 and 2003. These data
were converted to raster format with a pixel size of 25x25 m.
Later, we considered the variables that could be used according to the available data. After
having analyzed other models, we decided to consider the following variables:
•
Accessibility (distance to roads).
•
Height.
•
Distance to forested land.
•
Distance to main urban core.
•
Distance to peripheral urban cores.
•
Distance to the shore.
•
Distance to main roads.
•
Distance to secondary roads.
•
Distance to other roads.
•
Distance to railways.
•
Distance to railway station.
84
•
Distance to port.
•
Distance to residential land uses.
•
Distance to commercial land uses.
•
Distance to industrial land uses.
•
Distance to parks and green areas.
•
Distance to institutional land uses (educational centers, cultural centers, administrative
buildings, public sport facilities, welfare and health centers and churchs).
•
Orientation.
•
Slope.
•
300 m radius neighborhood. To calculate the neighborhood, we have considered a square
region around each central cell with a cell radius equivalent to a distance of 300m
measured between the edge and the center of the area. In this area, we applied equation
(1), where I is 0 if the cell is non-urban land and 1 if the cell is urban land; w is a coefficient
that weights the value according to the distance of each cell in the neighborhood to the
central cell, computed as follows: the neighborhood window is divided into several
concentric squares, separated by a cell, w will have the same value in each square and it
will decrease with the increase in distance to the central cell. The decrease in the value of
coefficient w will be related to a line whose slope has the same value as the fractal
dimension2 of Ribadeo village.
N = ∑w× I
(1)
•
Zoning (the current urban planning instrument in the municipality is the Local Development
Plan 1977, which establishes only the zoning of Ribadeo urban core).
•
Shape index. The shape index is a spatial index that defines a relation between the
perimeter and the area of a patch on a map (equation 2). In the present study, the areas of
built land were defined and the value of the shape index of each urban patch was assigned
to the nearest non-urban pixels.
IF = perimeter / 4 area
•
Area of the cadastral parcels.
•
Shape index (equation 2) of the cadastral parcels.
(2)
Following the method used by Aguilera-Benavente (2006), we used Idrisi software to estimate
logistic regressions for each variable, using a binary map that considered the built up area
between the years 1995 and 2003 as independent variable. Roads were not considered in
these maps. A mask was used to exclude the following elements from the regressions: built up
cells in 1995, water surfaces and the 100m coastal fringe protected by the Spanish Coastal Act.
Such elements were excluded from regressions because the variables present in those zones
2
The fractal dimension is determined considering ring areas concentric to the urban core center, so that
the ratio that relates built up area and total area is calculated in each ring. Then a line is fitted to the
values of the ratios obtained. The slope of the line is the fractal dimension of the urban area considered.
85
did not influence urban development during the period considered. In order to reduce
computation time and spatial dependence between variables, not all the cells were used in the
regressions. Systematic sampling was used to choose the cells that would be considered in the
regressions.
Logistic regression considers the probability of a cell being urban and determines the relation
between such a probability and the variables considered by fitting the following equation (3):
P( y = 1 | X ) =
exp(∑ BX )
1 + exp(∑ BX )
(3)
where:
•
P is the probability of the dependent variable being 1 (urban).
•
X are the independent variables. X = ( x 0 , x1 , x 2 ..., x k ); x 0 = 1
•
B are the estimated parameters. B = (b0 , b1 , b2 ..., bk ).
By linearizing the equation is linearized, the following expression is obtained (equation 4):
Ln( P /(1 − P)) = b0 * x 0 + b1 * x1 + b2 * x 2 + ... + bk * x k + error
(4)
This equation can be fitted by using a linear regression.
The results were analyzed by using two parameters:
•
Pseudo R2 indicates the degree of fit of the regression. If the value was greater than 0.2 the
fit was considered to be good.
•
ROC considers the degree of relation between the independent variables and the
dependent variable. If the value was greater than 0.5 it was considered that there was a
dependence.
•
Once the variables were analyzed, those with the highest ROC were chosen for use in a
multiple logistic regression model, thus allowing us to know the weight of each variable in
urban growth.
Then, we used the weights obtained and Cheng and Masser (2004) model as a reference, and
applied the following transition rule in the cellular automaton (equation 5):
P = (∑ ri × Xi + n × N ) ×ν
(5)
Where:
•
P is the probability of change.
•
ri is the coefficient of each variable i obtained by logistic regression.
•
Xi is the value of variable i.
•
N is the neighborhood.
•
n is the neighborhood coefficient obtained by logistic regression.
•
ν is the stochastic variable, which introduces randomnesss in the system and is calculated
from equation (6):
86
RA = 1 + (− ln γ )
α
(6)
Where γ is a random number between 0 and 1 and α is a coefficient that controls the degree of
randomness introduced in the model. The value of α used in the model was equal to the fractal
dimension of Ribadeo (Aguilera, 2006).
The urban area map for 1995 was used as a starting point, and the model was iterated a
number of times equivalent to the number of years between 1995 and 2003. At each iteration,
the urban area was determined by dividing the built up area in the considered period by the
number of years. At each iteration, the cells with the greatest probability of urbanization
changed to urban use until the previously calculated area was reached.
Variable
ROC
300m radius neighborhood
0.8463
Distance to institutional land uses
0.8162
Distance to main urban core
0.8049
Distance to commercial land uses
0.7996
Distance to forested areas
0.7966
Distance to main roads
0.7857
Distance to railway stations
0.7827
Distance to residential land uses
0.7773
Distance to port
0.7734
Distance to industrial land uses
0.749
Accessibility
0.7461
Height
0.736
Distance to other roads
0.7356
Distance to parks and green areas
0.7336
Distance to peripheral urban cores
0.7233
Zoning
0.6759
Distance to the shore
0.6719
Shape index
0.6667
Distance to railways
0.6557
Distance to secondary roads
0.6106
Area of cadastral parcels
0.6057
Shape index of cadastral parcels
0.5864
Orientation
0.3793
Slopes
0.0401
Table 1. Variables chosen to perform the simulation.
The variables with the highest ROC were chosen (Table 2).
87
From among the variables included in Table 3, distance to main urban core, distance to railway
station and distance to port were eliminated because of their correlation with distance to
institutional land uses. All these elements are close to one another and near Ribadeo urban
core. The variable ‘distance to institutional land uses’ was not removed because it showed the
highest ROC. In addition, distance to commercial land uses, distance to industrial land uses
and distance to residential land uses were eliminated because they were correlated with
neighborhood, whereas distance to main roads was removed because it was related to
accessibility.
A new logistic regression was calculated to obtain the relative weight of each selected variable
and to determine its influence on land change from non-urban to urban (Table 3).
Variables
Coefficient
Error parameter
-2.06422057
Neighborhood
2.15979110
Accessibility
-24.39321381
Height
-3.63615921
1.34067087
-3.53720871
Table 2. Coefficients obtained for each variable by multiple logistic regression. Negative coefficients suggest that the
higher the value of the variable, the lower its influence on urban growth, whereas positive values indicate the
opposite relationship.
•
ROC = 0.8760
•
Pseudo R2 = 0.2260
Because the value of pseudo R2 is greater than 0.2, it can be considered that the fit is good.
The value of ROC is also high.
In table 3, we can observe that distance to roads plays a key role in the urban growth process.
The fact that neighborhood does not have as much weight in this model as in other revised
models may be due to the sprawl of growth during the study period.
The coefficients of the variables were rounded and applied in equation 4. After 8 iterations,
results were crossed with 2003 data and analyzed (figure 1).
88
Figure 1. Comparison between real data and simulation results with a cell size of 30x30 m. Red cells: non-urban
land, black cells: urban land in 1995, yellow cells: urban land in 2003 unpredicted by the model, green cells: wrong
predictions, orange cells: correct predictions of the model.
With this model, we did not intend to make an accurate simulation of reality, but to observe the
growth patterns obtained in order to be able to analyze the growth dynamics that generates
such patterns and draw some conclusions about which elements should be considered when
implementing models in similar areas. This is the reason why visual analysis of data is
considered more important than statistical analysis.
The results suggest that the ring road north of Ribadeo attracts a lot of growth. However, in
reality, this road gives access to Puente de los Santos and is fenced. Therefore, there is no
direct access to it, such that it does not attract growth. For this reason, the model was run
another time without considering the ring road in accessibility maps.
Removing the ring road modified one of the variables, such that its weight in the model varied.
Consequently, the logistic regression was performed a second time to calculate the new
coefficients. The following results were obtained (Table 4):
Variables
Coefficient
Error term
-2.00074547
-4.35845586
Height
-2.71871732
2.35060711
Accessibility
-29.20497587
Neighborhood
1.09300021
Table 3. Coefficients of each variable calculated by logistic regression, using the variables selected for the model. In
this simulation, the ring road has not been considered.
After the model was run a second time and the results were crossed with the urban land map
for the year 2003, we obtained the following results (fig. 2):
89
Figure 2. Comparison between real data and simulation results with a cell size of 30x30 m, not considering the ring
road. Red cells: non-urban land, black cells: urban land in 1995, yellow cells: urban land in 2003 unpredicted by the
model, green cells: wrong predictions, orange cells: correctly predicted cells.
The results of the simulation have improved. For example, the growth that appeared in the
small villages along the provincial road heading northwest and inside the urban core is
approximate to the actual core growth trend.
In addition, we can observe unpredicted growth to the southeast and southwest of the urban
core. Maybe these areas were favored because their slopes are oriented to the estuary. In the
first case, the location of the parcels near the urban center and the sea could have favored the
urbanization of these parcels. In the second case, the parcels are near Ribadeo and they are
well communicated with the village.
These factors should be considered in further analysis, including the visual basins and the
proximity to the shore in the analyzed variables, despite the fact that the analyzed variables
show a lower ROC than the variables used. The results of the model could also be improved by
considering different land uses in the simulation in addition to urban and non-urban land. Thus,
more detailed models could be implemented, which could better reproduce the dynamics and
avoid the mistakes derived from assigning the same growth attraction power to all urban land
uses or by considering that the variables influence the location of the different land uses in the
same way.
2. Conclusions
The results of this work have allowed us to draw some conclusions about the future
development and application of more suitable CA based models for the simulation of the
singularities of the Galician urban system.
Logistic regressions show that one of the variables with the highest weight in the model is
distance to roads. After having run the model, we can observe some distortions derived from
considering that all roads attract urban growth in the same way. To improve the results,
90
different road proximity variables based on the kind of road should be used. In the case of ring
roads and highways, the distance to access points could be used as a variable.
Neighborhood is the variable that most influences urban growth, but its relative importance with
respect to other variables inside the model is low. In the neighborhood used in this study, only
the proximity to urban land has been considered and no difference has been made between the
different land uses that constitute urban land. If we analyze variables such as proximity to
industrial, commercial or residential areas, we observe that they are closely related to growth.
However, such variables have not been used because we have assumed that they were
represented in the neighborhood. Models will be more accurate if the influence of different land
uses on the neighborhood is considered.
All these conclusions point to the need to develop more complex models that better simulate
growth in small urban areas.
Because most of the CA models used for urban simulation have been applied to the analysis of
large cities, they use a coarser spatial scale than the scale used in this work. Besides, a
relatively small time scale has been considered because meaningful growth cannot be
appreciated in an 8-year period. These two issues hamper the application of CA models to the
simulation of growth in characteristic Galician urban areas.
To overcome this problem, wider temporal and spatial scales should be used when designing
cellular automata for predicting the growth of small urban areas, such that the patterns are well
defined and allow for simpler analysis and identification. Thus, we will be able to make
simulations with simpler models that can be better analyzed.
However, we must not forget that by increasing the time scale we may cover different periods in
which growth may depend on different variables and, consequently, the model should be
calibrated for each of these periods. The same applies to the use of small scales; since the
geographical area covered is greater, there will be more heterogeneity in the processes and
variables that condition urban growth. For this reason, it is necessary to consider a balance,
defining more or less homogeneous areas and considering temporal spaces where changes in
urban trends have not occurred, such that the performance of the models is good and reliable
predictions can be made.
References
Aguilera Benavente F., 2006. Predicción del crecimiento urbano mediante sistemas de
información geográfica y modelos basados en autómatas celulares, Geofocus, Vol. 6, pp. 81112.
Berling-Wolff, S. and Wu J., 2004. Modeling urban landscape dynamics: A review, Ecological
Research, Vol. 19, pp.119-129.
Cheng J. Q. and Masser I., 2004. Understanding spatial and temporal processes of urban
growth: cellular automata modelling. Environment and Planning B-Planning & Design, Vol. 31,
pp.167-194.
Sui, D. Z. and Zeng H., 2001. Modeling the dynamics of landscape structure in Asia's emerging
desakota regions: a case study in Shenzhen. Landscape and Urban Planning, Vol. 53, pp.3752.
91
Verburg P. H., Schot P., Dijst M. and Veldkamp A., 2004. Land use change modelling: current
practice and research priorities. Geojournal, Vol.61, pp.209-324.
White R. Cities and cellular automata. Discrete Dynamics in Nature and Society, 1997, pp. 111125.
Wu F. L. and Martin D., 2002. Urban expansion simulation of Southeast England using
population surface modelling and cellular automata. Environment and Planning A, Vol. 34,
pp.1855-1876.
Andrés Manuel García Lamparte
Laboratorio do Território. Escola Politécnica Superior de Lugo. Pavillón II
Rúa Benigno Ledo s/n. Campus Universitario s/n. C.P. 27002 Lugo, Spain
Telephone number: +34 982 25 22 31 ext. 23292
Fax: +34 982 28 59 26
URL: http://laborate.usc.es
92
DEVELOPMENT OF AN EXPERIMENTAL PROCEDURE TO ANAYLSE
THE TEMPERATURE FIELD IN THE WELDED JOINT OF
STRUCTURAL STAINLESS STEEL
Miguel, V.
Universidad de Castilla-La Mancha
Estrems, M.
Universidad Politécnica de Cartagena
Martínez, E.
Martínez, A.
Universidad de Castilla-La Mancha
Abstract
In this work, an experimental procedure to determine the temperature distribution in sheet buttwelded joints has been established. Related to light structures, the experiments are focussed
on stainless steel elements welded by GTWA procedure. The proposed methodology lets the
authors analyze the influence of welding parameters on the heat affected zone. The
experimental method can be applied to other types of structural steels and arc welding
processes.
Keywords: Temperature field, electric arc welding, metal structure, stainless steels
1. Introduction
The metallic structure is a design and construction alternative which over the last few decades
has been developing into a structural solution. Nowadays its use has been fully consolidated.
The variety of industrial applications of welding systems have led them to substitute the
traditional nuts and bolts and rivets in the construction sector, not just because welding is an
easier and faster operation, but also because the draughtsman can combine sheets, plates,
bars, tubes, profiles etc. limitlessly, to apply a great number of design possibilities which will
enable to improve the relation between resistance/weight and stiffness/weight with greater
economic viability (Benhayon, 1994).
Stainless steels are very important in the construction of equipment for the process industry as
well as for building. These steels are used instead of conventional ones due to their excellent
properties, such as: resistance to corrosion, hardness as low temperatures and good properties
at high temperatures.
93
The high resistance to corrosion of stainless steels is due to the passivation which they go
through with the formation of a surface film of impermeable chrome oxide which isolates the
steel from the corrosive medium.
Stainless steels have a thermal conductivity lower than that of carbon steels, which leads, in
comparative terms, to more pronounced temperature gradients and greater permanent
deformation. Applied to the welding operations, a slower diffusion of the heat through the base
metal represents that the welded zone will remain hot for a longer time and permits the
possibility of precipitation of chromium carbides on the grain edges if the transition from 800ºC
to 500ºC during cooling is not carried out sufficiently rapidly. Moreover, due to the greater
concentration of the heat in the zone of the bead, less heat is required to produce the fusion, as
compared to the welding of ordinary construction steels; which means that in similar conditions,
the welding speed can be greater (Gomez de Salazar, 2003).
In relation to the coefficient of thermal dilatation, this is much higher in stainless steels than in
carbon steel, which generates a greater complication for their welding, from the point of view of
deformations and residual tensions after cooling (Mazur et al., 2002).
In electric arc welding processes, the energy is applied in a localized zone, reaching
temperatures far superior to the fusion temperature of the base material. The large temperature
differences which are established between the zone where heat is applied and adjacent zones,
together with the good thermal conductivity, in general, which the metals to be welded show,
originates an important thermal flow which conditions the behavior of the welded structures,
modifying the properties which result, both in the weld bead as well as in the zones near to it.
When the harmful effects that the thermal cycle of the welding can produce in the welded piece
are analyzed, the interest does not lie necessarily in the zone of fusion or the bead, but also in
the zone near to this, denominated the heat affected zone (HAZ), which usually presents a
greater degree of weakness in its mechanical properties than the bead itself. Therefore,
knowledge of the evolution of the temperature in the vicinity of the weld bead, during the
welding process and in the cooling process, is essential in order to determine the degree to
which the material is affected and serves to establish strategies which minimize these thermal
effects, such as using preheating, post-welding thermal treatments, the positioning of hot
points, etc.
The principal results that can be determined with the experimental thermal analysis are the
evolution over time of the temperature field in any point of the pieces to be welded, as well as
the isothermal map of any instant during the welding process. The corresponding effects of the
temperature on the material are quantified in different conditions (González et al,2006; Mazur et
al, 2002; Alhama et al, 2005).
The present study analyses the results of the temperature field of a welded joint of AISI 304
structural stainless steel by means of the TIG technique without the contribution of material.
2. Materials
The material tested experimentally is AISI 304 austenitic stainless steel, X5CrNi18-10
according to the UNE norm (Aenor, 2006). The fusion point for this steel is estimated to be at
1400º C and the chemical composition of the steel tested is indicated in table 1. In tables 2 to 4
the physical properties are shown (Aenor, 2007); the variation in mechanical properties with the
temperature; and the thermal properties at different temperatures. The test-tubes used for the
tests are 75 mm long and 30 mm wide, the thickness varies depending on the test carried out.
94
I
N
R
.07
%
%
%
.045
.015
.11
8
%
%
%
%
I
%
Table 1: Composition of the AISI 304 steel.
Density
Thermal
Thermal Capacity
Electrical
Conductivity
7.9 kg/dm3
15 W/m K
Magnetizable
Resistivity
0.73 Ω .mm2/ m
500 J/ kg K
No
Table 2: Physical properties of the AISI 304 steel.
Temperature (º C)
20
100
200
300
400
500
Elasticity Modulus( GPa)
200
194
186
179
172
165
Table 3: Values of the elasticity modulus for the AISI 304 steel at different temperatures.
Temperature º C
100
200
300
400
500
Thermal Dilatation C.(10-6 K-1)
16
16.5
17
17.5
18
Table 4: Thermal dilatation coefficient of the AISI 304 steel at different temperatures.
3. Welding Procedure
The welding procedure used is that of electric arc with protection by inert gas and nonconsumable electrode or TIG procedure, also denominated GTWA, habitually utilized for the
welding of stainless steels. As the protection gas, commercial argon C-50 has been selected,
supplied by Carburos Metálicos. The volume of gas utilized is 12 l/min., and the welding has
been done without contribution material. The welding has been carried out “in position”, that is
to say, in a horizontal position. The electrode employed is made of thorium-lanthanum of 1.6
mm in diameter, with a fusion point estimated to be at 4000ºC. The type of current selected is
continual with direct polarity. At all moments the point of the electrode has been guaranteed to
be sharp according to the established norms for the type of current employed to maintain the
arc stable. A small electrode diameter has been selected in order to concentrate the arc and
obtain a reduced fusion bath. The zone of the welded joint was superficially prepared with a
brush with stainless steel bristles to eliminate any type of surface oxidation which might
complicate the stability of the electric arc during the welding. Finally, to eliminate possible
95
traces of dirt from the base material, acetone was used to clean the area to be welded. Due to
the thickness of the plate, the welding could not be carried out without preparing the edges.
The welding variables considered are indicated in table 5.
I
A
Intensity of the current
V
V
Voltage
e
mm
Thickness of the plate
v
m/s
Lineal velocity of welding
Tabla 5.Experimental welding variables.
4. Description of equipment and instrumentation
4.1. Thermocouples
Initially, type K thermocouples were used encased in a sheath of stainless steel, figure 1,
although finally the measurements were made with type K thermocouple wire to increase the
response speed. To improve the contact between the tip of the thermocouple and the base
material, the plate was drilled to insert the wire, as can be seen in figure 2.
Drilled
Thermocouple
5 mm
Thermocouple
Figure 1: Encased thermocouples.
Figure 2: Thermocouple Position.
To measure the real distance between the drills made on one side of the bead, a profile
projector has been used. Figure 3 shows three drills of 1 mm made in one of the tests to insert
the thermocouples.
4.2. Temperature logger
For the measurement of the plate temperature a USB TC-08 logger for thermocouples has
been employed, connected to a PC which permits compiling, analyzing, and visualizing the data
from eight channels. Moreover, the data logger offers connection through USB port. To log all
96
the temperature data that are registered with the TC-08, we utilized Picolog commercial
software for Windows 5.15.6. It has a wide range of temperature scales, with a vertical
resolution of 20 bits.
HAZ
Bead
5 mm
5 mm
5 mm
Figure 3: Distance thermocouples-bead.
4.3. Welding equipment
The equipment employed (figure 4) is a transformer-rectifier; model MAGIC WAVE 2600, made
by Fronius. This equipment permits the execution of manual TIG welding, as well as with
coated electrodes and allows the possibility of choosing the type and polarity of the current. The
electricity tension is rectified by means of transistors and the control is Fuzzy type.
4.4. Control of the welding speed
To maintain a constant welding speed a milling machine has been used with a tool made
expressly to house the welding machine’s torch fitted into the head. In this machine we can
select several forward velocities and thus consider that as a test variable (figure 4).
4.5. Support for the base material
To carry out the tests a wooden support has been produced to hold the plate and thus avoid
deformations during the welding. The support is housed on the milling machine table. Wood
has been chosen for the support in order to avoid the existence of heat losses by conduction.
5. Description of the test method
The tests are initiated with the preparation of the base metal to be welded, as indicated in
section 2. The following step consists in positioning the thermocouples in the base metal. It
must be ensured that the thermocouple remains in contact with the base metal in order to
97
guarantee the correct measuring of the plate temperature at all times. Different methods have
been adopted for the positioning of the thermocouples but finally that indicated in figure 1 has
been selected. To guarantee the contact during the initial manipulation of the plate, adhesive
resin has been employed.
Milling
machine
Torch
Bench
Welding
machine
Thermocouple
Data logger
Figure 4. Experimental equipment.
Next, the base metal is placed in the wooden support and held in place by clamps to thus avoid
deformation of the plate. All this is then finally placed on the milling machine table and bolted
into place.
Following that, the TIG torch is positioned by means of the corresponding tool to the head of
the milling machine. Then the mass clamp is placed on the metal to be welded, the gas volume
and the intensity of the welding current are regulated, the velocity of movement of the milling
machine table is programmed, and finally the thermocouples are connected to the data logger,
which is in turn connected to the PC.
Once the test set is prepared, the arc is charged, the feed velocity is switched on (milling
machine table) and the test commences. While the weld bead is made the temperatures are
recorded. Finally, the data obtained are exported to a spreadsheet for later analysis.
6. Results and Discussion
6.1. Preliminary Tests
To fine-tune the test method, initially the first experiments were carried out with eight
thermocouples encased in a sheath of stainless steel. With this system it was detected that the
98
thermocouples furthest from the weld bead scarcely varied in temperature and that, moreover,
the highest temperature registered by said thermocouples did not surpass 500º C, as can be
seen in Figure 5.
Figure 5. Result of the test with 8 thermocouples (preliminary tests). Registers
temperature-time.
In addition, in order to be able to move the plate towards the encased thermocouples, a tool
was designed made in aluminium (figure 6). This tool was excessively rigid and did not allow for
compensating the deformations which are produced in the plate during the welding process.
Thus, contact between the thermocouples and the plate was lost, and non-real temperate
readings were obtained.
Loss of contact
thermocoupleplate
Thermoco
uple
holder
Figure 6: Tool used to hold the encased thermocouples (preliminary tests).
99
6.2. Final tests
Table 6 indicates the welding variables considered in the two tests carried out with the final
procedure.
Test I
Test II
Description
I
79
120
A
Intensity of the current
V
15
20
V
Voltage
η
0.7
0.7
EBA
3.485
5.263
kJ/mm
Gross Energy Contributed
e
2
4
mm
Plate thickness
v
0.0034
0.0034
m/s
Lineal weld velocity
Yield of heat contributed
Table 6. Experimental welding variables.
In figure 7 the temperature values registered according to time at 5, 10 and 15mm distance
from the axle of the bead and the weld are represented. A cooling time of 200 s is considered,
which corresponds to the time in which the temperatures at the different points become the
same.
Temparatura
( ºC)
Temparatura
( ºC)
Resultados Experimentales
Resultados Experimentales
1200
1200
1100
1100
1000
1000
900
900
800
800
700
700
600
600
500
500
400
400
300
300
200
200
100
100
0
0 0
0
Exp. 5 mm
Exp. 5 mm
Exp. 10 mm
Exp. 10 mm
Exp. 15 mm
Exp. 15 mm
50
50
100
100
150
150
200
250
300
200
250
300
Tiempo ( s)
Tiempo ( s)
350
350
400
400
450
450
500
500
Figure 7. Experimental results test 1.
100
One of the important parameters to determine in welded joints of stainless steels is the time
taken between the temperatures of 800 and 500ºC during cooling, that is to say, the well-known
t 8/5. The determination of this parameter is shown in figure 8.
Temperatura (ºC)
Tiempo Enfriamiento 800-500ºC
1100
1000
900
800
700
600
500
400
300
200
100
0
t8/5
1
51
101
151
201
251
301
351
401
Tiempo ( S)
Exp. 5 mm
Exp. 10 mm
Exp. 15 mm
800 ºC
500º C
Figure 8. Determination of the parameter t 8/5.
The tests allow to also determine the variation of the temperature with the distance from the
weld bead, as is indicated in figure 9.
Tmax Exp-Distancia cordon
Temperatura (º C)
T max
1100
1000
900
800
700
600
500
400
300
200
100
0
5
10
15
Distancia ( mm)
Figure 9: Variation of the values of maximum temperature with the distance to the weld bead.
In figure 10 the experimental results from Test II can be seen, according to the parameters from
table 6. In this test, the thickness of the plate and the intensity of the current have been
modified. It can be observed that, although the intensity employed has been greater than in the
previous test, the maximum temperature reached is considerably lower. This demonstrates the
sensitivity of the phenomenon to the plate thickness, which, being greater in this case,
101
generates greater thermal dissipation acting as a thermal cooler. In this last case one more
thermocouple has been considered, so that a greater distance to the main bead has been
analyzed (fourth thermocouple).
Resultados Ensayos
1000
900
Temperatura
800
700
600
500
400
300
200
100
0
0
50
100
150
200
250
300
350
400
450
500
Tiempo
Termopar 1
Termopar 2
Termopar 3
Termopar 4
Figure 10. Registers for temperature time for an intensity of current of 120 And a plate thickness of 4 mm.
7. Conclusions
An experimental procedure has been designed to calculate the distribution of temperatures in
the zone adjacent to the weld bead of a structural stainless steel.
It has been seen that it is more adequate to measure the temperatures in the zone of the weld
bead by means of type K thermocouple wire, since thermocouples encased in a sheath of
stainless steel do not provide good results due to the elevated thermal inertia of the cover.
During the welding time an important increase occurs in the temperature in the zone of the joint,
this produces a deformation of the base material. With this experimental analysis the
temperature in the HAZ can be calculated in order to reduce the deformation of the base
material.
By means of this analysis of the temperature field, the dimension of the HAZ can be calculated.
The data obtained with these tests can serve to validate the calculation of the HAZ with a
numerical method.
This experimental procedure can be used for other types of steel and for other welding
procedures by electric arc such as MIG/MAG.
References
Alhama F., Alcaraz D. and González C.F., 2005. Rev. Metal. Madrid Vol. Extr. pp.304-308.
102
Benhayon M, 1994. Introducción al Proyecto de las Construcciones Metálicas Soldadas. II
Congreso Colombiano de Soldadura y III Exposición Industrial (Colombia), Centro Argentino de
Tecnología de la Soldadura. CATS
Gómez de Salazar J.M., Moreno P., 2003. “Soldabilidad de aceros inoxidables bajos en Ni”
Congreso de Ciencia e Ingeniería de Materiales. Lugo (Spain)
González R., Suárez J.C., Vicario F.J. and Muñoz A., 2006. Rev. Metal. Vol.42 pp.255-269.
Mazur Z., Gonzalez G. , Urquiza G., Salazar O., Mariño C. and Hernandez A., 2002. Rev.
Metal. Vol. 38 pp.14-22.
UNE-EN 10088-1:2006. Aceros Inoxidables. Parte 1: Relación de Aceros Inoxidables
UNE-EN 10028-7:2007. Aceros Inoxidables Parte 7: Productos planos de acero para
aplicaciones a presión.
Acknowledgements
The authors would like to thank the Laboratorio de Ciencia e Ingeniería de Materiales del
Instituto de Desarrollo Regional de Albacete of the UCLM, for the support received in carrying
out this work.
Eusebio José Martínez Conesa.
Profesor Ayudante de Universidad.
Departamento de Arquitectura y Tecnología de la Edificación.
Área: Construcciones Arquitectónicas.
Escuela Técnica de Ingeniería Civil.
Universidad Politécnica de Cartagena.
Pº de Alfonso XIII, 52, Edificio de Minas Despacho nº 15
Cartagena, Spain
Phone: (+34) 968325666
Fax: (+34) 968325931
103
VALIDATION OF A MODEL OF TYPIFICATION OF AGRARIAN
EXPLOITATIONS BY TECHNICALS OF DISCRIMINANT ANALYSIS
Riveiro, J. A.
Marey, M.F.
Álvarez, C.J.
University of Santiago of Compostela
Abstract
Galicia (no Spain) is an example of agriculture characterized by the existence of a variety land
uses, different productive models, dimensional diversity and geographical dispersion of farms,
in a highly heterogeneous territory in terms of constraints for farming development. Any action
aimed to improve production efficiency in this activity, must fulfill the characterization of the
production models, and needs of a methodology for data analysis, and the obtention of
technical-economical and financial results based on models developed from representative
farms in the various areas. For this reason, a methodology of grouping farms in different types
was developed, based on variables from the microdata Agricultural Census, and using a
combinatory analysis technique subject to constraints. In this paper, is described the validation
process of this methodology by applying Discriminant Analysis techniques on population
census of commercial Dairy farms in Galicia, using characterizing variables of production
systems.
Keywords: Productive Agricultural Management, Rural Development, Discriminant Analysis,
Types of farms, modelling of farm production systems
1. Introduction
Agriculture Galician (northwest Spain), is characterized by the existence of many land uses and
a great diversity in the productive structure of land exploitations. In the middle-twentieth century
in Galicia, the agrarian activity was based on smallholder and a destined production almost
entirely to subsistence, with a great structural complexity and with more than 70% of the
population employed in this sector, in more of 433.000 exploitations (Bouhier, 2001). In 1999,
there were 270.053 exploitations in 696.691 ha of useful land surface (2.58 ha. per such
exploitation) destined for the largest of cases to the production for own consumption, with very
different dimensions, from those with less than 1 ha, compared with those that made it possible
the marketing of production (I.N.E., 1999). This regional situation has endured a largely
structural transformation in the last years. At present, EEEA (Survey about the structure of the
agrarian exploitations) in 2005, (INE-EEEA, 2005) shows that the number of exploitations are
reduced to 91,813 while the cultivated area (UAA) is maintained in values similar (about
700,000 ha.). On the other hand, it is retained a degree of structural complexity and usually, in
104
each exploitation there is more than one specific productive guidance, which integrates various
land uses, and is sometimes confused the own consumption with the commercial production.
In a highly dynamic setting in terms of transformations is referred, in 2002, the Administration of
Galicia promoted the realization of studies of Productive Land Management, which took place
on the 21 of the 53 regions of Galicia, (UTE EIDO-USC, 2004). To meet the needs arising in
the development of works, emerged different patterns of help, among them stressed Gestop ®,
a model of analysis of the different agroforest existing and potential uses in Galicia, (Riveiro et
al., 2005). This model was used to generate technical and economic results that could be
integrated with other indicators (structural, social and environmental) allow developing a
support system in decision-making to prioritize the activities to promote in every territory in the
region.
The application of Gestop ® required information from the current situation, so that the provided
results were real useful. It was necessary to develop other systems and models to provide
information, specifically in regard to the typification, classification, territorial analysis and
characterization of land exploitations, as tools of support in processes of Productive Land
Management.
With a view to improving the initial tool, a combinatorial method was developed. Such a
combinatorial method used official census data to classify farms based on the different
production processes, sizes and geographical locations of farms, among other factors. The
combinatorial method proved very useful as a source of differentiated data for the decision
support system. Yet, validating the results using robust statistical methods was considered
essential in order to guarantee the suitability of the model for use.
In the international literature there are many different proposals of models and developed
systems as an aid in planning for agrarian activity. Le Show & Benoit, (Le Ber & Benoit, 1998),
devised a model based on agronomic quality rules, which allows describing the spatial
organization of land uses, creating maps of uses. Hardin et al. (Hardinan et al., 1990), use the
Cluster analysis for the identification and classification of systems of exploitation in a region of
China. Such analysis allowed for the identification of three representative groups (slope was the
principal component for discerning the groups) and seven subgroups. This was useful for
determining the geographic boundaries of different systems of exploitation, and to choose the
appropriate areas for the introduction of new technologies and also to make taxonomy of
systems of exploitations. Because a large number of production systems would be obtained if
we assumed that each farm is unique in its characteristics, such an approach is not realistic or
efficient. Köbrich et al. (Köbrich et al., 2003) raise a system of taxonomy from typical or
representative situations, using statistical techniques of multivariate analysis, selecting the
factors to consider in the analysis Cluster. Such techniques were implemented with data of
farms in Chile and Pakistan. In both cases, the quantitative and qualitative information was rich,
such that some variables had to be excluded before performing a principal component analysis
of the remaining variables and obtaining the factors that should be considered in the cluster
analysis. The main discussion issue focused on where the tree (dendrogram) could be cut in
order to obtain the optimal number of groups using heuristic procedures. Kostov and McErlean
(2006) tried to find the appropriate number of clusters by using the mixture of distributions
model (MDM) as a clustering technique that proved effective in classifying dairy farms in
Northern Ireland. Another perspective of classification, which is provided by Girard et al. (Girard
et al., 2001) propose a model which allows cataloging the systems of exploitations based on
the practices of land use, accepting the temporal and spatial analysis. The authors identified
the attributes that characterized land use practices and defined the main models or farmland
105
use categories by combining the identified attributes based on different farm models. The
model was applied to a set of sheep farms in the South of France. The results obtained from
this categorization approach allowed for the analysis of farms in terms of time and space.
Some researchers considered the size of the exploitation as a suitable variable for typological
classification and the analysis of different aspects. Tavernier and Ptolemy (Tavernier &
Tolomeo, 2004) analyzed exploitations according dimensional ranges, finding a significant
positive correlation between the small farms and the practices of sustainable agriculture.
Demircan et al. (Demircan et al., 2006) conducted an analysis of the economic structure of
cattle farms in Turkey, based on their dimensional classification. This allowed them to detect
significant differences in technical and economic obtained results, in relation to the size of the
farm. Judez et al. (Judez et al., 2001) also based on a dimensional classification of the
exploitations from the number of UDEs, to analyze the effects of the application of the called
"Diary 2000" on beef cattle farms.
2. Background
In a process of Agrarian Productive Management requires differentiated identification and
evaluation of different productive systems. In the described work is classified and established
groups of exploitations with some specific specialty or are studied systems or rules that allow
such classification, however, none of them gives the appropriate results to the established
objectives in the context of an Agricultural Productive Management. That is why was developed
an original procedure for the clustering of exploitations in disjoint classes, taking into account
the differences between agrarain exploitations of a region (typification by combinatory analysis),
(Riveiro et al., 2008) and its temporary development, (Álvarez et al., 2008) that is based in the
hypotheses that this differentiation will depend on the number and type of uses belongs to
exploitation (they define their productive orientation, degree of complexity and diversity), and its
productive dimension (usually associated with different degrees of mechanization,
specialization and productive efficiency), which will lead to different productive systems. All
these results can be used by the Administration in the decision support system for Agricultural
Production Planning.
The typification methodology of exploitations based on Combinatorial Analysis has advantages
over other techniques of statistical analysis. In this sense, Iraizoz et al. (Iraizoz et al., 2007)
found problems in the formation of groups by Cluster techniques, from the multicollinearity
among variables, which is a common feature of agrarian systems, regarding to the variables
that characterize the exploitations. This forces frequently to carry out an analysis of major
components to eliminate some of these variables. The technique that is objective to validate, is
based on combinatorial analysis, and allows the selection of the variables to use accordance
with the objectives of clustering, and to establish the rules of the game through the restrictions
to the process of generation of groups, from the available information for each exploitation.
The combinatorial methodology uses the principal and most extensive source of information
regarding the volume of exploitations in question, that are the Agricultural Census, and the last
done in Spain was in 1999, (I.N.E., 1999), which are available the microdata. As there is a
extensive information for each individual exploitation, coupled with the consequent possibility of
incorporating in a rational way, other information with a degree of equal or major aggregation,
make of the Agricultural Census the suited source to deal different studies or research and at
different scales. Censuses are compiled for every EU (European Union) country in order to
106
have official statistics at the regional and other levels. Given the relevance of such censuses,
some non-member countries have been compiling agricultural censuses. For example, China
compiled the first National Agricultural Census in 1997. The census revealed substantial
differences among farm types and allowed for the redefinition of the geography of Chinese
agriculture (Fanfani and Brasili, 2003).
In the census, (I.N.E., 1999), each exploitation is characterized by the following: identification of
the exploitation, ownership, legal personality and management, total surface and parcelization,
tenure of land, irrigation, land use, crop, machinery, livestock, installations for the storage of
fertilizers, labor and marketing of production .Considering the Agricultural Census as a
reference source for researchers, with the volume of information that it provides and for its
international character, it was opted for the use of the microdata, specifically those made in
1999, (I.N.E., 1999), as main source of data for development of typification procedures of
agrarian exploitations.
2.1. Typification by Combinatorial process
Based on a mathematical combinatorial process subjected to different restrictions, and is
performed by an application that is developed in Excel which automatically generates various
possible combinations of use, making the composition and number of exploitations in which is
present in each of them. This process is carried out in successive stages, with the possibility of
obtaining different intermediate or final results, as described by Riveiro et al. (Riveiro et al.,
2008). Once completed, will generate a specific number of combinations of land uses, whose
analysis allows establishing the types of representative exploitations, of which are known the
integrated land uses and the number of exploitations that are present. These types collect from
simple combinations of two uses until complex combinations of several land uses with a
maximum of 11 (the main and 10 secondary).
2.2. Results of the Combinatorial process with dairy farms
The Agricultural Census of 1999 collects for the area of study (Galicia, NO Spain), information
of a total of 29.180 dairy farms that sell milk. Of this population has been excluded all
exploitations with fewer than 10 productive units, arguing that proposals do not address to
productive commercial approaches. Thus, the analyzed population was reduced to 16.445
exploitations.
Applying the referenced methodology allowed to differentiate the galician exploitations with the
presence of dairy cattle, in three representative types: (T1) exploitations of cattle with a dairy
production linked to the land and based in the forage crops grasses, with the presence of the
maize crop, whatever of that there are other additional activities such as beef rearing, potatoes
crop or forest uses, among the most common. In these exploitations will not exist cattle of beef
in a greater number than 4 animals (T2) identical to the T1 but without the presence of the
maize crop, and (T3) consisting of exploitations that differ from previous (T1 and T2) by the
presence of cattle of beef in a greater number than 4 animals.
Moreover, regarding the distribution of exploitations as defined types and taken dimensions,
serves the expressed values in Table 1. Exploitations of the types T1 and T2 together represent
91% of all concerned. The evolution of the sector since 1999, corresponding to the census
data, has led to a significant decrease in the mixed exploitations T3, (INE-EEEA., 2005) tending
to his disappearance, which minimizes the importance of this type of exploitations.
107
The objective of this article is to validate this procedure, using a statistical methodology of
Discriminant Analysis (DA) that allows contrasting the existence of differences between the
identified groups by Combinatorial Processes subjected to restrictions. The D.A. is a technical
of Multivariant classification very used by researchers for the classification of products and
agrarian productions with different objectives. Davidson et al. (Davidson et al, 2007) used it for
comparison of the ecological status of different agroforest models, Liu et al. (Liu et al., 2006)
used it in combination with spectroscopy techniques, for the classification of wines from
different origins belongs to the same variety; Cozzolino et al. (Cozzolino et al., 2008) to analyze
the reliability and accuracy of a spectroscopy technique as a tool to verify and authenticate the
type of used silage in feed for ruminants; Bowler et al. (Bowler et al., 1996) to explore different
models of farms in the North of England. The D.A. was also used to study the different
responses to diseases regarding with different types of crop models, (Van Schik, 1998) or
compare different models of management of exploitations and their economic performance,
(Arriaza & Gómez-Limón, 2003). In addition, Hodge, (Hodge, 1978), used these techniques to
study the environmental impacts posed by different types of farms, and Somda et al, (Somda et
al, 2005) used it to study the viability of farms in terms of use of family labor and its economic
structure.
3. Methodology of clustering by D.A.
The described methodology, based on Discriminant Analysis will be used for the purpose of
validating the methodology of typification through a Combinatorial process. Both methodologies
of typification (Combinatorial process and D.A.), are made from the selection of the
exploitations with an use in common.
The discriminant analysis starts in a data table of n individuals who have been measured p
independent variables, which act as a profile of characteristics of each of them. An additional
qualitative variable, dependent or qualifying, with two or more categories, defines the group to
which the individual belongs. It is thus a table of n • (p +1) cases in which each figures with a
profile and an allocation to group. From the table of n • (p +1) it will be obtained a mathematical
discriminant model against which will be contrasted the profile of the new individual whose
group is unknown to, according to a numerical result, be assigned to the most likely group.
The discriminant function of Fisher (D), (equation 1), it is obtained as linear function of k
explanatory variables. The discriminant scores are the obtained results by giving values to X1,
X2,..., Xk in that equation. The mathematical problem is to obtain the coefficients of weighting uj
that best allow distinguishing the individuals of the different groups.
D = u1 X 1 + u 2 X 2 + ... + u k X k
(1)
The discriminant cutoff point (C) (Equation 2) is used to classify each individual in its
corresponding group. The considered criteria for the selection of the variables are Wilks
Lambda (λ ) (equation 3) and the F statistical. The results are interpreted according to the
following criteria: if the value of λ is close to unity, the groups will be mixed and the set of
independent variables will not be adequate to build the discriminant function, because the
higher rate of variability would be represented by the variability within the groups.
D = u1 X 1 + u 2 X 2 + ... + u k X k − C
(2)
108
λ=
Within group sum of square
Total sum of square
(3)
The approach of Wilks Lambda identifies the best discriminant variables but the F statistic will
determine those variables that should be to consider in the model. For this, if F> 3,84 the
variable will be considered, otherwise will be eliminated. The value of the F statistic is
determined by equation (4), where n is the number of exploitations; g, the number of groups; p
the number of independent considered variables; λp the calculated λ before including the
selected variable and λp+1 the calculated λ after including the selected variable.
⎛ n − g − p ⎞ ⎛⎜ λ p +1
⎟⎟ ⋅ 1 −
F = ⎜⎜
λp
⎝ g − 1 ⎠ ⎜⎝
λ p +1 ⎞
⎟
λ p ⎟⎠
(4)
For a population n with p independent variables, each exploitation can be considered as a point
on a p-dimensional space. The coordinates of each point will be obtained from p independent
variables.
Based on the representation of n exploitations in an area Rp, it is claimed to obtain a new
space of minor dimension (Þ) such that, by projecting the point cloud on this space, the
corresponding points to exploitations including in different groups, are removed. In the studied
case there are only two groups (T1 and T2), which will be obtained a discriminant function
(equation 5), in which, the value Di corresponds to the discriminant score for each exploitation,
Xp the value of variable p in the exploitation i, and â some estimated coefficients so that the
obtained values for Di carry out the following conditions: a) maximizes the obtained auto values
by the equation 6, b) minimize the calculated Wilks Lambda (λ)by equation 3, and c) maximize
the canonical correlation according to the equation 7.
Di = â0 + â1 X 1 + â2 X 2 + ... + â p X p
Autovalore s =
n=
Between group sum of square
Within group sum of square
Between group sum of square
Total sum of square
(5)
(6)
(7)
Once obtained the discriminant function (equation 5), that can be used to make a classification
of the same used cases to obtain the function, allowing checking the effectiveness of the
discriminant function to make the classification. If the results are successful, the discriminant
function can be used to classify future cases (exploitations) of that, knowing the values of the
independent variables, are not known to group which they belong.
When groups of different sizes are managed, it is necessary to compensate the inequality to
classify the cases. If is assumed that discriminant scores attend a normal distribution, it can be
calculated the associated probability with an event in each of the used groups in the analysis
(conditional probability), equation 8. To classify each new case where is not known to which
group belongs, is necessary to compare the conditional probabilities that correspond in each
group of analysis, using the Bayes theorem of the equation 9.
109
P = (D > di G = g k )
P( g k d i ) =
(8)
P( d i g k )P(g k )
(9)
g
∑ P( d
k =1
i
g k )P(g k )
4. Results of validation of typologies by D.A.
The validation will be performed using the same data source, the agrarian census. The
information that this provides allows selecting some direct variables or to establish calculated
variables (as relations between direct variables), which may serve to establish groups of
exploitations with significant differences in their productive systems.
Of the three defined types from the combinatorial methodology, the typology T3 does not need
validation about the differences in their productive system according to the other two types,
because the systems of management of the beef cattle are different than dairy cattle, and so,
its exclusive presence in these exploitations will tick significant differences. Moreover, as this
typology is also represented by a small number of exploitations, even tend to disappear, it is
chosen by their exclusion and focus the discriminant analysis in the exploitations of the
typologies T1 and T2. These represent a whole (Table 1), a total of 15.004 units. Of this
amount a total of 36 exploitations are invalid for errors in the values of some of the variables.
From 14.798 valid exploitations, are established two groups: one with 7.456 exploitations on
which the discriminant analysis forms a random sample of approximately 50% of the total the
valid exploitations. The 7.512 remaining exploitations are reserved to test the quality of the
resulting model.
Size dairy class
nº V. L.
Type 1
Number
Type 2
%
Number
Type 3
%
Number
Total
%
Number
%
10-24
3.906
64,1%
6.507
73,0%
1.173
81,4%
11.586
70,5%
25-39
1.462
24,0%
1.723
19,3%
200
13,9%
3.385
20,6%
40-54
462
7,6%
497
5,6%
43
3,0%
1.002
6,1%
55-69
131
2,1%
108
1,2%
7
0,5%
246
1,5%
> 69
133
2,2%
75
0,8%
18
1,2%
226
1,4%
Todas
6.094 100,0%
8.910 100,0%
1.441 100,0%
16.445 100,0%
Table 1. Results of the typification by combinatorial analysis.
.
110
T1
(3.041 valid)
Variables
T2
(4.415 valid)
T3
(7.456 valid)
Average Stand. Dev. Average Stand. Dev. Average Stand. Dev.
ST
18,631
13,938
16,416
11,473
17,320
12,583
SAU
13,598
11,123
11,869
8,859
12,574
9,881
UGM_Bov/UTAT
15,056
19,946
16,795
144,411
16,086
111,851
2,562
1,222
3,926
45,272
3,370
34,851
11,813
10,708
18,707
176,607
15,895
136,108
%_SF_Meadow
0,433
0,056
0,498
0,009
0,472
0,048
%_Corn
0,133
0,112
0,004
0,017
0,057
0,096
%_Rearing
0,238
0,203
0,202
0,202
0,216
0,203
27,383
25,215
27,032
23,357
27,175
24,131
1,743
1,494
2,033
1,879
1,914
1,738
45,497
11,127
47,094
11,227
46,442
11,213
S_unitary_plot
1,263
3,044
1,139
2,476
1,189
2,722
Sf_Tot/S_Tot
0,710
0,196
0,681
0,223
0,693
0,213
UGM_Bov/SF_Tot
CV_Trac_Pr/SF_Tot
NºPlots
NºPlot/ha_ST
Responsible Age
Tabla 2. Stadistics of group.
Initially were selected the 13 following variables: ST, (Total area of exploitation), UAA (Utilized
agricultural area); UGM_Bov / UTAT (Major Cattle Units per Unit of Agricultural Labor);
UGM_Bov / SF_Tot (Major Cattle Units per hectare of Forage Total Area); CV_Trac_Pr /
SF_Tot (Horsepower in Own Tractors per hectare of Forage Area Total); _SF_Meadow%
(Percentage of forage area occupied with Meadows); _Corn% (Percent of the Forage Area
busy with Forage Corn); _Rearing% (percentage of animals in period of Rearing); Nº Plots
(Number of Plots in the Exploitation); Nº plot/ ha_ST, (Number of Plots per hectare of Total
Area of the Exploitation); Responsible_age (Age of Responsible of the exploitation);
S_unitary_plot (Unitary Surface of agricultural Plots) and Sf_Tot / S_Tot (Ratio between of
Forage Area of the exploitation and its Total Area).
The Table 2 shows the distribution and the statistics of group (T1 and T2) for the 13 variables.
The use of F statistic of the equation 4, for a value of n =7,456 exploitations and g=2 groups,
allowed selecting as variables to include in the analysis, the 8 considered statistically significant
(F> 3,84), reflected in Table 3, which will allow to establish the ownership of the exploitations to
one or another analyzed group (T1 or T2). The other variables of the 13 originally considered
have been excluded because the value that takes the F statistic is minor than the output value
(F <2,71).
111
Variable
Tolerance
F exclusion
(λ)
%_Corn
0,991
5.735,804
0,978
SAU
0,111
12,925
0,553
Sf_Tot/S_Tot
0,435
66,984
0,558
ST
0,103
19,650
0,554
S_unitary plot
0,778
2,766
0,553
NºPlot/ha_ST
0,398
33,761
0,555
NºPlots
0,372
26,330
0,554
Responsible Age
0,974
5,640
0,553
Table 3. Inclusion of variables in the analysis.
The discriminant function indicated in equation 5 is applied for a value i =7,456 exploitations
and a value p=7 independent variables. In Table 4 presents the corresponding results for the
test of significance of the discriminant obtained function (equation 5).The values of lambda
0,500 are low, evidence of differentiation between groups, which confirmed high levels of
statistic, and above the value of significance less than 0,05%.
F exact
Step
(λ)
nºp
nºF
nº Ex.
1
0,568
1
1
7.454
5.673,065 7.454
0,000%
2
0,563
2
1
7.454
2.893,854 7.453
0,000%
3
0,560
3
1
7.454
1.949,483 7.452
0,000%
4
0,556
4
1
7.454
1.485,823 7.451
0,000%
5
0,556
5
1
7.454
1.191,413 7.450
0,000%
6
0,555
6
1
7.454
995,553 7.449
0,000%
7
0,553
7
1
7.454
860,213 7.448
0,000%
8
0,553
8
1
7.454
753,861 7.447
0,000%
Stadistic G.L. Significance
nºp: number of variables in the discriminate function
nºF: number of discrimínate functions
nºEx.: number of considerated exploitations
G.L.: Degrees of freedom
Table 4. Test of significance of the discriminant function.
112
In Table 5 shows the results of the all comparisons by pairs of groups, for the F statistic and its
significance. Due that the values of F are high and the level of significance inferior than 0,05%,
can be indicated that all the included variables in the discriminant analysis differentiate
significantly the typologies, two to two.
Group
Step Group
T1
F
1
2
3
4
5
6
7
8
Signif.
T1
T2
2.893,854 0,000%
2.893,854 0,000%
1.949,483 0,000%
1.949,483 0,000%
T1
T2
1.485,823 0,000%
1.485,823 0,000%
T1
T2
1.191,413 0,000%
1.191,413 0,000%
T1
T2
995,553 0,000%
995,553 0,000%
T1
T2
860,213 0,000%
860,213 0,000%
T1
T2
Signif.
5.673,065 0,000%
T1
T2
F
5.673,065 0,000%
T1
T2
G.L.
T2
753,861 0,000%
753,861 0,000%
7.454
7.453
7.452
7.451
7.450
7449
7.448
7.447
F: stadistic
Sig.: significance
G.L.: Degrees of freedom
Table 4. Comparisions of groups by pairs.
The obtained autovalor for the discriminant function (equation 6) is 0,810, and the obtained
value for the canonical correlation (equation 7) is 0,669. Both values are close to the unit, which
is a clear indication that the discriminant obtained function allows distinguishing the average of
the groups T1 and T2, indicating that the points relating to the exploitations of both groups have
a high degree of differentiation according to the discriminant scores (Di).
113
Coeff. Variable
Group
T1
T2
1,047
1,012
a1
ST
a2
SAU
-1,122
-1,088
a3
%_Maíz
31,385
6,410
a4
NºParcelas
-0,012
-0,023
a5
NºParcel/ha_ST
1,729
1,897
a6
Edad_Responsable
0,423
0,430
a7
S_unitaria_parcel
-0,035
-0,013
a8
Sf_Tot/S_Tot
41,902
40,057
a0
(Constante)
-30,938 -27,760
Table 6. Value of the coefficients of the discriminant function.
The obtained value for the Wilks Lambda (λ) by equation 3, is 0,553, which is contrasted of
hierarchical way the significance of the obtained function by the null hypothesis, getting a
significance of 0.00 (<0,05), which suggests that the model discriminates well between the two
groups.
The centroids put the typology T1 in the positive part with a value of the function equal to 1,084,
and to the typology T2 in the negative part with a value -0,747.
The Table 6 shows the values of the estimated coefficients for equation 5 of the discriminant
function. These will allow calculating the values that will determine which typology is included
such exploitation of the tested population. The Table 7 shows the obtained results of ranking for
both cases, selected and new or not selected. In selected cases, were able to classify correctly
84,7% of the exploitations (Mardia et al., 1994). The percentage of correct classification of the
used exploitations in the validation sample (new cases) reaches values of 62,8% for the type
T1, and 99,7% for type T2, representing a 84,8% of exploitations. In set it is achieved a good
result of classification, which stands at 84,6% of exploitations whose classification matches as
both methodologies, discriminant analysis and combinational process.
114
Group
Recount
Selected cases
Percentage
Recount
Not selected cases
Percentage
Forecast group
Total
T1
T2
T1
1.918
1.123
3.041
T2
20
4.413
4.433
T1
63,07% 36,93% 100,00%
T2
0,45% 99,55% 100,00%
T1
1.909
1.133
3.042
T2
15
4.473
4.488
T1
62,75% 37,25% 100,00%
T2
0,33% 99,67% 100,00%
Table 7.Results of the clasification.
These results are similar or even better than those obtained by other researchers using
techniques of discriminant analysis, for example Liu et al., (2006), achieved an overall rate of
correct classification of wines from 76,6%. Stewart et al., (Stewart et al., 1979) regarding the
prediction of land use and depending on the number of considered groups and selected
variables, classifies correctly between 51,65 and 62,7% of cases.
5. Conclusions
The developed methodology for to group exploitations according to different types, based on a
combinatorial algorithm, is validated when after the completion of the Discriminant Analysis, the
results show a high coincidence in the classification.
As shown in this paper, the development of new heuristic classification or clustering methods
aimed at conducting production planning initiatives suited to each of the resulting farm groups
demands validation actions that verify the quality of results.
This is of great importance because the validated methodology of grouping, based on a
combinatorial algorithm, allows establishing different groups of exploitations from a public
source of information such as the Agrarian Census, covering all agrarian exploitations in the
territory. Moreover, given that the methodology generates the cases and then the selection of
groups is made through a manual process of analysis and valuation, the resulting types are
more realistic than those obtained by other statistical methodologies subjected to the
restrictions of the number values of different variables and to the subjectivity of the imposed
weights by unrelated criteria with the differing characteristics of the productive processes that
govern the exploitations.
115
References
Álvarez C.J., Riveiro J.A., Marey M.F., 2008. Type, classification and characterization of farms
for agricultural production planning. Spanish Journal of Agricultural Research. Vol. 6 (1), pp.
125-136.
Arriaza M., Gómez-Limón J.A., 2003. Comparative performance of selected mathematical
programming models. Agricultural Systems. Vol. 77 (2), pp. 155-171.
Bouhier A. 2001. Galicia Geographical test of Analysis and Interpretation of an old Agricultural
complex Tomos I y II, (Translation of Benxamin Casal Vila), Xunta of Galicia - Counseling of
Agriculture Livestock and Forestry, Caixanova. Santiago of Compostela.
Bowler I., Gordon C., Alasdair C., Ilbery B., Alastair S., 1996. The development of alternative
farm enterprises: A study of family labour farms in the Northern Pennines of England, Journal of
Rural Studies. Vol. 12 (3), pp. 285-295.
Cozzolino D., Fassio A., Restaino E., Fernández E., La Manna A., 2008. Verification of silage
type using near-infrared spectroscopy combined with multivariate analysis, Journal of
agricultural and food chemistry. Vol. 56 (1), pp. 79-83.
Davidson N.J., Close D.C., Battaglia M., Churchill K., Ottenschlaeger M., Watson T., Bruce J.,
2007. Eucalypt health and agricultural land management within bushland in the Midlands of
Tasmania, Australia, Biological Conservation. Vol. 139 (3-4), pp. 439-446.
Demircan V., Binici T., Koknaroglu H., Aktas A.R., 2006. Economic analysis of different dairy
farm sizes in Burdur province in Turkey. Czech Journal of Animal Science. Vol. 51 (1), pp. 8-17.
Fanfani, R., Brasili, C., 2003. Regional differences in Chinese agriculture: Results from the
1997 first national agricultural census. J. Peasant Stud., 30(3-4), 18.
Girard N., Bellon S., Hubert B., Lardon S., Moulin C.H., Osty P.L., 2001. Categorising
combinations of farmers' land use practices: an approach based on examples of sheep farms in
the south of France, Agronomie. Vol. 21 (5), pp. 435-459.
Hardinan R.T., Lacey R., Yang M.Y. 1990. Use of clustes-analysis for identification and
clasification of farming systems in Qingyang county, central North China, Agricultural Systems.
Vol. 33 (2), pp. 115-125.
Hodge I.,1978. Application of discriminant-analysis for evaluation of local enviromental-impact
of livestock production, Agriculture and Environment. Vol. 4 (2), pp. 111-121.
INE. Basis of Agricultural Census of microdata 1999, [Online], In: Internet <http//www.ine.es>
(Consultation, 28 october 2005).
INE-EEEA. Survay about the estructure of the Agricultural Exploitations 2005, [Online], In:
Internet <http//www.ine.es> (Consultation, 10 January 2007.
INE-Met. Methodology of the Agricultural Census [Online], In: Internet <http//www.ine.es>
(Consultation, 28 October 2005).
Iraizoz B., Gorton M., Davidova S., 2007. Segmenting farms for analysing agricultural
trajectories: A case study of the Navarra region in Spain. Agricultural Systems. Vol. 93 (1-3),
pp. 143–169.
Judez L., Chaya C., Martinez S., Gonzalez A.A., 2001. Effects of the measures envisaged in
"Agenda 2000" on arable crop producers and beef and veal producers: an application of
116
Positive Mathematical Programming to representative farms of a Spanish region. Agricultural
Systems. Vol. 67 (2), pp. 121-138.
Köbrich C., Rehman T., Khan M., 2003. Typification of farming systems for constructing
representative farm models: two illustrations of the application of multi-variate analyses in Chile
and Pakistan. Agricultural Systems. Vol. 76 (1), pp. 141-157.
Kostov, P., Mcerlean, S., 2006. Using the mixtures-of-distributions technique for the
classification of farms into representative farms. Agricultural Systems, 88(2-3), 528-537.
Le Ber F., Benoit M., 1998. Modelling the spatial organization of land use in a farming territory.
Example of a village in the Plateau Lorrain. Agronomie. Vol. 18 (2), 1998, pp. 103-115.
Liu L., Cozzolino D., Cynkar W.U., Gishen M., Colby C.B., 2006. Geographic classification of
Spanish and Australian tempranillo red wines by visible and near-infrared spectroscopy
combined with multivariate analysis, Journal of agricultural and food chemistry. Vol. 54 (18), pp.
6754-6759.
Riveiro J.A., Álvarez C.J., Marey M.F., Marco J.L., Alvarez C.J., 2008. Procedure for the
classification and characterization of farms for agricultural production planning: Application in
the Northwest of Spain, Computers and Electronics in Agriculture. Vol. 6 (1), pp. 169–178.
Riveiro J.A., Álvarez C.J., Miranda D., Pereira J.M., 2005. “Profitability and production
requirements for land use allocation of farming and forestry land” Biosystems Engineering. Vol.
90 (4), pp. 477-484.
Somda J., Kamuanga M., Tollens E., 2005 Characteristics and economic viability of milk
production in the smallholder farming systems in the Gambia, Agricultural Systems. Vol. 85 (1),
pp. 42-58.
Stewart A., Reeds L.G., 1979. An Application of Discriminant Analysis to Agricultural Land Use
Prediction, Economic Geography, Vol. 55 (2), pp. 114-122.
Tavernier E.A., Tolomeo V., 2004. Farm type and sustainable agriculture: Does size matter? ,
Journal of sustainable agriculture. Vol. 24 (2), pp. 33-46.
UTE EIDO-USC., 2004. Regional Studies of Productive Agricultural Management of 21 regions,
Xunta of Galicia. Counseling of Agricultural Policy and Rural Development. Santiago of
Compostela.
Van Schik C., Dijkhuizen A.A., Benedictus G., Barkema H.W., Koole J.L., 1998. Exploratory
study on the economic value of a closed farming system on Dutch dairy farms, The Veterinary
Record. Vol. 142(10), 1998, pp. 240-252.
Acknowledgments
The authors are grateful to the Ministry of Education and Science of Spain, for the financing of
this research through the project "Integration of information in a methodological model applied
to decision-making in the management of agrarian productive management”, with reference:
AGL2006 - 04789/AGR.
117
Carlos José Álvarez López
Polytechnic University. University of Santiago de Compostela.
Agroforest Engineering Department
Universitary Campus s/n, 27002, Lugo, Spain.
Phone: (+34) 982 285900 ext 23262
Fax: (+34) 982 285926
118
TOPIC AREA: PRODUCT ENGINEERING
119
APPLICABILITY OF THE B-CUBE MODEL TO THE SERVICES
SECTOR
Chulvi, V.
Vidal, R.
Universitat Jaume I
Abstract
With the concept of globalization and the delegation of tasks that require large amounts of
manpower, the core of the economies of the larger towns and cities of the major world powers
has shifted to the tertiary sector (services sector). In this field, it is not enough just to remain
competitive, it is also necessary to continue to improve in order not to be pushed out of the
market by others. Functional design, which is widely and successfully used in all engineering
fields, has been proved to be a useful tool in the tertiary sector when it comes to designing
companies and establishing relations between departments based on the functions they have
inside the company. B-Cube is a new approach to functional design through the FBS
framework. This approach proposes a three-dimensional scheme that uses definitions as
Behaviours concepts. The key to this approach is that a Behaviour is not defined by a word or a
taxon, which could lead to ambiguity and misinterpretation, but rather it is defined as a threedimensional vector (X, Y, Z), which is set by its characteristics and qualities. In this work we
present the process for elaborating a functional design of a company in the services sector
through the B-Cube model, and illustrate its use with a practical application.
Keywords: FBS, Behaviour, functional design, B-Cube, business & management
1. Introduction
In recent years the tertiary sector has grown considerably, above all in first world countries. So
much has this been the case that this sector has become the engine driving the economies of
the larger towns and cities of the major world powers. One of the main causes of this trend is
globalization and the ensuing transfer of tasks with higher labour requirements, like agriculture
or manufacturing, to developing countries. This can be corroborated by observing the
percentages of participation of the third sector in the GDP (Gross Domestic Product) in
countries like the United States of America, Japan, Germany, Spain and Portugal, where this
sector contributes over 50% of their GDP (OECD, 2000; INE). For this reason, improving
competitiveness is very important for enterprises in this sector so that they are not pushed out
of the market. In other words, such enterprises need continuous improvement and innovation
(Bharadwaj and Menon, 2000; Hipple, 2003; Jacob et al., 2001; Stevens et al., 1999).
Functional design has been broadly and successfully used within all engineering fields
(Chakrabarti and Blessign, 1996; Liu et al., 1999; Zhang et al., 2002a), but nowadays it has
also been shown to be a useful tool in the services sector (Earl, 2001). The FBS (Function121
Behaviour-Structure) framework has established itself as one of the most effective and widely
extended frameworks with which to include all the parts that characterize functional design, i.e.
functions, behaviours and structures (Umeda et al., 1996; Umeda et al., 2005; Umeda et al.,
1990; Zhang et al., 2002b). Within this framework, Function represents the function or functions
performed by the design; Structure represents the set of physical elements that compose the
solution; and Behaviour acts as a link between the other two, and refers to the specific way in
which a Function is performed by a Structure.
The B-Cube (Behaviour Cube) model (Chulvi et al., 2008) is a three-dimensional tool that has
been developed in order to improve the functional design process, allow it be automated, and
make it possible to link it with CAI (Computer Aided Inventing) and environmental assessment
tools. The development of the B-Cube model is based on the upper-level ontology DOLCE
(Descriptive Ontology for Linguistic and Cognitive Engineering) (Masolo et al., 2003), the
functional proposal developed by Garbacz on this ontology (Garbacz, 2006), and Rasmussen’s
concept of SRK (Skills, Rules and Knowledge) for defining human behaviours (Rasmussen,
1983). The main characteristic of B-Cube is that Behaviours are not defined using a single word
or taxon. Instead, they are defined as three-dimensional vectors (X, Y, Z) that will determine
their characteristics and qualities.
The application of functional design to organize enterprise structures is a new field in which
several studies have been started in the past few years (Bottazzi and Ferrario, 2006; Masolo et
al., 2005; Masolo et al., 2004; Singh and Gero, 2007). This paper shows the process for
elaborating the functional design of a service sector enterprise using the B-Cube model, and an
example is also given to show how it can be applied.
2. B-Cube and Abstract Qualities
The B-Cube model was developed from the three basic definitions of DOLCE, i.e. endurants,
perdurants and qualities, with the support of Garbacz’s functional proposal on them, which were
adapted to fit the model. An endurant is defined as the entity or element (Structure) referred to
by the input into B-Cube. It is supposed that the number of elements in the universe tends to
infinity. Endurants are divided into physical endurants (PED) and non-physical endurants
(NPED). A perdurant (P) is one of the characteristics that define a Behaviour. It describes the
kind of behaviour that the aforementioned entity or element refers to. The values for perdurants
in the B-Cube model are: achievement (non-cumulative and atomic), accomplishment (noncumulative and non-atomic), state (cumulative and homeomeric) and process (cumulative and
non-homeomeric). For a better understanding of these concepts, cumulative is when the
mereological sum of two cases of the same type maintains that same type; homeomeric is
when all the temporal parts are described by the very same expression used for the whole
occurrence; and atomic is when the case is immeasurably short in time.
On the other hand, qualities are characteristics linked to another entity, and they are used to
help to define a Behaviour. There are three different kinds of qualities: temporal (TQ), physical
(PQ) and abstract (AQ). The next value needed to set a Behaviour is determined by a PQ when
the Behaviour affects a characteristic of a PED, or an AQ if it affects an NPED. PQ values are
fixed as: spatial location (the PED’s position in space); topological connectedness (the kind of
connection achieved by the PED); energy (the energetic state of the PED); magnitude (a
physical magnitude of the PED is affected); and signal (related to any sort of signal or
information flow). AQ values are skills, rules, and knowledge, and these will be explained
below. Finally, TQ constitutes the third axis. It refers to the direction in which the Behaviour
affects the SoA (State of Affairs) of the PQ or AQ mentioned above. So, the values for TQ are:
initial SoA (PQ or AQ is at the beginning and it is reduced or disappears); immutable SoA (the
Behaviour does not vary the PQ or AQ); and final SoA (the PQ or AQ is achieved at the end of
the action).
122
Figure 1 shows a graphic representation of the B-Cube model with all its values as names, but
when actually working with it numerical values will be used instead of the full terms. That is to
say, on the X axis “spatial location” = 1, “topological connectedness” = 2, “energy” = 3, and so
on. For example, instead of defining a Behaviour as (Energy, State, Initial SoA), for the sake of
convenience it will be transcribed as (3, 2, 1); or (Spatial location, Achievement, Terminal SoA)
will be given as (1, 4, 3).
Figure 1: B-Cube model.
Within the functional design of a company, the members of staff can be seen from two different
perspectives when it comes to defining them. On the one hand, they are considered to be
physical entities (PED), since they are clearly tangible structures. Their properties, therefore,
correspond to the physical field (PQ). On the other hand, however, there are the duties that
these workers carry out in the company, that is to say, the "role" that they play (Masolo et al.,
2005; Masolo et al., 2004). In this case we are talking about an NPED, the properties of which
are to be considered as lying on the abstract level (AQ). As an example, it can be seen that a
person taking a box of tomatoes to market has a Behaviour (1, 3, 3), where the first value (1)
corresponds to the PQ, while the same person trading (role: merchant) with the same tomatoes
presents a Behaviour (-2, 3, 1), where the value -2 corresponds to the AQ.
In order to understand the definitions of the AQ correctly, it is necessary to resort to the terms
coined by Rasmussen to define human behaviours: skill, rule and knowledge (Figure 2). Skill is
a previously-learned behaviour which does not need any conscious control, although such
control may have been needed during the learning process. Rule needs a procedure or “written
rule” that guides how the task is to be carried out. It does not entail an exhaustive intellectual
involvement, but the subject must bear in mind the rules to be followed at all times. Knowledge
refers to those actions that need a high conceptual level. Table 1 shows examples of the
application of Rasmussen’s concept of SRK and of the application of the above-mentioned
terms within the B-Cube model.
123
Figure 2: Algorithm for determining human behaviour.
Rasmussen
Skill, Rule and Knowledge
define a human Behaviour
B-Cube
Skill, Rule and Knowledge define an
AQ, but they do not define a
Behaviour by themselves
Examples (applicable to)
Skill
Rule
Riding a bicycle
Driver
Seeing
Secretary making a clean copy
Playing the violin
Assembly of strings on a violin
Cooking following a recipe
Racing driver
Recognizing
Secretary answering the phone
Playing the violin (with a score) Violin tuner
Knowledge
Planning
F1 engineer
Identifying
Manager (making decisions)
Composing a score
Composer
Table 1: Examples of the use of SRK.
3. B-Cube in a small enterprise in the services sector
In order to verify the applicability of the B-Cube model to the functional design of an enterprise,
a small vehicle dealer was chosen as an example. The main function of the enterprise is
defined as “selling vehicles” and as elements that are external to the enterprise, i.e. the
environment, it is necessary to define the customers and the supplier. Four different Behaviours
124
must be fulfilled so that a company can carry out the above-mentioned function (Figure 3),
namely, selling, understood as meaning that the element moves towards the customer (1, 3, 1);
buying, the element comes towards the company (1, 3, 3); invoicing, the element enters the
enterprise (1, 4, 3); and paying, the element leaves the enterprise (1, 4, 1). Since a Behaviour
is, by definition, the specific way in which a function is performed, the function of selling can be
exercised in many different ways. In this case, the vehicle dealer, the above-mentioned function
takes on a definite concrete form with these four Behaviours, and in those that will be seen
during the development of the present section.
Figure 3: Behaviours derived from Function.
At the Structure level, the first thing that is required is an element that is capable of carrying out
behaviours (1, 3, 1) and (1, 3, 3), that is, the vehicles. Yet, other structures are also needed
inside the enterprise in order to sell or purchase these vehicles (Figure 4). In this figure,
relations between Structures that carry out a Behaviour and this Behaviour are represented by
thick discontinuous lines ending in arrowheads, and relations between Structures that affect in
a secondary way and the Behaviour affected are shown as a thin dotted line ending in dots.
125
Figure 4: Behaviours (1, 3, 1) and (1, 3, 3) related to the Structure and Environment levels.
In the same way, it is necessary to establish another Structure to be able to carry out
Behaviours (1, 4, 1) and (1, 4, 3): Euros (€). These are fully defined with the Structures ‘trader’
and ‘administrative assistant’, and with the environment elements ‘customer’ and ‘supplier’
(Figure 5).
Figure 5: Behaviours (1, 4, 1) and (1, 4, 3) related to the Structure and Environment levels.
126
At this level, it is necessary to define the roles and the elements that are going to perform them
in order to be able to continue with the functional level of an enterprise. First of all, only two of
the defined Structures are capable of carrying out roles, i.e. ‘trader’ and ‘administrative
assistant’. Abstract or role Behaviours are derived from the previously defined Behaviours on a
second level. Thus, in order to manage the sale of a vehicle, both the ability to process the sale
(-1, 3, 1) and the ability to negotiate the conditions under which it will take place (-2, 1, 2) are
needed. The Behaviours ‘placing orders’ (-2, 4, 3) and receiving payment (-1, 3, 3) are
established in a similar way. At this level, a differentiation is drawn between ‘paying' in the
sense of making a payment to the supplier (-1, 3, 1) and that of elaborating the payslips for the
workers (-1, 3, 1). Apart from Behaviours derived from the first level, which can be
accomplished by the existing Structures, the condition “designing an enterprise” requires a
Behaviour of management and coordination (-3, 2, 1) over the rest of the role Structures. This
new Behaviour needs another Structure that is capable of being carried out with it: the director.
This Structure must be reanalysed with the rest of the Behaviours in the scheme in order to see
whether it is related to any of them. If this is the case, then it is related with the Behaviour (1, 4,
1). Figure 6 shows the direct relations between the role Behaviours and the Structures capable
of carrying them out. The parameters (X, Y, Z) of the derived Behaviours do not need to be
similar to the corresponding ones on the upper level, since the Structure that is directly affected
is not the same.
Figure 6: Role Behaviours and their directly affected Structures.
Lastly, Figure 7 shows the definitive scheme representing the final result of the functional
design of the enterprise chosen as an example in this paper. Regarding derived Behaviours,
secondary relations have been omitted in order to simplify the scheme, but these relations
correspond to the other Structures affected by the first-level Behaviour. That is, Behaviour (1, 3,
1) has ‘vehicles’ as its primary Structure and ‘trader’ and ‘customer’ as the secondary ones. So,
since the Behaviour (-1, 3, 1) has its primary relation with ‘trader’, then its secondary relations
are with ‘vehicles’ and ‘customer’.
127
Figure 7: Functional design of a service-sector enterprise through the B-Cube model.
4. Conclusions
The process of developing a functional design for a service-sector enterprise has been shown
in this paper. This functional design was carried out within the FBS framework and through the
B-Cube model. An example has also been used to show how it can be applied in enterprise
structure organization. Thus, Figure 8 shows the Structures needed to establish the
organization chart of an enterprise. This will be performed with the Structures that are capable
of carrying out with role Behaviours (those which are in the negative part of the X axis of the BCube). In this example, a trader, an administrative assistant and a director are needed for the
organization to work properly. The volume of the other Structures will determine the number of
the first ones. That is, if we look at the flow of vehicle transactions in the enterprise, which is
directly related to Behaviour (1, 3, 1) and which is at the same time indirectly related with the
‘trader’, the higher the number of vehicles on sale, the more traders will be needed. So, it is
necessary to establish quantity ratios between Structures in order to know how many vehicle
sales are needed to increase the number of traders.
It can be appreciated from the development of functional design that structures capable of
having a direct relation with role Behaviours only have this kind of relations with them, but they
can have secondary relations with the rest of the Behaviours. Classifying the Behaviours on
different levels helps in the visualization and comprehension of the functional design, although
there is no direct relation between the kind of Behaviour and the level it is placed on. Therefore,
although in the example the Behaviour (-3, 2, 1) appears on the second level, this is only to
make the scheme clearer and easier to understand by showing all role Behaviours on same
visual level. It can be seen that it is not derived from any other Behaviour, so it is considered to
be primary level Behaviour.
128
(X, Y, Z) parameters of the derived Behaviours do not have to be related to first-level ones
because the Structure which they act directly on are different. Thus, X parameters are normally
different because they come from a direct quality of the Structure. The Y parameter refers to
the kind of Behaviour, so it is completely independent with regard to other Behaviours.
However, as Z parameters are related to the action direction, both first-level and derived
Behaviours will normally have the same value, but this is not a hard and fast rule. For example,
on looking at Figure 6 it can be seen how practically all the second-level Behaviours share their
Z value with their origin Behaviour. So, if the Behaviour buy (1, 3, 3) a vehicle means that this
vehicle will be inside the enterprise at the end of the action (terminal SoA Æ Z = 3), it is logical
to expect its derived Behaviour ‘place orders’ (-2, 4, 3) to have the same direction, since the
final result of the action will be the vehicle inside the enterprise. The exception can also be
seen in the same example. In the Behaviours derived from selling (1, 3, 1) it can be seen that
one of them, negotiating (-2, 1, 2) does not share the direction, since it does not necessarily
entail the vehicle’s leaving the enterprise (sold) or its remaining there (not sold). Therefore, it
can be deduced that although the Z values usually coincide between one Behaviour and its
lower-level derivates, it can be said that the three coordinates remain fully independent
between levels.
Logically, although the example in the present paper has focused on a services sector (or third
sector) enterprise, the procedure can be applied to functional design of enterprises from other
sectors (first sector or raw materials, and secondary sector or industry). This work is part of a
project that has as its objective the creation of a computer application for functional design
automation. Thus, the B-Cube model is perceived as an optimization of the process of defining
the structural organization of an enterprise and its resources.
References
Bharadwaj, S, Menon, A., 2000. Making innovation happen in organizations: individual creativity
mechanisms, organizational creativity mechanisms or both?. Vol. 17, pp. 424 - 434.
Bottazzi, E, Ferrario, R., 2006. Preliminaries to a DOLCE Ontology of Organizations.
International Journal of Business Process Integration and Management.
Chakrabarti, A, Blessign, L., 1996. Representing functionality in design. Artificial intelligence for
engineering design, analysis and manufacturing (AIEDAM), Vol.10, pp. 251-253.
Chulvi, V, Vidal, R, Cebrian-Tarrasón, D., 2008. B-Cube. XII Congreso Internacional de
Ingeniería de Proyectos (AEIPRO). Zaragoza.
Earl, M., 2001. Knowledge Management Strategies: Toward a Taxonomy. Journal of
Management Information Systems Vol. 18, pp. 215-233.
Garbacz, P. 2006. Towards a standard taxonomy of artifact functions. Applied Ontology Vol. 1,
pp. 221-236.
Hipple, J., 2003. So, you want to start an innovation effort. Research Technology Management
Vol 46, 11-13.
Jacob, M, Tintoré, J, Torres, X., 2001. Innovación en servicios. Fundación Cotec, Madrid.
Liu, Y, Chakrabarty, A, Bligh, T., 1999. A strategy for functional design. ICED´99. Munich.
Masolo, C, Borgo, S, Gangemi, A, Guarino, N, Oltramari, A., 2003. WonderWeb Deliverable
D18. Laboratory For Applied Ontology - ISTC-CNR.
129
Masolo, C, Guizzardi, G, Vieu, L, Bottazzi, E, Ferrario, R., 2005. Relational roles and quaindividuals. AAAI Fall Symposium Roles, an Interdisciplinary Perspective: Ontologies,
Languages, and Multiagent Systems. Arlington, Virginia, pp. 103–112.
Masolo, C, Vieu, L, Bottazzi, E, Catenacci, C, Ferrario, R, Gangemi, A, Guarino, N. 2004.
Social roles and their descriptions. Ninth International Conference on the Principles of
Knowledge Representation and Reasoning,. Whistler, Canada.
Rasmussen, J., 1983 Skills, rules, and knowledge; Signals, signs and symbols, and other
distinctions in human performance models. IEEE Transactions on systems, man, and
cybernetics, SMC-13, pp. 257 - 266.
Singh, V, Gero, JS., 2007 Computational modeling of temporary design teams. In: Y Gang, Z,
Qi and D Wei (eds), editor. CAADRIA07. Southeast University, Nanjing, China, pp. 49-56.
Stevens, G, Burley, J, Divine, R., 1999. Creativity + business discipline = Higher profits faster
from new product development. The Journal of Product Innovation Management, the juried
publication of the PDMA (Product Development and Management), Vol.16, pp. 455-468.
Umeda, Y, Ishii, M, Yoshioka, M, Shimomura, Y, Tomiyama, T., 1996. Supporting conceptual
design based on the function-behaviour-state modeller. Ai Edam, Vol. 10, pp. 275-288.
Umeda, Y, Kondoh, S, Shimodura, Y, Tomiyama, T., 2005. Development of design
methodology for upgradable products based on function-behaviour-state modelling. Artificial
Intelligence for Engineering Design, Analysis and Manufacturing (AIEDAM), Vol. 19, pp. 161182.
Umeda, Y, Takeda, H, Tomiyama, T, Yoshikawa, H., 1990 Function, behaviour, and structure.
In: Gero, J, editor. Applications of Artificial Intelligence in Engineering V. Berlin: Springer, pp.
177-194.
Zhang, WY, Tor, SB, Britton, GA., 2002a. Automated functional design of engineering systems.
Journal of Intelligence Manufacturing, Vol. 13, pp. 119-133.
Zhang, WY, Tor, SB, Britton, GA, Deng, YM., 2002b Functional design of mechanical products
based on behaviour-driven function-environment-structure modelling framework. Singapore, p
8.
Acknowledgements
The authors would like to acknowledge funding by the Spanish Ministerio de Educacion y
Ciencia (ref. DPI2006-15570-C02-00) and the FEDER funds from the European Union. We are
also grateful to the other members of the Engineering Design Group (GID) for their help with
this work.
130
Vicente Chulvi Ramos
Grupo de Ingeniería del Diseño (GID)
Departamento de Ingeniería Mecánica y Construcción. Universitat Jaume I.
Avenida Sos Baynat s/n. 12071 - Castellón (Spain)
Teléfono: +34 964 72 9252 Fax: +34 964 72 8106
URL: www.gid.uji.es
Rosario Vidal Nadal
Grupo de Ingeniería del Diseño (GID)
Departamento de Ingeniería Mecánica y Construcción. Universitat Jaume I.
131
PARADIGMS OF EXPLORATION AND PARADIGMS OF
EXPLOITATION.THE ARTIFICIAL DISC CASE
Barberá-Tomás, D
De los Reyes, E.
Jiménez-Saéz, F.
Institute of Innovation and Knowledge Management INGENIO (CSIC-UPV)
Abstract
Technological paradigms are heuristic models which guide organizations in their learning and
searching process for solutions of technological problems, as the seminal contribution of
Giovanni Dosi puts it. In the organizational learning literature, this model of solution is divided in
exploration paradigms and exploitation paradigms. Although Dosi’s analysis identifies properly
the exploitation paradigms, it does not happen the same with the exploration paradigms. This
article pretends to identify and analyse the heuristic behaviour of two technological paradigms
(both exploration and exploitation ones), used to solve a specific technological problem: the
substitution of a diseased intervertebral disk with an artificial prostheses. For this we will study
carefully the design characteristics of both paradigms; also, a quantitative analysis of a patent
citation database of this kind of artefacts is performed.
Keywords: technological paradigm, artificial disc, product design
1. Introduction
In his seminal article, Giovanni Dosi (1982) proposed to transfer some of the historicist notions
of philosophy of science developed by TS Kuhn (1969) to technological evolution. A Kuhnian
scientific paradigm includes the set of theories that shape the kind of problems to be studied by
‘normal’ science and, in this context, for Dosi a technological paradigm is a “model and pattern
of solutions to solve some technological problems, based on certain principles derived from the
natural sciences and certain materials" (italics in original), where the words in italics relate to
the particular nature of each paradigm. In addition to this technological adaptation of the
philosophical concepts of science to technology, Dosi’s early work has had a profound
influence on evolutionary economics, which deals mainly with technological change.
The evolutionary perspective suggests that the behaviour of economic agents in contexts of
change (e.g. technological change), must be based on learning and searching processes and
on the bounded rationality of the agents (Simon, 1957). Technological paradigms constitute the
set of heuristic rules that guide these processes of learning in the environment, which in its turn
will shape the technological trajectories followed by the agents involved in the process of
innovation. In modern economies, these agents are corporations (Dosi, 1988).
132
The processes of organizational learning and search have been characterized as exploration or
exploitation heuristics (March, 1991). Exploitation consists of learning through the refinement
and extension of previously existing knowledge, (Larrue, 2003) while learning through
exploration occurs through the combination of existing knowledge and new knowledge
(Abernathy and Clark, 1985). Exploitation includes local searches, close to the technological
point of departure of the searching agent; exploration implies a process involving 'long jumps' to
remote areas of the technological space (Hovhannisian, 2004). In the same way, technological
paradigms can be characterized by these two search heuristics: while exploitation paradigms
are based on the refinement of existing knowledge and local search, exploration paradigms are
characterized by the combination of new with existing knowledge, the former originating in
remote areas of technological space.
The original work by Dosi identifies exploration paradigms with an organization’s areas of
expertise and, therefore, with existing knowledge, which exploitation refines (Dosi: 1982: 155).
However, Dosi situates exploratory paradigms as originating exclusively in the scientific
advances that act as a deus ex machina. This trend to "exogenization" of exploration may be
due to the ambiguous nature of the innovation process. Innovation is inherently uncertain in
nature and impact: were it certain and known about it would not be innovation (Silverberg and
Verspagen, 2005). However, the emergence of new technologies is far from a purely stochastic
phenomenon and usually occurs in very specific areas of the technological space, regardless of
the degree of novelty (Silverberg, 2002). These seemingly paradoxical properties are explained
partially in terms of the creation of knowledge through exploration, as a phenomenon that
emerged due to interactions between the organization and others agents in the innovation
system. Thus, instead of "science" being an abstract black box which produces new knowledge
in unexplained ways, this perspective enables the study of learning by interaction among the
players, which may be scientific institutions but also could be users or providers or other
organizations (Kamp, 2002). In other words, the "new knowledge" that emerges from
interactions is the essence of organizational learning by exploration, as confirmed by several
empirical studies devoted to joint ventures (Koza and Lewin, 1998).
In order to study exploratory paradigms, therefore, it is necessary to focus on the interactions
among the players in innovation system under review. These relationships are strongly
sectorally-specific (Malerba, 2002) and can be categorized in the taxonomy devised by Pavitt
(Pavitt, 1984), which classifies industries into four sectors according to the agents involved: (1)
supplier dominated; (2) specialized suppliers, (3) scale intensive; and (4) science-based.
This paper identifies and analyses heuristic paradigms of exploration and exploitation involved
in the resolution of a particular technological problem: the replacement of an anatomical disc by
an intervertebral disc prosthesis. Section 2 provides a detailed study of the case of the artificial
disc, which distinguishes the technological paradigms of exploration and exploitation based on
existing capabilities and patterns of interaction among agents. In section 3, the analysis of a
patent database for these devices deepens our understanding of the heuristic rules and
searching behaviour in both types of paradigms. Section 4 discusses the major findings and
limitations of our study.
133
2. The identification of paradigms: a case study
2.1 Introduction
Arthroplasty of the intervertebral disc (i.e., artificial disc replacement using an implantable
device) is a surgical procedure used in the treatment of degenerative disc disease (hereafter
DDD). DDD is the major cause of pain and disability in developed countries (Errico, 2005).
Intervertebral disc arthroplasty has been postulated as being a more efficient alternative
treatment for DDD than other surgical procedures, such as vertebral arthrodesis, which has
been used systematically since the late 1970s to treat DDD (Acosta et al., 2005). Arthrodesis
involves the fusion of the two vertebrae adjacent to the degenerated disc. Arthrodesis
eliminates abnormal movement of the damaged disc, thus removing the pain this causes.
However, fusing two previously articulated vertebrae involves several biomechanical alterations
in the behaviour of the vertebral column: (i) the movement of the instrumented articulation must
be incorporated into the discs of the adjacent vertebral articulations; (ii) the loads that were
absorbed by the now immobilized disc must be absorbed by the discs in the adjacent
articulations. These kinematic and dynamic alterations following arthrodesis can provoke
‘adjacent disc degeneration syndrome’: following fusion, excess movement and load, result in
the appearance of DDD in the discs adjacent to the fused ones, frequently requiring some kind
of surgical intervention (Denoziere and Ku, 2006). The insertion of an artificial disc is to try to
avoid these problems associated with vertebral fusion.
In the history of artificial discs, one fact in particular stands out. As previously stated, an
artificial disc must fulfil two tasks in order to prevent the DDD syndrome of the segments
adjacent to the disc that was originally symptomatic: it must not cause any alteration to the
kinematic or dynamic conditions of the anatomic intervertebral discs in these segments.
However, as we show in the next section, commercially available discs meet only one of these
prerequisites: preservation of the kinematic properties of adjacent discs. None of the artificial
discs that have been produced commercially are able to mimic the dynamic properties of the
anatomical disc in being able to absorb loads (Oskouian et al., 2004), since they are all
constituted of articulations between materials (polyethylene, Cr-Co, steel…) with much smaller
load absorption capacities than the anatomical disc.
The medical literature has formulated a hypothesis to explain this shortcoming (Szpalski et al.,
2002): designs for vertebral disc articulations were influenced by the enormous efficacy of the
implants used to replace hip articulations. In the early 1960s Sir John Charnley designed the
hip prosthesis, which some years later became the standard implant in the treatment of
degenerative hip disease. Charnley’s hip prosthesis (and succeeding designs) was integrated
by two spherical articulation elements (‘ball-and-socket’) made of rigid materials. This rigidity
does not imply a functional shortcoming, since the anatomic articulation of the hip does not
have the viscoelastic ‘shock absorbing’ properties of the intervertebral disc.
Hence, transferring the hip prosthesis principles to disc replacement, as in the artificial discs
commercialized to date, implies the assumption of an important deficit: if the artificial disc
cannot absorb loads, the corresponding overloads will be charged on the adjacent discs at the
instrumented level, which can lead to the degeneration of these discs (McNally, 1996). This
possibility of adjacent degeneration is the main reason why arthrodesis (vertebral fusion) is
considered to be a suboptimal procedure in the treatment of DDD.
However, in the history of artificial disc design we can find a considerable amount of past and
current technological activity in other directions than the hip-like artificial disc, although without
134
the same market success. Szpalski et al. (2002) reviewed currently marketed artificial discs,
and also several devices tested only in vitro or in animal models, and devices that had been
clinically tested, but never commercialised. They found that there were many artificial disc
designs based on the principle of ‘shock absorbing’ biomaterials that actually enabled the
reproduction of viscoelastic properties of the anatomical disc, but that they were limited to nonmarket applications. They concluded that artificial disc designs can be categorised according to
two principles: shock absorbing ‘dynamic’ designs, or the already described ‘hip like’ devices
commercialised to date. Other reviewers of artificial discs, such as Lee and Goel (2004), agree
with this classification. In this paper, we employ the term ‘paradigm’ rather than ‘principle’ to
refer to the ‘dynamic’ and the ‘hip like’ solutions.
2.2 The exploitation paradigm.
In characterizing the paradigms, we referred to local search and refinement and the extension
of existing knowledge to define the exploitation heuristic process. Here, what seems to be
important is the prior existence of prior knowledge. So, in trying to identify exploitation in the
invention of the artificial disc, we need to concentrate on those activities that employ knowledge
that is already possessed by the innovative agent.
Again, as already mentioned, many of the artificial discs that have been developed (and all
those commercialised to date) ‘apply the design principles commonly used in total hip
arthroplasty’ (Fraser et al., 2004). These hip design principles can be identified with the above
mentioned local search strategies, because the hip-like paradigm for the artificial disc consists
basically of recombining the components of hip prostheses and adapting them to spinal
anatomy (Santos et al., 2004). This reutilization of an early technology for use in other
applications than the original corresponds to what we characterize as the exploitation paradigm.
The following paragraphs propose some historical justifications for agents developing implants
within the hip-like paradigm, to refine the knowledge previously developed in the design of hip
prostheses.
Since Sir John Charnley’s development of a hip prosthesis in the 1960s, the orthopaedic
industry has been marked by developments in total hip replacement to the point that some
reviewers refer to pre-Charnley and post Charnley eras. Charnley’s discovery also became the
basis for the subsequent development of prosthetic knees, shoulders and elbows (Fraser et al.,
2004).
The first disc prosthesis to be marketed in the world (and the only one until 1999) was the SB
Charité, based on a 1982 design in the form of two metallic platforms encrusted into the
adjacent vertebrae, which were articulated with a polyethylene body. A German company
(Waldemar Link), dedicated to the manufacture of hip implants since the late 1960s, was on the
development team. This is an important part of the story: in the mid 1980s, artificial hip joints
were already successful and used technological capabilities that were built on the ‘ball-andsocket’ paradigm established by Sir John Charnley. Waldemar Link brought its technological
experience in artificial hip manufacture. Following incidents of fractures of the prostheses
endplates in the non-forged stainless steel, a new Co-Cr cast alloy was developed and began
to be used, with the SB Charité becoming commercially available in Europe in 1987 (Link,
2002).
Ultimately, the paradigm, which we call "hip like", allowed the creation of knowledge (through
refinement to and extension of existing knowledge) in the invention of a new prostheses for the
replacement of a different articulation (the disc), using the benefits of tribological knowledge
135
previously developed with great success for synovial joints such as hip and knee
(Szpalski,2002). Therefore, we can identify this paradigm as the exploitation paradigm.
2.3 The exploration paradigm
In our definition of exploration activities in the first section, we indicated the long jump search
and the recombination of knowledge as properties of the heuristic process. We also discussed
that the exploration of new knowledge is not random: there are search areas that are more
likely to be explored. These areas correspond to the location of the innovative agents that are
part of the knowledge base specific to the sector (Malerba, 2002).
The medical equipment industry has often been conceptualized as a capital input in the medical
services sector (Geljins et al., 1996). There is also evidence that medical surgical equipment is
an area where user innovation is relevant (see, e.g. Kahn, 1991). Therefore, it seems justified
to classify this sector, according to Pavitt’s (1984) taxonomy, as a typical ‘specialised supplier’
sector. In such sectors, innovative activities relate mainly to product innovations, which enter
other sectors as capital inputs; firms within this classification operate in close contact with
users. Specifically related to medical surgical equipment, recent empirical studies have proved
that surgeons are the source of radical new paradigms and influential innovations (Lettl et al.,
2006). For medical technology, the users are the professionals: they use the medical
technology in their lives, and can apply their ‘professional’ knowledge to develop innovative
solutions. The physical properties of the spine can be regarded as the ‘basic science’ of
orthopaedic spinal surgery, as they are the professional knowledge indispensable to
interpreting, analysing and, if necessary, correcting abnormal human anatomy and physiology
(White y Panjabi, 1978). The ‘shock absorbing’ properties of the spinal anatomical disc have
been known about since 1954, when Hirsch and Nachemson (1954) published the results of
compression tests. Nachemson (1962) was the first person to propose replacement of
interverbral discs, and reported trials using silicon (a material with shock absorbing properties)
prostheses.
Basalla (1989) has shown that the origin of some invention is the imitation of natural objects (or,
as the author called "naturfact”). In the case of the disc prosthesis, this mimetic force is even
more powerful, since the device is functionally designed to replace a "naturfact": an organ of
the human body. This imitative mechanism can be seen clearly in the Nachemson’s journey
from the original description of the dynamic properties of the intervertebral disc in 1954 until the
first test of an artificial prosthesis in 1962. This biomechanical knowledge possessed by
advanced users as surgeons is the remote area of technological space where the explorative
interaction takes place.
3. The heuristic properties of paradigms: analysis of patent citations
3.1 Design of research, data and analysis.
Section 2 identified and discussed the paradigms of exploration and exploitation in the invention
of the artificial disc; in this section we apply quantitative methods to study the heuristic
properties of these paradigms. To do this, we use the results from an analysis of patent
citations. Our database consists of 303 patents for artificial discs recorded between 1972 and
2004, derived from a keyword search of US Patent and Trademark Office (USPTO) and the
European Patent Office (EPO) patents. After a review of the technical information with the help
136
of two experts (a spine surgeon and an engineer involved in designing artificial discs) we
classified the patents into hip-like or dynamic paradigm. Thus, we include “richer information
than the categories [listed by the International System of Patent Classification (IPC)] allow”
(Fleming and Sorenson: 1026).
In addition to a detailed description of the patented device or process, patents include citations
to one or more patents. These citations are mainly a legal instrument. They are an indication of
the knowledge described as patentable, and that part of the knowledge claimed by a previous
patent(s). Also, as proposed by several authors (see Jaffe and Trajtenberg, 2002, for a review)
citation to earlier patents indicates that the knowledge developed in the cited patents was used
to build the knowledge contained in the citing patent.
This information is useful in studying the heuristic properties of paradigms. If the inclusion of a
citation to an earlier patent means a "construction" of knowledge, we can assume that these
patents are in the proximity of multidimensional technology space (Silverberg and Verspagen,
2005) of the citing patent. Indeed, for the creation of new technological knowledge it is
necessary for closely related knowledge in the same technological space to be already
“discovered” (Silverberg, 2002). We can identify this neighbouring knowledge through patent
citations.
In Section 1 we referred to some of the limitations of exploratory paradigms, and especially that
searches in remote areas of the technological space are more uncertain than in the case of
exploitation, where searches are conducted in areas adjacent to the technological space in
which the agent is innovating. This uncertainty is reflected by the larger number of citations to
patents in exploitation areas. Therefore, we can hypothesize that:
Hypothesis 1: Patents belonging to the exploitation paradigm are more frequently cited than
patents in the exploration paradigm.
On the other hand, contiguity of the technological space can acquire different meanings
depending on the dimensions of the technological space studied. Citations in patents are
included by the inventors but can also be added by the patent examiner. Jaffe et al. (2002),
based on interviews with inventors and patent examiners, find that patent citations refer to two
types of prior knowledge: application-related and technology-related. If patent X cites patent Y,
this might mean that patent X represents an alternative way of doing something that has
already been discovered (e.g., patent X is for a mousetrap incorporating a titanium spring which
improves the performance of the patent Y trap which uses zirconium). Alternatively, it could be
that patent X uses a similar technology to patent Y, but for a different application (patent X
relates to a mouse trap incorporating a titanium spring and patent Y is for a pen incorporating a
titanium spring). We consider the first case "application-related" and the second case
"technology-related."
According to this classification, citations between exploitation and exploration paradigms are
application related, since, as discussed in Section 2.1, both technologies are unique: if a device
is driven by the operational principle of one of the paradigms (kinematics of rigid bodies or the
viscoelastic behaviour of materials) it cannot relate to the other, so that citations across
paradigms can never be technology-related. This is the "incommensurability" of technological
paradigms referred to by Dew (2006), translating Kuhn’s concept of incommensurable scientific
paradigms. Assuming that this aspect of technological space (application-related) is the same
for both paradigms, there is no reason, in principle, why the numbers of application-related
citations to be different (exploratory to exploitation, and vice versa). Thus, we can hypothesize
that.
137
Hypothesis 2: The number of cross-citations between paradigms will be the same regardless of
whether the citation is from the exploratory to the exploitation paradigm or vice versa.
This distinction between citations is the basis for our third and fourth hypotheses. According to
the theoretical framework described in Section 1, exploitation paradigms refine existing
knowledge. Therefore, in the exploitation paradigm, all other citations will be intra-paradigmatic,
i.e. will be to patents within the same paradigm.
In the case of the exploratory paradigm, citations to existing knowledge can be applicationrelated and intra-paradigmatic citations. However, the element of new knowledge that
characterizes the exploratory heuristics is not included in these categories. Citations to new
knowledge will refer to extra-paradigmatic patents not related to artificial discs, as the
technology-related search will have extended to farther areas of the technology space.
Hypotheses 3 and 4 reflect this proposition.
Hypothesis 3. Patents cited in the paradigm of exploitation (in addition to patents related to the
application) belong to the same paradigm.
Hypothesis 4. Patent citations in the exploratory paradigm include (in addition to patents related
to the application) patents for technologies unrelated to either paradigm.
3.2 Resultados
Table 1 shows the results of our analysis:
Cites to
exploration
patents
Cites to
Number
Citation
number
Media
Cites to
explotaition
patents
Exploitation
patents
161
457
2,83
227 (49,67%)
54 (11,82%)
176 (38,51%)
Exploration
patents
150
336
2,24
34 (10,12%)
130 (38,69%)
172 (51,19%)
extraparadigmatic
patents
Tabla 1: Descriptive statistics of the patent database.
Hypothesis 3 is not confirmed by these results; hypothesis 3 suggests that the exploitation
paradigm does not make extra-paradigmatic citations, based on the idea that the exploitation
heuristic refines the existing knowledge, and that therefore intra-paradigmatic and applicationrelated patents will be cited. However, 38.51% of our patent citations refer to extraparadigmatic patents not describing artificial discs. The deductive argument related to
Hypothesis 4 is similar, therefore this hypothesis is also rejected.
138
4. Discussion and limitations
This paper identifies the exploration and exploitation paradigms used to solve a technological
problem (replacement of an anatomical disc by an artificial prosthesis) and analyse the heuristic
searches of both paradigms. The aim was to deepen our understanding of Dosi’s technological
paradigms, through the concepts of organizational learning and behavioural theory of
organizations, at the sectoral level.
However, our statistical analysis contradicts our hypothesis that the exploration paradigm only
refines knowledge. Since the existence of the two paradigms described in our case study has
been confirmed by other historians writing about the development of the artificial disc (Szpalski
et al., 2002; Lee and Goel, 2004), the problem must lie in our second theoretical assumption
related to the type of knowledge that can be identified with an agent in the innovation system.
The concept of learning by interaction in the innovation system helps to resolve these
problems: the type of agent with knowledge can be indicative of the type of heuristics that we
can assign to a paradigm (exploratory and exploitative), but only based on a historical analysis
to determine the type of agent that originally possessed the knowledge, since the same
interactions between the players leads to the knowledge being distributed throughout the
innovation system, which explains the failure of Hypothesis 3 and 4. This type of historical
analysis is well suited to the case study described in Section 2, but it invalidates our analysis of
patent citations in Section 3. Although this type of research is used in the study of patent
databases (a recent example is Lettl et al., 2006), the statistical refutation of some of our
hypotheses proves that a dynamic framework is best suited to studying the essential historicity
of evolutionary analysis (Nelson and Winter, 1982).
5. References
Abernathy, W. and Clarck, KB, 1985. Innovation: Mapping the Winds of Creative Destruction,
Research Policy, 14, pp 3-22
Acosta, F.L. Jr, Aryan, H.E., Ames, C.P., 2005. Emerging directions in motion preservation
spinal surgery. Neurosurgical Clinical North America 16(4), pp. 665-669.
Basalla, G., 1989. The Evolution of Technology: Cambridge Studies in the History of Science,
Cambridge.
Denoziere, G., Ku, D.N., 2006. Biomechanical comparison between fusion of two vertebrae and
implantation of an artificial intervertebral disc. Journal of Biomechanics, 39(4), pp 766-775.
Dew, N., 2006. Incommensurate technological paradigms? Quarreling in the RFID industry.
Industrial and Corporate Change 15(5), 785-810
Dosi, G., 1982. Technical paradigms and technological trajectories, Research Policy, Vol.
11(3), pp147–162.
Dosi, G., 1988. The nature of the innovative process, pp. 221-239 in: Dosi, G., Freeman, C.
Nelson, R., Silverberg, G., Soete, L. (eds.) Technical Change and Economic Theory (London:
Pinter).
139
Errico, T.J., 2005. Lumbar disc arthroplasty. Clinical Orthopaedics & Related Research 435,
106-117.
Fleming, L., Sorenson, O., 2001. Technology as a complex adaptive system: evidence from.
patent data. Research Policy 30(7), 2001, 1019–1039
Fraser, R.D., Ross, E.R., Lowery, G.L., 2004. Freeman, B.J., et al. Lumbar disc replacement.
AcroFlex design and results. Spine J 4, pp 245S-251S.
Frenken, K., 2006. Innovation, Evolution and Complexity Theory. Cheltenham UK and
Northampton MA.
Gelijns, A., Rosenberg N., Laubach, G., 1996. Implantable and Surgical Medical Devices.
National Academy Press, Washington DC.
Hirsch, C., Nachemson, A., 1954. A new observation of the mechanical behaviour of lumbar
discs. Acta Orthopedica Scandinava 23, 1954, 254.
Hovhannisian, K., 2004. Depth and Breadth of Search on a Technology Landscape, DISA,
University of Trento.
Jaffe A. y Trajtenberg, M. 2002. Eds. Patents, Citations and Innovations. MIT Press,
Cambridge, Massachusetts, USA.
Kahn, A. 1991. The dynamics of medical device innovation: An innovator's perspective, in: [28]
Genjins A., Halm E. (Eds.) The Changing Economics of Medical Technology, The National
Academies Press, Washington DC, 89-95.
Kamp, L., 2002, Learning in wind turbine development: A comparison between the Netherlands
and Denmark, PhD thesis, Utrecht University.
Koza, M. P.and Lewin, A. Y., 1998. The co-evolution of strategic alliances. Organization
science, vol. 9, nº 3, p. 255-264.
Kuhn, T. S., 1969. La estructura de las revoluciones científicas, México D.F.: Fondo de Cultura
Económica.
Larrue, P., 2003. Coping collectively with the exploration—exploitation trade-off in research
consortia. Pp. 150-178. In Saviotti, P. Applied Evolutionary Economics: New Empirical Methods
and Simulation Techniques Edward Elgar Publishing.
Lee, C., Goel, V., 2004. Artificial disc prosthesis: design concepts and criteria. The Spine
Journal, Volume 4, Issue 6, pp S209-S218
Lettl, C., von Wartburg, I., Rost, K., 2006. Technological merit of user invention: achieving
impact by exploration or exploitation? Brown-Bag seminar, Prof. G. von Krogh, June 22t, ETHZurich, Switzerland.
Link, H. D. History, design and biomechanics of the LINK SB Charité artificial disc Eur Spine J,
11 (Suppl. 2), pp S98–S105
Malerba, F., 2002. Sectoral systems of innovation and production. Research Policy 31, 247264.
March, J., 1991. Exploration and exploitation in organizational learning, Organization Science,
Nº2, pp. 71-87
McNally DS, Shackleford IM, Goodship AE, Mulholland RC., 1996. In vivo stress measurement
can predict pain on discography. Spine. Nov 15;21(22), pp 2580-7
140
Nachemson, A., 1962. Some mechanical properties of the lumbar intervertebral disc. Bull Hosp.
Joint Dis. 23, pp130-132.
Nelson, R.R., Winter, S.G., 1982. An Evolutionary Theory of Economic Change (Cambridge MA
& London: Belknap Press of Harvard University Press).
Oskouian RJ, Whitehill R, Samii A, Shaffrey ME, Johnson JP, Shaffrey C, 2004. The future of
spinal arthroplasty: a biomaterial perspective. Neurosurgical Focus, Vol. 15, 17 (3):E2
Pavitt, K., 1984. Sectoral patterns of technical change, towards a taxonomy and theory.
Research Policy 13, 343-373.
Santos, E.G., Polly, D.W., Mehbod, A., Saleh, K.S., 2004. Disc arthroplasty: Lessons learned
from total joint arthroplasty. The Spine Journal 4, pp 182S-189S.
Silverberg, G., 2002. The discrete charm of the bourgeoisie: quantum and continuous
perspectives on innovation and growth, Research Policy, Vol. 31, no. 8-9, 1275-1291.
Silverberg, G. and Verspagen, B., 2005. Self-organization of R&D search in complex
technology spaces, ECIS Working Paper 05.07.
Simon, Herbert. A Behavioral Model of Rational Choice, in Models of Man, Social and Rational:
Mathematical Essays on Rational Human Behavior in a Social Setting. New York: Wiley, 1957
Szpalski, M., Gunzburg, R., Mayer, M., 2002 Spine arthroplasty: A historical review. Euro Spine
Journal Supplement 2, pp. S65-84.
Vence Deza, X., 1995. Economía de la innovación y del cambio tecnológico: una revisión
crítica. Madrid: Siglo XXI de España.
White, A.A., Panjabi, M.M., 1978. Clinic Biomechanics of the Spine. J.B. Lippincott Co.,
Philadelphia, PA.
Acknowledgements
This work has been supported by the Plan Nacional de I+D+i, 2004-2007 (SEJ200508603/PSIC), financed by the Spanish Ministerio de Educación y Ciencia (MEC), “Influencia de
la cultura organizacional y los flujos de conocimiento en el rendimiento científico de grupos de
I+D tecnológicos. Aplicación a la Universidad Politécnica de Valencia.”
David Barberá Tomás [email protected]
141
MODELS FOR AIRBAG SIMULATION IN VEHICLES ADAPTED FOR
DISABLED DRIVERS
Masiá, J.
Eixerés, B.
Dols, J.
Abstract
The car control adaptations are used in vehicles in order to facilitate the driving to persons with
physical handicaps. This does not have to suppose a decrease of the passive safety that is
required to the vehicles. In order to analyze this relation there will be characterized the different
control adaptations that are in use together with the different devices of passive safety that can
be mounted in the vehicles in diverse cases of impact in order to generate models of
simulation. The methodology used to generate this simulation consists of the first phase in
which there develops the three-dimensional model of the driving place. For it, there has been
used a commercial software of three-dimensional design. Once realized this one divides, the
model is imported to the finite elements software in which meshing is generated. Finally, a
dynamic simulation software is used to assign the most important characteristics like material
properties, contact interfaces, gas expansion models, airbag fold types, etc.
Keywords: airbag, passive security, disabled driver, steering control devices, finite elements
1. Introduction
An increasing number of components are used in vehicles. These components reflect a
functional design that meets various requirements such as passive safety. These requirements
should also be met by any control adaptations fitted to enable disabled people to drive. While
the main requirement of these adaptations is to enable individuals with various disabilities to
drive, their use need not imply a decrease in the vehicle passive safety level. Consideration
should also be given to recycling these components in compliance with established
environmental policies.
The redesign of these components is considered so that they meet functional requirements and
increase the passive safety level for disabled drivers. The environmental repercussions of the
materials and processes used in design and manufacturing are also considered.
To undertake such a redesign, the various situations that may occur are analysed – together
with the passive safety devices that are currently fitted in vehicles. Load hypothesis is analysed
from the functional and passive safety perspectives. Airbags and steering adaptations have an
142
important impact on the passive safety level of a vehicle and these factors are analysed in this
article.
2. Objectives
The work begins with a series of study stages in which the ultimate goal is to recover the
materials used when these vehicle parts reach the end of their useful life. To analyse the
behaviour of these adaptations in normal service, as well as in the event of an accident, the
various loads to which these adaptations are subjected must be studied.
This study will verify the levels of force to which these adaptations are subjected in normal use.
This will give rise to the establishment of several loading hypotheses which will facilitate the
design of these adaptations from a functional perspective – the applied loads being those
considered worst from the functional point of view.
Moreover, during the redesign of these adaptations those loads which are caused by passive
safety devices fitted on the vehicle should be specified. In addition to verifying how these
adaptations are affected, it is essential to assess whether these adaptations reduce the passive
safety of the vehicle. A growing number of devices are being fitted to prevent injury in the event
of an accident. Therefore, the assembly of these devices must be examined to see if they affect
passive safety. Steering wheel airbags are a clear example. These devices can subject
adaptations to high loads that can cause their deformation or breakage. In addition, the safety
function of an airbag may be affected by these adaptations if they intrude into the passenger
safety cell.
3. Methodology
The process used to evaluate the adaptations and assess the need for a redesign is a finite
element software simulation of the dynamics of the model of the accident (Morgan et al., 1998).
To achieve this, analytical models are available which have been calibrated and homologated
for impacts, impact barriers, as well as impacts on dummies that simulate the vehicle
occupants, etc. An airbag is one of the most important elements of passive safety. There are
many airbag systems in use, but one of the most important airbags is the unit that prevents
frontal impact. The simulation of the deployment of this airbag is a major factor in the evaluation
of adaptations (Figure 1).
Figure 1: Modelling an explosive airbag opening.
143
The methodology used to generate this simulation consisted of an initial phase in which a threedimensional model of the driving position was developed. A commercial 3D software design
package was used for this task. Once this task was accomplished, a finite element model was
imported and a finite element mesh generated (see Figure 2). Finally, the main constraints of
the model were specified in the dynamic simulation software – such as the materials and their
characteristics, contact interfaces, gas expansion models, technique of folding the airbag, etc.
Figure 2: Model of a steering wheel with a double knob.
Geometric parameters from commercial vehicles which take into account the minimum
distances in passenger safety cells were used for modelling the driver's position. The range
was set taking into consideration that these vehicles must accommodate drivers with differing
sizes and abilities. Other important factors include: the shape of the airbag, the steering wheel
angle of inclination, airbag folding technique, and rate of pressure inflation.
Two fundamental aspects in the deployment of the steering wheel airbag were evaluated: the
airbag folding technique, and the modelling of the thermodynamic gas expansion (Motevalli and
Bedewi, 1996).
The way in which an airbag is folded is one of the factors influencing its deployment; and the
folding technique also affects the time/pressure curve of the airbag. Figure 3 shows three of the
folding techniques used for the simulations. These include the use of fine pleats, internal folds,
and a spiral fold.
144
Figure 3: Airbag folding techniques.
Differing models representing various laws of behaviour were used to model the expansion of
the gases. The most widely used are simple, adiabatic, and hybrid.
Only the interaction of the airbag with the adaptation device was analysed. The driving position
was the nominal or standard driving position, and the distance of the driver from the wheel, and
therefore the front airbag, was set at the most appropriate distance for the correct functioning of
the airbag. However, it is often considered that a disabled driver using an adapted device sits
closer to the steering wheel than is normal.
Firstly, a simulation was made of a conventional airbag deployment on a steering wheel without
any fitted adaptation. This enabled the distances occupied by an expanding airbag to be
measured, as well as with the timings. Figure 4 shows a simulated steering wheel airbag
deployment sequence made using dynamic finite element software.
Figure 4: Modelling an airbag opening.
The central flat surface of the steering wheel is taken as a reference, and when the airbag
inflates the distance is measured from the halfway line or the outside edge of the airbag. This
145
measurement serves as a benchmark for subsequent analysis when the various devices being
tested are fitted to the steering wheel.
If no adaptation is fitted, then the outer edge of the airbag is parallel to the steering wheel's
central surface and the distance is the same in every position. Once adaptations are fitted, the
minimum and maximum distances are measured with respect to the central surface of the
steering wheel, in order to determine by how much the airbag has changed position.
Figure 5: The studied adapted steering controls.
We carried out the simulation using four different types of steering control adaptations: a simple
knob; a remote control knob; a forked knob, and a double knob. These adaptations are typical
of the variety available on the market, and according to a previous study on the influence of
such devices on passive safety, these types of adaptations are among those most likely to
interfere with the proper functioning of the frontal air bag. This interference is caused because
they are clamped to the steering wheel itself and intrude on the interior space of the wheel. As
shown in Figure 3, these devices can be placed in any position on the steering wheel according
to the ability and comfort of the driver.
The simple knob is relatively smaller than the other devices and also interferes less. This knob
enables the steering wheel to be turned with one hand. The remote control knob is larger,
although it does not apparently create more difficulties than the simple knob. The remote
control enables the lights, indicators, window-wipers, etc., to be activated without taking the
hands from the wheel.
The forked knob is larger than the others and its design may affect the operation of the airbag.
It is used in cases of tetraplegia. Finally, the shape and size of the double knob would seem to
suggest that it could be the design most likely to interfere with the deployment of the airbag.
This design is also used for cases of tetraplegia, where the drivers are totally disabled in the
hand and wrist (Dalrymple, 1996).
4. Results
The pressure/time curves have been obtained from the interior of the bags (Figure 6) in order to
assess the characteristics of the airbags and how they inflate. The figure below illustrates how
the pressure develops over time – from detonation to the bag being filled with air. Depending
on the model used, this representation enables graphics to be produced that reveal how the
bags fill and expand under pressure. A clear indication is offered by the shape of the curve
during the catapult stage. This air filling process is crucial to avoid injury following an impact
(Werner, Roberson, Ferguson, Digges and Day, 1996).
146
Pressure in N/m2
Airbag filling
Catapult phase
Detonation phase
Time in milliseconds
Figure 6: Pressure/time curve of the airbag.
Figure 7 shows one of the developed models. Representative models of airbags in action are
produced by combining various folding techniques with the expansion models.
Figure 7: Simulation of an airbag deployment.
Table 1 shows the average distances values obtained according to the various fitting positions.
147
Type of adaptation
Knob
Remote control knob
Forked knob
Double knob
Reference surface on steering wheel
Minimum
Maximum
Timings (ms)
distance (mm)
distance (mm) Dist. min
Dist.
max
10
33
8
14
16
41
9
22
7
122
7
26
3
112
6
24
Table 1: Maximum and minimum distances are shown from the reference surface on the steering wheel to the
outside edge of the airbag. Timings are also shown.
5. Conclusions
We have studied various combinations of expansion and deployment of airbags and this has
enabled us to obtain the time/pressure curve produced in the interior of the airbag. This curve
offers a clear indication of the shape of the explosive opening of the airbag. Reliable models
have been obtained for comparison with the actual results – and so enable an analysis of the
level of passive safety available to physically handicapped individuals in the driving position.
The study suggests that the interference of the control adaptations in the area of airbag
deployment significantly affects the positioning of the airbag following a collision. This means
fitting such adaptations causes a notable decrease in the passive safety of disabled drivers.
The forked knob and double knob are the adaptations that most intrude into the space occupied
by a deployed airbag and so produce the worst positioning of the airbag.
References
Dalrymple G., 1996. Effects of assistive steering devices on airbag deployment, National
Highway Traffic Safety Administration, SAE International Congress and Exposition. SAE paper
960223, 1996.
Morgan, R.M., et al., 1998. The Effect of Redesigned Air Bags on Frontal USA NCAP, 16th
ESV Conference, Windsor, Canada, June 1998.
Motevalli, V. and Bedewi, N.E., 1996. Airbag Modeling, © GWU/NCAC. 1996.
Werner, J., Roberson, S., Ferguson, S., Digges, K., Day, A., 1996. Air Bag Deployment
Frequency and Injury Risks, SAE # 960664. February.
Acknowledgements
We would like to thank the R&D&I Linguistic Assistance Office, Universidad Politécnica de
Valencia (Spain), for translating this paper.
148
Jaime Masiá Vañó
Instituto de Diseño y Fabricación
Universidad Politécnica de Valencia. Campus de Alcoy
Plaza Ferrándiz y Carbonell, s/n. 03801 Alcoy (Alicante)
Telephone: 96 652 84 77
Email: [email protected]
149
CYLINDRICAL PARALLEL GEAR TRAINS DESIGN BY MEANS OF
GENETIC OPTIMIZATION
Sánchez Caballero, V.
Colomer Romero, R.
Pla Ferrando, A.V.
Martínez Sanz, S.
Abstract
In this paper a genetic algorithm (GA)-based optimization procedure for the design of gear
transmissions is presented. For gear design, simultaneous discrete (p.e. pitch) and continuous
variables nonlinear related are used. However, most optimization methods are suited for
continuous design variables. The approach presented uses Gas as a tool to achieve not only
the optimal design, but also near-optimal designs. First, the optimization problem is formulated.
It must be multiobjective (maximum strength, minimum energetic losses, etc) and restricted. A
mechanism to transform the constrained problem into unconstrained thought penalty functions
is proposed. Recommendations on the objective function and penalty terms are proposed. Next
a design variables coding and decoding method, as well the genetic operators of reproduction,
crossover and mutation are presented. Finally, it is analyzed an example in which the
developed genetic algorithm has been used, comparing the obtained results from a previous
optimization.
Keywords: Genetic Algorithm, optimization, transmission, gear
1. Introduction
During last thirty years, lots of methods to solve specific mechanical optimization problems
have been developed (Gallager et al, 1973; Hillier et al, 1990). However, a fully efficient robust
and simple method applicable to the full range of optimization problems have not been found
(Rajeev et al, 1993).
Gear transmission problems represent one of the most complex optimization problems that can
be found, not only for the range of the optimization objectives, but also the relationship between
different variables which take part in this process.
Variables involved in the design process can be discrete (p.e. modulus, number of teeth, etc),
or continuous (p.e. helix angle, width, teeth displacement, etc). However, most of traditional
optimization methods are only suitable for continuous variables.
Getting maximum and minimum in an optimization problem where discrete variables are
involved, is much more complicated and difficult than when continuous take part. Only a few
methods (Arora, 2002) are suitable to fulfill this kind of optimization, among them full
enumeration techniques, integer programming, branch and bound algorithms and dynamic
programming can be stand.
150
When in addition continuous and discrete variables are mixed and non linear related, the
spectrum of methods are even more reduced.
During last fifteen years, it has been produced a growing interest in optimization methods
based in which has been called soft computing, among them can be outstand evolutionary
algorithms and simulated annealing.
From this group of techniques, not only it has not been found any of them which have been
proved more profitable in any case, but each problem has one or several suitable techniques.
Existing optimization methods in gear design (A. Daidie et al, 1993) search for a general
optimum for a general mechanics applications, without allowing other kind of optimization like
cost or specific power minimization, efficiency maximization, etc…
2. Objectives
This work intends to develop a general technique which can be used in any case and can be
adapted to specific optimization conditions in any case.
From these techniques, previously described, evolutionary algorithms were chose. This group
of techniques is also classified in: evolutionary programming, evolutionary strategies and
genetic algorithms.
From this group of techniques, genetic algorithms were chose because of its implementation
easiness and because, in a way to certain extent, all other can be developed in the future from
this one.
Must be considered, as has been told, that it is not possible to foresee which technique is the
best suitable beforehand. Only employing all of them will be possible to state which of them is
the most suitable for this case.
3. Genetic Algorithms description
Genetic algorithms are based on the strategy of model development based on genetic evolution
mechanisms based on Darwinian theory (David et al, 1988). The main characteristics of genetic
algorithms (GAs) are based on best fit survival and environment adaptation.
The starting point is an random generated population, which design variables are coded
following the genetic laws. Each design variable is represented by a gene or genotype, which is
normally coded in binary code. This binary representation is called phenotype. All the
genotypes (or phenotypes) form an individual which is named chromosome.
Next the algorithm generates, following several rules, new populations formed by best adapted
to environment individuals. In other words, those individuals which best fit the objective function
are selected.
Basic rules which rule new population generation are: reproduction, crossover and mutation.
This is achieved through the implementation of genetic operators of reproduction, crossover
and mutation, whose nature is mainly stochastic.
Main advantage of this method in front of traditional ones settles in the fact that it explores
simultaneously all the solution domain. In any case, convergence to a maximum is assured. It
might not be the absolute maximum, but it has more probabilities to obtain an absolute
maximum than traditional criterion based on gradient. This is because of this method allows
151
some individuals of the population to jump in the optimum search in spite of the most ones are
focused on the search of the same optimum. This jump is generated by the mutation operator.
Convergence speed, and the global mínimum/máximum depends mainly on the parameters
and the kind of the three genetic operators.
4. Gear optimization problem formulation
Optimum teeth geometry determination is a complex operation. The parameters quantity which
takes part of the calculation formulation is significant. Gear calculation methods are also
numerous: AGMA, ISO, DIN, VDI, AFNOR, Henriot, Niemann, etc…
It is intended to develop an automatic cylindrical gears sizing method, which gets an optimum
result with minimum designer effort.
In near future it is intended to implement this algorithm in a gear design software.
4.1 Parameters
Parameters in standard optimization problems are: Transmissible power Ptrans, reduction ratio
i=n2/n1 (where n1 represents input speed and n2 output speed) and endurance N. Reliability,
surface finish, input speed ni and safety coefficient n are also needed.
4.2 Design variables
Main design variables are: width b, gear teeth number z1 and z2, teeth displacement
coefficients x1 y x2, an normal modulus (cutting tool) m0, helix angle β, and gear materials
MAT1 y MAT2.
Only teeth displacement and helix angle are continuous variables, while all other variables are
discrete.
4.3 Calculation Formulae
Details about gear calculation theory can be found in standards ISO 6336 Method B and
Henriot’s book (Henriot, 1979) Graph Trend equations corresponding to the calculation method
have been obtained.
gs1 and gs2 are sliding speeds in first and last contact point. They are obtained from the
peripheral gear speeds u1 and u2. These speeds are obtained through turning gear speed and
curvature radius in the considered contact point:
ui = ωi ⋅ ri
(1)
Sliding speed is the difference between both peripheral speeds::
152
vg = u2 − u1 = ω2 ⋅ r2 − ω1 ⋅ r1
(2)
Pinion specific sliding is the ratio between specific speed and the contact point speed:
g s1 =
u2 − u1 u2
ω ⋅r
r
1
= −1 = 2 2 −1 = i ⋅ 2 −1 = i ⋅
−1
α'
'
a
⋅
sen
ω1 ⋅ r1
u1
u1
r1
−1
ra22 + rb22
gs2 =
u1 − u2 u1
ω ⋅r
1 ⎡ a '⋅ senα ' ⎤
= −1 = 1 1 −1 = ⋅ ⎢
− 1⎥ − 1
u2
u2
ω2 ⋅ r2
i ⎢ ra21 + rb21 ⎥
⎣
⎦
(3)
(4)
Where a’ is the working center distance, α’ is the working contact angle and rai ,rbi are tip and
base radius.
Specific sliding is wear responsible, so in the optimization process both gear specific sliding
must be balanced in order to achieve an equal wear.
Strength calculation considers the possibility of flank surface and foot teeth failure.
Surface failure is prevented by Hertz theory implementation applied to two cylinder surface
contact, whose radii are the instantaneous flank curvature radii. Maximum transmissible power
in order to prevent surface failure Ppres is obtained by comparison between the teeth flank
contact stress and the maximum contact material strength.
Foot teeth failure is prevented by the teeth flexure calculation. Maximum transmissible power in
order to prevent foot teeth failure Prup is obtained by comparison between foot flexure stress
and the maximum admissible material strength.
It must be checked that both transmissible powers are higher than the desired one, with an
adequate security factor. Likewise, both transmissible powers must be as equated as possible.
4.4 Particular goals of the optimization problem
The optimization problem requires a series of particular goals as well as the general goals
previously enunciated.
These particular goals are:
1. To equate, with certain tolerance range, specific sliding for both gears. This factor has a
direct influence on wear and efficiency.
2. To equate transmissible powers Ppres, Prup according to a failure criterion.
3. To insure a minimum transversal contact radius.
4. To restrict maximum peripheral speed.
5. To insure that interference does not take place during the assembly.
6. To insure that the width/diameter ratio of the gears is within a possible range.
4.5 Problem algebraic formulation
Based on previous description, the problem can be algebraically formulated:
153
−
To minimize the gear weight:
F = f ( b, z1 , z2 , m0 ) = b ⋅ ⎡⎣ m02 ⋅ ( z12 + z22 ) − ( d i21 + d i22 ) ⎤⎦ = W
(5)
Where di1 and di2 are the hole diameter of the gears
Subject to:
G1 ( MAT , b, m0 , z , ω , K ....) = Prup − n ⋅ Ptrans ≤ 0
(6)
G2 ( MAT , b, m0 , z , ω , K ....) = Ppres − n ⋅ Ptrans ≤ 0
(7)
G3 ( MAT , b, m0 , ω , z , K ....) = Prup − Ppres ≤ 0
(8)
G4 ( g s1 , g s 2 ) = g s1 − g s 2 ≤ 0
(9)
G5 ( a, amin ) = amin − a ≤ 0
(10)
G6 ( z1 , z1min ) = z1min − z1 ≤ 0
(11)
G7 ( z1 , z2 , v ) =
v ⋅ z1
100
G8 ( b, m0 , z1 ) =
z22
− 10 ≤ 0
z12 + z22
b
− 2, 5 ≤ 0
m0 ⋅ z1
G9 ( m0 ,... ) = san − 0, 2 ⋅ m0 ≤ 0
G10 ( z1 , z2 , i ) =
z1
−i ≤ 0
z2
z1 ∧ z2 ∈ impares
(12)
(13)
(14)
(15)
(16)
bL ≤ b ≤ bH ∈ z1L ≤ z1 ≤ z1U ∈ z2L ≤ z2 ≤ zU2 ∈ (17)
x1L ≤ x1 ≤ x1U ∈ x2L ≤ x2 ≤ xU2 ∈ Where G1, G2 y G3 represent the break and surface contact strength condition, as well as the
equality condition between them. G4 represents the specific sliding equally condition between
gears. G5 shows the non interference condition between gears, assuring the minimum center
distance between gears. G6 assures that the teeth number of the pinion is higher than the
minimum needed, in order to avoid that undermine is produced by the cutting tool. G7
represents the gear peripheral speed limitation. G8 prevents against an excessive gear width.
G9 prevents against an excessive gear sharp tip. G10 forces the relation between the number of
teeth and the reduction ratio.
154
Constraint (16) forces one gear to have an odd teeth number.
Remainder constraints (17) settle the nature and limits of different variables.
As can be appreciated, the design variables appear in the problem formulation as explicit form
but also implicit.
5. Genetic algorithm formulation for problem solving
Solving optimization problems by means of genetic algorithms does not generate only one
algorithm to solve a specific problem. Reproduction, crossover an mutation operators are not
the only ones, but also exists several ways to implement them.
Likewise, the way of treat the fitness or penalty function or the constraints is not also solely.
Several approaches could be taken: static penalty functions, dynamic, adaptative,
coevolutionary, sudden death, etc. …
This is way different algorithms are obtained, whose convergence speed and capacity to obtain
the global optimum change between them.
Determination of the optimum algorithm to solve a specific kind of problem requires to carry out
several essays with different types of operators. This benchmarking is done by means of “class
problems” which represent the class of problem in question.
5.1 Design variables coding
In order to make easier the coding process, and speedup the calculations, the variables which
take part on the gear design process are not coded altogether. From the whole variables range,
the previously exposed variables are selected. It would not be excessively complicated to add
more variables to the whole, however some variables as the working parameters described are
used to be determinate a priori by the designer in the optimization process.
This way, the genotype (or chromosome) of any individual of the population would be ever
settled by the following vector:
(18)
{b z1 z2 m0 x1 x2 β MAT 1 MAT 2}
Binary coding of the variables in binary form is made in a different way depending if the variable
is continuous or discrete.
The bit chain length for a discrete variable is calculated using this equation:
2n = λ
(19)
Where n is the coding needed number of bits and λ is the number of possible values that can
take the variable.
In this way, if the number of standardized modulii of the preferential class is of 34, 6 bits will be
needed to code this variable. The total number of values that could be coded would be of 26=
64 values. Provided that the number of values needed in the list of modulii is about 34, it would
be 30 “gaps” in the list. In order to avoid this disadvantage, its usual to fill the gaps repeating
the first 30 elements of the list either are filled with 30 list random elements.
Bit chain length for a continuous variable is obtained by means of the Hajela’s equation (Hajela,
1992):
155
2n ≥
XU − X L
AC
(20)
Where XU y XL are the upper and lower limits of the continuous variable X and AC is the
required precision in the discretization progress of the continuous variable.
This way if discretizing the gear displacement variable x is desired, with a lower and an upper
range between -0,5 y 0,5 with at last 0,025 precision it would be needed:
2n =
0,5 − ( −0,5 )
1
=
= 40
0, 025
0, 025
(21)
Therefore, 6 bits would be needed to be able to code this parameter with the desired precision.
The number of discretizing number of divisions would be of 64, so the step between values
would be of:
δ=
0,5 − ( −0,5 )
= 0,015625
64
(22)
As can be seen in the solution, the required discretization resolution is higher than the initial
required one.
Following the described method, all variables are coded with following resolutions: width 4 bits,
number of teeth 6 bits, modulus 6 bits, tooth displacement 6 bits, helix angle 4 bits and material
3 bits.
When coding the variables in binary form it would be obtained the phenotype, above which
genetic operators would be applied:
GENOTYPE {b
z1
z2
m0
x1
x2
β
MAT 1 MAT 2}
(23)
PHENOTYPE {1001 110011 101111 100101 101111 101111 1101 011 101} (24)
5.2 Penalty / Fitness function
Fitness of the individuals in consecutive generations is measured by means of the penalty
function. There are several differences between de previously stated penalty function (5) and
the one used to assess the individual fitness.
Penalty function is used in order to convert a constrained optimization problem (or
multicriterion) into a non constrained problem. This is achieved by introducing the constraints
into the initial fitness function, so if any constraint is violated, a penalty proportionally to the
violation grade is added.
This modified fitness function is called penalty function.
There are several approaches when building the penalty function: statics, dynamics,
adaptatives, coevolutionaries, etc. … The one used in this optimization was a quadratic static
function. Penalty function formulation was made in this way:
m
F = W ⋅ ∏ Φi
(25)
i =1
Where Φi is the penalty value for constraint Gi, and W is the total weight.
Calculation of the penalty function values was carried out by means of the following quadratic
expression:
156
Φ i = 1 + ri ⋅ ( qi − 1)
(26)
Where ri is the “i” constraint quadratic penalty foreground and measures the importance given
to the constraint violation.
qi is defined as:
pi
⎧
si
≤1
⎪1
( pi )max
⎪
qi = ⎨
pi
⎪ pi
si
>1
⎪ ( pi )
p
(
)
i max
⎩
max
(27)
Where pi is the constraint parameter and (pmax)i is the limit value for that one.
5.3 Genetic operators
Reproduction or selection operator:
There are several ways to implement this genetic operator: proportional reproduction; ranking
group or tournament selection, etc. …
Group selection has been used in this study, in which individuals are sorted according to its
fitness form best to worse.
Next, population is divided into two groups, setting a selection probability to each group
Crossover operator:
Like all other operators, there are several models for crossover operators: fixed, flexible,
uniform, adaptative, etc. …
In this study one point chromosome cut by genotype has been chosen. This is one of the
simplest and is form by cutting parents chromosome through selecting the cut point between
phenotypes in the bit chain P.e.:
FATHER {1001 110011 101111 100101 101111 101111 1101 011 101}
(28)
MOTHER {1011 100011 100111 111101 101001 101001 1001 001 111}
(29)
CUT
CHILD1 {1001 110011 101111 111101 101001 101001 1001 001 111}
(30)
CHILD2 {1011 100011 100111 100101 101111 101111 1101 011 101}
(31)
Mutation operator:
This operator allows exploring other areas from the solution domain, yet when most of solutions
tend to an optimum. This is what allows jumping from a local optimum to another one where le
convergence progress became poor due to a local optimum focus.
For its implementation, not may individuals are chosen afterwards the selection an crossover
processes. The ratio of selected individuals is determined by the mutation probability which
usually is not higher than 1% of the population. Typical value is 0,1%.
157
Once individuals have been selected, one or more aleles (chain bits) are chosen and then their
values are swept from 0 to 1 (or vice versa). P.e.
NATURAL CHILD {1011 100011 100111 100101 101111 101111 1101 011 101} (32)
MUTATED CHILD {1011 100011 101111 000101 101111 101101 1101 011 101} (33)
The evolution mechanism is repeated trough several generations until no increase on individual
fitness is produced. The highest fitness member will be the optimum design.
6. Implementation example
This test analyses a typical gear optimization problem from the bibliography (Marcelin, 2001). A
gear pair optimization is planned in order to accomplish the following design specifications:
Power to transmit: 73 kW.
Input angular speed: n1 = 2800 rpm
Reduction ratio: i = ½
Manufacturing quality: ISO 6
Endurance: 3650 h. Normal work, without impacts.
Pinion material: 34CrMo4 hardened (34CD4 AFNOR, F 125 UNE).
Gear material: 20MnCr5 cemented (16NC6 AFNOR, F 158 UNE)
Following table shows reference values, a previous optimization obtained by other author and
the one obtained by the described method.
In order to check the optimization method improvement, pinion and gear material have been
fixed in the analysis. Strength and fatigue security coefficients have been defined as reference
values.
158
Variable
Reference
Solution
Initial optimization
Improved
optimization
Modulus (mm): m0
3
2.5
2.5
Width (mm): b
27
20
45
Pinion number of teeth: z1
24
29
17
Gear number of teeth: z2
48
116
38
Pinion displacement: x1
0.16
0.3
0.2273
Gear displacement: x2
-0.16
0.008
-0.0038
0
0
Helix angle: β
36º
Pinion foot strength security
coefficient
1.4170
0.9559
1.4633
Gear foot strength security
coefficient
2.1737
1.3386
2.1090
Pinion foot endurance security
coefficient
0.5673
0.5618
Gear foot endurance security
coefficient
1.1567
Weight (kg)
4.31
0.5763
1.1179
1.1101
3.24
2.78
Table 1: Result summary.
As can be appreciated, weight reduction is about 35% with regard to initial solution and about
14% to the previously suggested one. Security coefficients are more balanced and no strength
variation with respect to original solution has been produced, while is something higher to the
previously suggested one.
Putting the material as a variable, including a penalty term which represents the economical
cost, would allow optimizing the problem even more.
7. Conclusions
From the initial design and after 240 generations, an optimum solutions range is found. Values
shown in table 1 represent the best solution, according to the obtained penalty function.
Described method is a really tool to help designer, who finally selects the optimal solution
between a range of optimum results. It must be taken into account that the constraint altogether
does not represent 100% of possible limitations or design criterions. Is then designer
responsibility to make the fine tuning by selecting between possible results the best suitable
that he considers.
159
Designer can also adapt the algorithm to an specific problem by simply adding more variables
to the problem, by coding them according to the described method.
In the future is expected to implement the calculus and the genetic algorithm into a software
which even would allow to compare optimum results form different calculation methods
References
Arora, J., 2002. Methods for discrete variable structural optimization, in Recent Advances in
Optimal Structural Design, S.A. Burns, Editor, ASCE. p. 1-40.
Daidie A., e.a., 1993. Dimmensionnement optimal d'un train d'engrenages à l'aide d'un logiciel
CAO. in PRIMECA. Ecole Centrale de Paris.
David, E.G. and H.H. John, 1988. Genetic Algorithms and Machine Learning. Mach. Learn.,
3(2-3): p. 95-99.
Gallagher, R.H.a.Z., O.C., 1973. Optimum Structural Design: Theory and Applicacions: John
Willey and Sons.
Hajela, P., 1992. Stochastic Search in Structural Optimization: Genetic Algorithms and
Simulated Annealing. p. 611-635.
Henriot, G., 1979. Traité Théorique et Practique des Engrenages., París: Dunod.
Hillier, F.S.a.L., G.J., 1990. Introduction to Mathematical Programming: McGraw-Hill.
Marcelin, J.L., 2001. Genetic Optimisation of Gears. Advanced Manufacturing Tecnology.
Rajeev, S.a.K., C.S., 1992. Discrete Optimization of Structures Using Genetic Algorithms.
ASCE Journal of Structural Engineering. 118(5): p. 1233-1250.
Acknowledgements
To Notre Dame University professor Mihir Sen, for rediscovering me the genetic algorithms.
Polytechnic School of Alcoy
Polytechnic University of Valencia
Department of Mechanical Engineering
C/ Plaza Ferrándiz i Carbonell 2, 03801 Alcoy (España)
Phone: +34 966 52 85 75
Fax: + 34 966 52 84 09
160
AN ONTOLOGY-BASED APPROACH TO INTEGRATING LIFE CYCLE
ANALYSIS AND COMPUTER AIDED DESIGN
Abad-Kelly, J.
Cebrián, D.
Chulvi, V.
University Jaume I
Abstract
One of the principal problems faced by engineering design today is the exchange of product
information across different applications and environments. Ontological engineering systems,
an evolution of KBE (Knowledge-Based Engineering) systems, seek to facilitate this integration
while incorporating additional design information. An ontology, in the engineering domain, can
be defined as an explicit specification of a shared conceptualization. This paper proposes the
integration of an ontology with a Computer Aided Design (CAD) program, while also accessing
a database of information on environmental impact. The proposed ontology is based on the
AsD (Assembly Design) formalism, which describes spatial relationships and features of CAD
models. The use of OWL (Web Ontology Language) and SWRL (Semantic Web Rule
Language) ensures machine interpretability and exchange across different environments.
Ultimately, the ontology will be used to represent a CAD model and related information (such as
joining methods, materials, tolerances) in formal terms. Concurrently, a database of information
on environmental impact of the materials, processes and transport involved will be accessed to
evaluate the model on an environmental level. As a practical illustration, the evaluation of an
underwater camera is used as an example.
Keywords: ontology, Knowledge Based Engineering, environmental, Computer Aided Design
Introduction
Effective collaboration and information sharing is essential across the various stages of product
development. The increasing use of Computer Aided Design (CAD) systems serves as a great
aid to this process, however, communication between programs is often difficult if not
impossible. This is mainly due the use of proprietary formats and a lack of true integration
across different platforms. A parallel issue in the design field is the growing importance given to
environmental concerns and impacts generated by products. Both private initiatives and
government regulations are placing a great emphasis on reducing contamination and the use of
resources.
This paper proposes a method that, with the aid of an ontology, communicates a threedimensional CAD model of a product with information related to environmental impact data. In
161
the first section, a general background is given of the terms and concepts that the paper is
based on. The following section provides a detailed explanation of the proposed method and
each of its constituents. In section 4, the results of a practical application of the method itself
are given, followed by a section of discussions and conclusions.
1. Background
1.1. KBE Systems and ontologies
The nature of product design as an inherently collaborative process and the increased
dependency on computer systems have placed great emphasis on the importance of effective
information sharing between designers. As a solution to this need, Knowledge Based
Engineering systems have been created in order to correctly capture, represent and structure
product-related information. These systems seek to facilitate the exchange and reuse of design
know-how, going beyond the limits of a CAD system to include design intent, objectives and
specifications, in addition to geometric models [Chapman & Pinfold, 1999]. However, many of
these systems lack the ability to efficiently utilize the knowledge itself, and are rarely more than
just elaborate information bases. As the successor to knowledge engineering, Ontological
engineering seeks to overcome the aforementioned deficiencies by providing methodologies
that allow for more effective knowledge use [Kitamura & Mizoguchi, 2004].
The definition of an ontology can vary according to the field in which it is applied. Within the
engineering and knowledge-sharing disciplines, an ontology can be described as a formal,
explicit specification of a conceptualization shared by an agent or group of agents [Mizoguchi,
2003]. That is, an ontology facilitates knowledge sharing and reuse by establishing an explicit
agreement of objects and relationships within a specific domain of discourse [Gruber, 1993].
The use of the term formal implies that the ontology is machine-readable, but without being too
cryptic for human understanding [Mizoguchi, 2003]. By establishing a common language for
use and interpretation of information, ontologies allow for efficient and organized exchange and
reuse of knowledge.
Within the engineering design domain, one can distinguish between task and domain
ontologies. The first refers to ontologies related to design activities themselves, while the latter
deal with the targets of the design process, whether products or manufacturing processes
[Kitamura & Mizoguchi, 2004].
1.2. Computer Aided Design
From the point of view of knowledge engineering, Computer Aided Design systems provide a
basis for representing and sharing a large amount of information related to a products and their
components. Within a three-dimensional model, data such as geometry, dimensions, relations,
structure and other information can be stored. Current CAD programs are being developed to
increasingly include more product-related data, such as CATIA PLM Express or DriveWorks for
SolidWorks. These modules allow for product models to contain additional information such as
configurations, rules, analysis and cost. However, no true integration of knowledge engineering
and CAD has yet been achieved [Skarka, 2007]. The use of a CAD model as the core of a
knowledge management system allows abstract information to be directly related to the product
itself, making use of the system much more visual and effective.
162
1.3. Environmental impact assessment
Another area that is currently of great interest within the engineering design community, and the
world at large, is the environmental impact created by products and processes. Environmental
criteria are being increasingly integrated into the design process at earlier phases, along with
the development of strategies for end of life management. Taking these criteria into
consideration at earlier stages in the design process reduces the need to make costly changes
at a later stage. While some of these changes can refer to material use, processes and
transport, others may relate to the geometry of the product itself.
One method for evaluating the environmental impact of a product is Life Cycle Assessment
(LCA) which assesses the impact generated by the product throughout the various stages of its
life cycle [Lewis H. et al, 2001]. This process consists of four main stages: definition of goal and
scope, inventory analysis, impact assessment and interpretation [Curran, 2006]. In order to
carry out an LCA it is necessary to define, among other things, the material and process flows
that occur during the life cycle of the product. For example, for a product that can be broken
down into components, the amount of material used for each component and the process used
to manufacture it must be defined. The impact is then calculated for each step of the product’s
life: obtaining the raw material, intermediate and final processes, disposal, and so on.
Several existing software tools, which incorporate large material and process databases, can
be used to facilitate carrying out an LCA [Lewis H. et al, 2001]. Even though many assumptions
and estimations are made, these tools help give a general idea of the main sources of impact
generated by a product and at which stage they occur. An LCA software tool of extended use is
SimaPro, developed by PRé Consultants. SimaPro contains a variety of databases and
methods for impact assessment, such as Eco-indicator 95 and EPS 2000.
At the present time, several tools exist to directly evaluate a product design from an
environmental perspective using different methods, among which is the LCA. These tools
include Green Design Advisor [Feldmann, K. et al., 1999], ECoDe [Lye S.W. et al., 2002],
LASeR [Ishii, 1994], DFD/E [Dewhurst and Boothroyd, 1992], and RecyKon [Meerkamm, 1995].
These analytical tools exist as stand-alone programs. That is, much like most of the software
used by product designers, they only deal with the evaluation itself and do not offer integration
with other environments or programs. Not only is it necessary to develop an automatic and
efficient way to evaluate a product from an environmental perspective, it is fundamental to be
able to integrate this method with other areas of product design.
2. Methods
2.1. CAD model
The first step of the proposed method consists of creating a three-dimensional model of each
part of the product that will be evaluated. It is fundamental to create models that precisely
represent each component, given that they will later be used for evaluation. For this step, the
authors have used SolidWorks 3D CAD software. Once each part is modeled, sub-assemblies
and an assembly of the entire product are created. Within each part file, a material is specified
and its properties (such as density, conductivity, Poisson’s ratio, etc.) are applied to the part. In
SolidWorks, materials can be chosen either from the standard database or a user-defined set of
materials. Upon completion of this stage, an accurate representation of the product geometry
and material is obtained.
163
2.2. Environmental impact assessment data
As previously mentioned, SimaPro software is a tool to assist in carrying out Life Cycle
Analysis. For this work, the Eco-Indicator 95, a versatile and comprehensive method, was
applied. Eco-95, developed by PRé Consultants for the Dutch government, groups impacts into
three categories: human health, ecosystem health and resources [Lewis H. et al, 2001]. Single
score results are given in Eco-Indicator points (Pt) where 1Pt is representative of one
thousandth of the yearly environmental load of one average European inhabitant [Goedkoop,
1996]. In order to analyze the impact of a material or process, an amount must be specified,
such as mass, surface area or length. At this phase of the method, only an analysis based on
mass is carried out. For each unit of mass, the material required and the relevant
manufacturing process is analyzed. In order to do so, a separate impact assessment of one
kilogram of each of the materials specified in SolidWorks was completed. Single Score results
are used, in order to facilitate comparison between different parts and materials. As a result, the
outcome of the LCA of each material is exported to an Excel spreadsheet.
2.3. Impact evaluation macro
At the present time, the link established between the CAD model and the results of the LCA is
carried out by a macro developed in Microsoft Excel. This software was chosen because of the
possibility of creating a macro that could easily be executed by a user with little or no
knowledge of either CAD or LCA. The macro was written in Visual Basic, a language supported
by the SolidWorks Application Programming Interface (API). In the Excel file that contains the
macro are two buttons that execute different commands when selected by the user. The button
labeled “Get Component Properties” lists the components of the CAD assembly model, the
material assigned to each one and its mass in kilograms. If no mass property is available for a
component, the returned value is “-”. The command button labeled “Evaluate Component
Impact” calculates the environmental impact of each component and returns a numeric value in
Pt. In the case of a sub-assembly within an assembly, a nil value is returned. This calculation is
based on the LCA data previously exported from SimaPro for each material. The result of this
simple and automated step is a concise evaluation of the impact generated by each
component, giving the designer a clear picture of the product’s environmental impact.
2.4. The ontology
An ontology consists of three main elements: classes, properties and individuals. The ontology
developed for this work is based on the program Protégé 3.2 and is represented using OWL
(Web Ontology Language). As part of the Semantic Web, the use of OWL as a common
vocabulary for ontology representation ensures data sharing and re-use across different
environments [Daconta M.C. et al., 2003; World Wide Web Consortium, 2004]. In order to
develop the OWL code in a more intuitive way, the aforementioned program is divided
according to the classification of an OWL ontology. Afterwards, the tools employed will also be
explained.
Classes:
OWL classes are interpreted as sets that contain individuals. They are described using formal
(mathematical) descriptions that state precisely the requirements for membership of the class.
The structure of the ontology, based on the Assembly Design (AsD) ontology [Kim K.Y, 2006],
consists of the following classes. Each class is defined by the indicated properties:
164
Structure: Elements with a spatial location and a degree of topological connection between
them. This class defines the physical elements that constitute a design.
Assembly: Elements that are a part of the structure of an object and consist of at least 2 parts.
Part: Elements that are part of the structure and may have a topological connection with other
parts in order to form an assembly.
Material: This class indicates the materials of which parts of a structure may be made. It is
essentially related to data properties, in order to completely define physical properties.
Topological Connectness: This class is based on the spatial relationships originally put
forward in by Ambler and Popplestone [Ambler and Popplestone, 1975]. It indicates the type of
connection that exists between two parts and is subdivided in the following classes:
•
Against: the mating surfaces touch at some point. This relationship is the most basic
spatial relationship and applies to any assembly. Any combination of two parts can
possess this property.
•
Aligned: two features are aligned if their centerlines are collinear.
•
IncludeAngle: the inclination relation holds for an include angle between two planar
faces in their positive normal direction. The rotation is clockwise with respect to a
normal of a picking face. The rotational axis has to be parallel to the normals of above
two planar faces.
•
ParallelOffset: the parallel relation holds between planar faces, cylindrical and spherical
features. In the case of two parallel planar faces, the outward normals are pointing in
the same direction. This relationship exists without physical contact of two features with
an offset distance.
•
ParaxOffset: this relationship is similar to parallel-offset but the outward normals of the
parallel planar faces are in opposite directions.
Properties:
Properties are binary relations on individuals - i.e. properties link two individuals together.
Object Properties:
These properties essentially exist in order to establish the necessary logical rules between
elements of a structure. The types are: has_Spatial_Location; isPartof; isAssembledBy; and
hasMaterial.
Datatype Properties:
The properties that are applicable to classes can be divided into the following two areas:
•
Physical properties, which are applied to the structure area. This information is obtained
from the macro that links SolidWorks to a spreadsheet. For example: density.
•
Environmental impact of the materials, applied to the information extracted from
SimaPro.
The Protégé interface contains a series of Tabs that allow for the ontology to be applied to
different areas. Below is a brief explanation of the Tabs that have been applied in order to
develop this work:
165
DataMaster Tab – DataMaster is a Protégé plug-in for importing schema structure and data
from relational databases into Protégé [The CO-ODE Project, 2007]. It allows for information
from the macro to be linked to the ontology.
TGViz Tab – TGViz Tab is a plug-in for Protégé which allows visualizing ontologies using the
TouchGraph library. TouchGraph provides a java library for rendering networks as interactive
graphs. This library has been modified and integrated with Protégé as a tab plug-in [Stanford
Protégé Wiki, 2007]. It allows for the different relations that exist between components of an
object to be defined.
OWLViz Tab – OWLViz is designed to be used with the Protégé OWL plug-in. It enables the
class hierarchies in an OWL Ontology to be viewed and incrementally navigated, allowing
comparison of the asserted class hierarchy and the inferred class hierarchy. OWLViz integrates
with the Protégé-OWL plug-in, using the same color scheme so that primitive and defined
classes can be distinguished, computed changes to the class hierarchy may be clearly seen,
and inconsistent concepts are highlighted in red. OWLViz has the facility to save both the
asserted and inferred views of the class hierarchy to various concrete graphics formats
including png, jpeg and svg [TouchGraph, 2005].
Individuals
Individuals represent objects in the domain that we are interested in. Individuals are also known
as instances. Individuals can be referred to as being ‘instances of classes’. In each of the
previously mentioned classes, the relevant instances are represented in order to organize the
available knowledge.
The ontology allows for the available knowledge to be organized according to the information
provided by the macro, and makes it possible to infer new knowledge and search the existing
one. This capability is provided for by the use of the Semantic Web Rule Language (SWRL). In
this way, the ontology achieves the integration of the structural information and the
environmental impact data.
166
Figure 1: The ontology.
3. Practical application and results
In order to illustrate the proposed method, the previously described steps have been applied to
an assembly of an underwater camera. This particular assembly consists of 9 parts
manufactured with different materials and processes, and held together by standard fasteners.
An image of the three-dimensional model of the assembly is shown below. A new database of
materials was created in SolidWorks, containing ABS, Spring Steel, Cast Stainless Steel and
Aluminum 6061 Alloy. Within each part file, a material was selected from this database.
167
Figure 2: CAD model of underwater camera assembly.
The next step is to carry out an LCA of each of the above materials (and their relevant
manufacturing processes) in SimaPro and export the results to an Excel spreadsheet. Below is
an example of the results of analyzing one kilogram of ABS with the Eco-Indicator 95 method.
Figure 3: Screenshot of LCA of ABS.
Once the CAD model and LCA are completed, the first part of the macro is executed in order to
obtain the mass properties for each component of the assembly. Next, the second command
button is selected, and the environmental impact of each component is obtained.
168
Figure 4: Excel spreadsheet of assembly.
The final result is a clearly visible analysis of the environmental impact generated by the
assembly. Due to the fact that each component is analyzed separately, the designer is able to
clearly identify which component has a greater impact in relation to its mass. In this case, it is
evident that the component “TV Camera” generates the greatest environmental impact. If
changes are made to the assembly, the macro is newly executed and any improvements in the
impact are instantly visible. Once all of the available information has been gathered, it is
incorporated into the ontology. As a result, the existing relations between the components are
specified in the terms used by the ontology, and the information related to each component is
displayed. The figure below is a part of the graphic representation of the ontology as applied to
the camera assembly (the entire image is too large). By structuring the available knowledge in
an ontology, it is now possible to carry out searches or even infer new knowledge.
Figure 5: Ontology representation of a sub-assembly of the camera.
169
4. Conclusions and discussion
Despite being an initial exploration, this work is an important step in a new direction that
explores the possibilities of integrating three-dimensional CAD models with Life Cycle Analysis
by means of an ontology. The growing importance of environmental criteria and the
competitiveness of product design today create the need for an effective way to incorporate
LCA into early design phases. Following a notable trend in engineering design, the use of an
ontology allows for the incorporation of additional knowledge and possible integration with other
programs. The developed ontology could be expanded to include cost analysis, manufacturing,
product functions and many of the other areas that product design encompasses. Rather than a
solution in itself, the research carried out demonstrates the feasibility of incorporating
environmental criteria into design without the use of specialized software. The fact that this
analysis is almost entirely automatic implies that tedious and manual analyses can be avoided,
and the effects of improvements on a design are instantly visible. Using a three dimensional
model as a central part of a knowledge management system gives users a much more visual
and effective way to interact with the system and the abstract data it contains.
In the future, the proposed method could be expanded to include a more complete
environmental analysis, such as alternative evaluation methods and data libraries. Impact could
also be calculated based on surface area and length as well as mass, depending on the
material and process chosen. Another possibility would be to develop the macro into an
independent User Interface, which would allow for more flexibility and the inclusion of additional
environmental criteria.
References
Ambler, A. P., and Popplestone, R. J., 1975. Inferring the Positions of Bodies from Specified
Spatial Relationships, Artificial Intelligence, Vol. 6, No. 2, pp. 157-174.
Chapman C.B. and Pinfold M., 1999. Design engineering - a need to rethink the solution using
knowledge based engineering, Knowledge-Based Systems, Vol. 12, pp. 257-267.
Curran M.A., 2006. Life cycle assessment: principles and practice, Environmental Protection
Agency, Cincinnati.
Daconta M.C. et al., 2003. The semantic web: A guide to the future of XML, web services, and
knowledge management, Wiley, Indianapolis.
Dewhurst, P. and Boothroyd, G., 1992. New software developments in design for assembly,
disassembly and service, International Forum on Design for Manufacture and Assembly,
Newport.
Feldmann, K. et al., 1999. The green design advisor: a tool for design for environment,
International Journal of Electronics Manufacturing, Volume 9, No. 1, pp.17–28.
Goedkoop M, 1996. The Eco-indicator 95 manual for designers, PRé Consultants, Amersfoort,
the Netherlands.
Gruber T.R., 1993. Toward principles for the design of ontologies used for knowledge sharing,
International Journal Human-Computer Studies, Vol. 43 (1993), pp. 907-928.
Ishii K., 1994. LASeR, Life-cycle Assembly, Service and Recycling – User’s Manual, Life-cycle
Engineering Group at Ohio State (LEGOS), The Ohio State University, Columbus, OH, 1994.
170
Kim K.Y, 2006. Ontology-based assembly design and information sharing for collaborative
product development, Computer-Aided Design, Vol. 38, No. 12, pp. 1233-1250.
Kitamura Y. and Mizoguchi R., 2004. Ontology-based systematization of functional knowledge,
Journal of Engineering Design, Vol. 15:4 pp. 327-351.
Lewis H. et al., 2001. Design + environment book: a global guide to designing greener goods,
Greenleaf Publishing Limited, Sheffield.
Lye S.W. et al., 2002. ECoDE – An environmental component design evaluation tool,
Engineering with Computers, Vol. 18, No. 1, pp. 14-23.
Meerkamm, H. ,1995. Design for recyclability – an analysis tool in the engineering workbench,
International Conference on Life Cycle Modeling for Innovative Products & Processes, pp. 501511.
Mizoguchi R., 2003. Tutorial on ontological engineering part 1: introduction to ontological
engineering, New Generation Computing, Vol. 4, No. 3, pp. 365-384.
Skarka W., 2007. Application of MOKA methodology in generative model creation using CATIA,
Engineering Applications of Artificial Intelligence, Vol. 20, No. 5, pp.677-690.
Stanford Protégé Wiki. DataMaster plugin. 2007.
http://protegewiki.stanford.edu/index.php/DataMaster
The CO-ODE Project. OWLViz Protégé plugin. 2007.
http://www.co-de.org/downloads/owlviz/
TouchGraph Vizualization Tab for Protégé. 2005. http://users.ecs.soton.ac.uk/ha/TGVizTab/
World Wide Web Consortium. Web ontology language (OWL). 2004.
http://www.w3.org/2004/OWL
Acknowledgements
The authors would like to thank the Spanish Ministry of Education and Science for their support
under the project numbered DPI2006-15570-C02-00 as part of the national I+D+i plan (2004 –
2007) y and the FEDER funds from the European Union. They would also like to thank the
Spanish Ministry of Environment for their support as part of project number 08I055. A final
thanks goes out to the rest of the Engineering Design Group (GID) at the University Jaume I for
their continued support.
Jessica Abad-Kelly
Engineering Design Group
Department of Mechanical Engineering and Construction
University Jaume I
Av. Sos Baynat s/n.
12071 - Castellón (Spain)
Phone: +34 964 72 9252 / Fax: +34 964 72 8106
171
URL: http://www.gid.uji.es/
David Cebrián
Engineering Design Group, Department of Mechanical Engineering and Construction, University Jaume I
Vicente Chulvi
Engineering Design Group, Department of Mechanical Engineering and Construction, University Jaume I
172
DESIGN PROCESS FOR AN ELECTROMAGNETIC POWER BRAKE
WITH ABS FUNCTION: CORPORATE KNOWLEDGE MANAGEMENT
Villanueva, P.
Martínez, P.
Pérez, A.
Public University of Navarre
Abstract
This paper will explain the methodology used to develop a new electromagnetic power brake,
which, used together with the brake master cylinder, will reduce driver force required during the
braking action. Until now, these mechanisms were assisted by engine vacuum, pneumatic
pressure or hydraulic pressure. The idea was to create a new power brake design that would
use the power from the vehicle battery to meet the pressure and volume requirements,
functioning by using the principles of electromagnetism. The design team applied the
company’s highly valuable knowledge in the field of certain power brake models and made
good use of its integrated corporate knowledge management framework. The result is a new
product with a braking system that is connected separately to each wheel and which provides
the pressure and volume requirements in accordance with the specific vehicle-road
environment. The new design includes a considerably simplified ABS electronic control.
Keywords: methodology, design, knowledge management, environmental impact, brakes.
1. Introduction
The pressure applied by a driver to the brake pedal must be boosted to adequately adjust the
pressure needed to stop or slow down the vehicle.
Normally to achieve this pressure, power brakes are designed that are regulated by the force
on the pedal. Depending on the vehicle system, these systems generally use vacuum, air or oil
pressure, the former being the most frequent system.
When the vehicle is also fitted with a brake ABS control unit, it needs a supplementary
hydraulic circuit with hydraulic solenoids, which are electronically regulated in accordance with
the wheel speed and slip to control the pressure sent to the brakes by the driver.
Depending on their requirements and usage, vehicles are fitted with a source of vacuum, air or
oil pressure. All vehicles have a battery; this can be used as a power source to obtain the
braking assistance that drivers need. In this project, the electronic control braking assistance
would function via electrical inputs without the need for any mechanical parts. This would
therefore potentially obtain a more direct regulation and dispense with the need for parts that
can produce annoying mechanical vibrations in braking processes with ABS application
requirements.
173
This paper will present the development process for a specific product. It will highlight the
design and development process, and how this can be used to improve the company’s
knowledge. When suitably classified and filed, this knowledge will help to manage and add to
the company’s experiences and knowledge.
Moreover, this paper will underscore how this power brake has environmental advantages
through reducing the waste generated at the end of its working life.
2. Design process
The product design process must be based on a method that identifies the stages of the
product design and development throughout the project’s entire life cycle.
Work methods are a useful tool to help industrial designers achieve their goal – to design an
efficient product that meets the needs of the customer and manufacturing company alike.
Pugh’s ‘Total Design’ philosophy was adopted to develop this power brake. The model used in
this projects includes four of the major stages in product design.
•
Market analysis and product feasibility
•
Design specifications
•
Conceptual design
•
Detail design
•
Prototypes and tests
•
Manufacturing
•
Sale
In this case, the market analysis and product feasibility stages were avoided as this project was
to initially develop a product using new technologies to then later study its technical feasibility.
And the project logically had to end with the prototyping and testing stage, which is when
conclusions could be drawn on the product’s technical and economic feasibility to see whether
it could be commercialised.
The life cycle of this development project therefore ran from the design specifications stage to
the prototype and testing stage
The product-specific knowledge of the company and its assets was applied in these different
stages as they were carried out.
3. Design specification (PDS)
In order to develop a useful product for immediate use, the aim was to create a prototype that
could be mounted on a current vehicle. The PDS therefore had to be drawn up with a specific
vehicle in mind.
Once the vehicle was chosen, a method was used to establish the product design
specifications (PDS).
174
In this case we did not have to meet the needs of a particular customer. The design team itself
was the customer. The RED method was used, with the following stages:
•
Intuitive search
•
Life cycle and environment
•
Sequential analysis of functional elements
•
Movements and forces
•
Internal reference products
•
External reference products (benchmarking)
The company’s knowledge could be used in any stage of this process.
1
2 3 4
5 6 7
8
Figure 1: Knowledge Repository.
With the prior knowledge that the prototype had to be adapted to vehicle (A), the repository was
consulted (figure 1, folder 1), as an example and to minimise the specifications, the following
functional requirements or PDS were defined.
•
Braking pressure: 75 bar
•
Driver force on the pedal: 300N
•
Volume to fill in the brakes: 5.34 cm3
•
Brake master cylinder: Ø 23.8 mm or 4.45 cm2
•
Cylinder pressure: 10 bar
•
Braking starts: 100 N
•
Oil: Dot 4
•
Operating Temperature: 20 ºC – 110 ºC.
•
ABS: Pressure regulation on each wheel
•
Pedal ratio: 5 to 1
Once the design specifications were defined and the knowledge was stored by the company,
information was provided on the way in which product performance level had to be reached
along with the details of the product operation:
Presión = Pressure; Volumen = Volume
175
80
Operating point
70
60
Pressure
area
Presión (bar)
50
40
30
20
Volume
Area
10
0
0
20
40
60
80
100
Volum en (%)
Figure 2 Volume-pressure curve (Repository Folder 2).
Fuerza pedal = Brake Pedal Force; Presión freno = Braking Pressure
Operating point
75 bar
ends
at
Presión freno
Braking
300 N
Braking
100 N
starts
at
300 N
Fuerza pedal
Figure 3 Pressure-Force Curve.
(Repository Folder 2)
176
4. Conceptual design
Once the PDS were defined, the next step was the conceptual design. Here, we had to bear the
following features in mind:
•
A single pressure generator behind the brake pedal
• A power brake per wheel that increased the pressure on each wheel differently so as to
separately regulate the pressure on each wheel.
•
Sensors on each of the wheels
•
Power supply for each of the power brakes from the battery.
• Power supply regulated by an electronic control that reacts to pedal force signals and
wheel signs of speed and deceleration.
Taking these features into account we arrived at the conceptual design shown in figure 4.
The brake master cylinder is unique, a first section of pressure is generated inside it to fill the
brake calipers or brake wheel cylinders. As can be seen in the volumes area of figure 2, the
pressure the cylinder has to achieve must be adjusted to the needs of that section and to the
force Fp, which is set according to regulations and specified in the PDS at 300N. The volume
(4.45 cm3), also contained in the PDS document, determines the stroke.
The brake master cylinder feeds 4 power brakes, one for each wheel, where an increase in
pressure takes place until the braking pressure is reached – established in the PDS document
at 75 bar.
The power brake starts to operate when the pressure reaches the end of the volumes area and
therefore only has to fill 40% of the wheel volume. As each wheel has its own power brake, only
¼ of that volume needs to be filled.
To increase the pressure reaching the wheels, an energy source is needed to create a
controlled force to add to the pressure that is transmitted from the cylinder.
That energy source is the vehicle battery. An electronic circuit was used to create this
controlled force: its inputs being the force exerted on the pedal and the wheel slip and its
outputs being the different currents transmitted to each power brake to regulate the brake assist
force.
Fp = Brake Pedal Force
Bateria = Battery
Pbom ba
Bom ba
Fp
Controlador = Control Unit
Servofreno = Power Brake
2
1
Fp
C o n tro la d o r
3
S e rv o fre n o
4
Pinza = Clip
Pf
Sensor = Sensor
I
Bomba = Master Cyclinder
w
P in z a
B A T E R IA
S e n so r
P Bomba
Pressure
=
Master
Cylinder
Pf = Brake Pressure
W = Angular Velocity
Figure 4: Conceptual design.
I = Current
177
The design team had prior experience in designing brake master cylinders and vacuum power
brakes. This experience is stored in the folder 5 of the Knowledge Repository.
This experience was also to be used for hydraulic part of the power brake and an alternative
solution had to be devised to vacuum, air and oil power brakes.
5. Conceptual design and sizing of the hydraulic components
5.1 Brake master cylinder.
Prior experience and knowledge was used to size the brake master cylinder. The brake volume
was a conditioning factor in the cylinder diameter. After making a calculations, we established a
Ø 23.8 mm brake master cylinder, which, based on the PDS, gives a pressure output of 29 bar
at the operating point. The remaining pressure to reach 75 bar had to be generated through the
power brake.
5.2. Power brake
Our experience led us to use a tried and tested solution. The conceptual design of this solution
is shown below in figure 5.
Servofreno
Circuito hidráulico en reposo
Pb
Pf
Servofreno
Circuito hidráulico trabajando
Figure 5: Power brake conceptual design.
•
Servofreno = power brake
•
Circuito hidráulico en reposo = quiescent hydraulic circuit
•
Circuito hidráulico trabajando = operational hydraulic circuit
•
Pf = brake pressure
•
Pb = Master cylinder pressure
The operating conditions are determined by the curve in figure 3. The power brake had to
provide the force necessary to raise the 29 bar pressure generated by the cylinder to 75 bar,
178
which in turn, had to determine the size of the whole system. The hydraulic part is shown in
Figure 6:
pushrod
piston
Fs
power brake
Pf
Pbomba
Figure 6: Balance of power in the hydraulic area.
Pbomba = master cylinder pressure; Pf = brake pressure; Fs = Power brake force
We made the relevant calculations and sized a Ø 12 mm piston and a Ø 5 mm pushrod, which
required a force of 675 N to be received from the power brake.
Figure 6 shows the conceptual design of the electromagnetic power brake and the forcegenerating magnetic element – an electrical coil that is powered from the battery and a
magnetic circuit with a fixed cover and mobile pushrod that generates the force to transmit to
the hydraulic circuit
Once the required force and displacements were calculated, the coil and circuit were sized. A
coil was designed with an operating current of 25 A.
Figure 7 shows the curves calculated for different currents and positions.
Force
800
750
700
650
600
550
500
Fs (N)
450
400
Current
350
300
250
200
150
100
50
Positio
0
-10
-8
-6
-4
-2
0
2
4
6
8
x (mm)
Figure 7: Curve of forces depending on the position and current.
179
6. Detail design
The total design (figure 8) was reached by using the conceptual design and the sizes.
The structural solutions of this design take into account similar designs included in the
Knowledge Repository, folder number 6.
Figure 8: Electromagnetic power brake.
Moreover, as this was a prototype that needed to be tested, this first design had to allow the
positions of the different parts to be easily and quickly changed.
7. Control: detail design
The conceptual design of the control unit had the following stages:
•
Input signal conditioning
•
Analog-to-digital conversion
•
Calculator
•
Power stage
•
Error check
Fp
U N ID A D D E C O N TR O L
1
2
3
B A T ER IA
Figure 9: Control unit Block design.
Fp = Brake Pedal Force; Unidad De Control= Control Uni t; Bateria = Battery
180
8. Control unit design details
This was the last stage. The design is shown below in figure 10
Figure 10: Control Unit Matlab Model.
•
P alta = high pressure
•
P baja = low pressure
•
P alta = high pressure
•
Intensidad = current
•
Fuerza = force
•
Curva fricción = friction curve
•
Peso = weight
•
Par Frenado = Brake torque
•
Velocidad vehículo = vehicle speed
•
Deslizamiento = slip
•
Par rozamiento = friction torque
•
Velocidad rueda = vehicle speed
•
Velocidad vehículo (angular) = Vehicle Speed (angular)
181
•
Distancia de frenado = Braking distance
9. Prototype and tests
To validate the product it had to tested to check it met the requirements set out in the PDS
document.
The prototype was built and placed on the test bench as shown in figure 11. A computer was
used to simulate the characteristics of a vehicle in transit.
Bomba
Fp
Pbomba
Pf
Fp
Etapa de
potencia
Servofreno
Pf
Pf
I
x
x
Batería
Bench
Apagado/
Encendido
Figure 11: Bench
•
Bomba = cylinder
•
Pbomba = cylinder pressure
•
Batería = Battery
•
Etapa de potencia = Power stage
•
Apagado / Encendido = Off / On
•
Servo freno = power brake
•
Pinza = clip
•
Esquema = Layout
•
Tarjeta de adquisición = Acquisition card
•
Ordenador = Computer
•
Controlador = control unit
•
Modelo rueda = Wheel model
•
Pf = Brake pressure
•
Fp= Brake pedal force
•
I = Current
Pinza
Tarjeta de adquisición
I
ORDENADOR
Controlador
Modelo
rueda
Pf
Esquema
182
•
X = Magnetic body displacemen
10. Results
A pressure-force curve was generated using the test results (figure 12). This demonstrates that
the product fulfils its function to increase pressure, which depends on the force.
And a slip curve was also generated, which shows the ABS function is successfully performed.
90
120
80
70
100
60
Pa (bar)
Pb (bar)
40
V(Km/h)
P (Bar)
80
50
Vr (rad/s)
Vv (rad/s)
60
30
40
20
20
10
0
0
0
0
10
20
30
40
1
2
50
3
4
5
6
7
t (s)
F (kg)
Figure 12: pressure-force curve and slip curve.
11. Environmental impact
In the electromagnetic power brake, the coil replaces the vacuum power brake and the brake
pressure modulation system has less parts and a set of valves, which operate with a high level
of vibrations and consume oil, is completely dispensed with.
12. Conclusions
•
The electromagnetic power brake design met the specifications set out in the PDS
document.
•
Both when drawing up the PDS document and seeking conceptual design and detail
design solutions, the company’s information and past experiences gathered together in the
Knowledge Repository were useful.
•
The design replaces other designs currently in use and has clear advantages in terms of
recycling and the use of less polluting materials
References
Aguayo, F., 2002. Metodología del diseño industrial., RAMA.
Arias-Paz, M., 2000. Manual de automóviles. Cie Dosswat.
Eder, EE. 2007, Design Engineering, CRC PRES.
183
Happian-Smith, J., 2003. Modern Vehicle Design, BH.
Luque, P., 2004 Ingeniería del automóvil. Sistemas y comportamiento dinámico THOMSON.
Dr. Paulino Martínez Landa. [email protected]
Dr. Pedro Villanueva Roldán. [email protected]
Dr. Amaia Pérez Ezcurdia. [email protected]
Public University of Navarre. Campus Arrosadía s / n 31006
Technical School for Industrial Engineering and Telecommunications
Projects and Rural Engineering Department
Phone: (+34) 948 169 330 / 948 169 330 331
184
Area: Environment
TOPIC AREA: ENVIRONMENT
185
Area: Environment
ANALYSIS OF ECO-EFFICIENCY PROGRESS IN INDUSTRIAL
PRODUCTION
Fernández-Viñé, M. B.
Gómez-Navarro, T.
Capuz-Rizo, S.F.
Universidad Metropolitana de Caracas. Venezuela.
Abstract
An eco-efficient business manages to produce high-quality, profitable goods and services, with
an environmental impact compatible with the Earth's carrying capacity. Eco-efficiency is
currently in a phase of expansion; numerous environmental management systems and tools,
support initiatives, and application experiences now exist. To understand how this strategy is
being developed and which driving forces enhance the implementation of eco-efficiency, a
comparative analysis of case studies both directly and indirectly related to eco-efficiency is
presented in this work. These surveys were undertaken in Canada, United Kingdom, Holland,
Finland, Spain and Latin America. In addition, an assessment of eco-efficiency performance in
small and medium sized business in Venezuela has been conducted. The first findings of this
assessment in Venezuela and their comparison to the other case studies are also discussed in
this paper. The comparison enables us to draw conclusions regarding the driving forces of ecoefficiency which are: training and motivation of human resources, the market (costumers,
suppliers, and competitors), environmental legislation, production technologies and
infrastructures, among others. The conclusions of this research highlight the differences and
similarities between regions, establishing cause-and-effect relationships between the driving
forces and the evidence of eco-efficiency found, allowing for a better understanding of how to
promote eco-efficiency in businesses.
Keywords: Eco-efficiency, Cleaner production, Ecodesign, EMAS
1. Introduction
Eco-efficient production represents the most important contribution of industries to improve
quality of life by developing added-value products without reducing business profits while
progressively reducing environmental impacts (UNEP-WBCSD,1998).
The path that leads companies from pollution control of the production processes to eco-design
and eco-innovation goes through the concept of eco-efficiency. Eco-efficiency is a strategy to
reduce environmental impact along the life cycle of the product (and not only during the
production stage) while increasing business competitiveness.
In 1991 the World Business Council for Sustainable Development defined Eco-efficiency as
187
Area: Environment
“the delivery of competitively priced goods and services that satisfy human needs and bring
quality of life while progressively reducing ecological impacts and resource intensity throughout
the product life cycle, to a level at least in line with the earth's carrying capacity” (UNEPWBCSD, 1998). Next is the list of eco-efficiency strategies put forward by the Atlantic Canada
Opportunities Agency (2002), which are essentially mirrored on those of the WBCSD.
ECO-EFFICIENCY STRATEGIES
- Reduce the amount of material required in the production of goods and services
- Reduce the amount of energy required in the production and delivery of goods and services
- Reduce toxic emissions
- Design products to be recyclable
- Maximize the use of renewable resources in the production and delivery of goods and services
- Design products that are durable and easy to repair
- Design products that serve multiple purposes and that can be easily upgraded
Table 1: Eco-efficiency strategies (Atlantic Canada Opportunities Agency 2002).
According to Coté et al. (Coté, Booth and Louis, 2006), Eco-efficiency is enhanced by factors
referred to as “driving forces”, “success factors” or “stimuli”. Briefly, eco-efficiency driving forces
can be enumerated and classified depending on whether they arise from inside or outside the
company, as shown in Table 2:
INTERNAL DRIVING FORCES
EXTERNAL DRIVING FORCES
Cost reduction
Consumer demand for “greener” products
Increase in the quality of products and services
Influence of the parties
involved in the process
Innovation
Access to capital
Increase in employee motivation
Competition
Personal
community
commitment/responsibility
to
or
stakeholders
Government regulations
Management of risks and liabilities
Public pressure
Maintaining or increasing market share
Global pressures, e.g. climate change
Table 2: Eco-efficiency driving forces (Coté, Booth and Louis, 2006).
2. Objective of the research
The aim of the present research is to identify the driving forces that contribute to improving ecoefficiency in SMEs in Venezuela. For this end a comparison analysis of eco-efficient practices
in different countries was performed. The level of development of the countries was identified in
188
Area: Environment
terms of level of implementation of eco-efficient practices and their relationship with barriers
and stimuli for eco-efficiency. This analysis allows a diagnosis of the situation of Venezuelan
SMEs as compared to the other countries under analysis.
The main objective of the present research work is, then, to review other surveys directly or
indirectly related to eco-efficiency and to compare the results with a preliminary diagnosis of the
level of implementation of eco-efficient practices in Venezuelan SMEs. The diagnosis was
obtained through a survey undertaken in small and medium-sized industries in the central
region of Venezuela, by far the most industrialized region of the country.
3. Methodology and Case Study
3.1. Analyzed Surveys
Table 4 shows the results of the comparison analysis. It presents the eco-efficient practices and
management systems implemented in 7 regions or countries (including Venezuela). The
surveys used in the comparison analysis are:
‐
‐
van Hemel and Cramer (2002), who analyzed 77 small and medium manufacturing
companies of The Netherlands belonging to the metal-mechanic, timber, plastic, textile
and electronic industrial sectors. The study analyzes SME eco-design barriers and
stimuli, an essential eco-efficiency tool (van Hemel and Cramer, 2002)
Coté et al. (2006); a survey of 25 SMEs in Nova Scotia, Canada, was conducted to
measure the levels of eco-efficiency in the SMEs of the region (Coté, Booth and Louis,
2006).
‐
Erkko et al. (2005); their study analyzes to what extent Finish companies with EMAS
(Eco-Management and Audit Scheme) statements have added eco-efficiency strategies
to their practices. The study is based on the EMAS reports published by 40 companies
in the chemical, paper, plastic and metal-mechanic sectors (Erkko, Melanen and
Mickwitz, 2005).
‐
A study carried out in April 2005 by NETREGS, the UK resource for the protection of the
environment, through a telephone survey of 5554 SMEs, which represent 99% of the
business activity developed in the country (Netregs, 2005).
‐
The work by Gómez (2004) analyzes the environmental performance of the SMEs in the
Region of Valencia, Spain (Gómez, 2004).
‐
Vives et al. (2005) conducted a comparison analysis in SMEs of different LatinAmerican countries (Argentina, Brazil, Chile, Colombia, El Salvador, Mexico, Peru and
Venezuela) in order to determine the level of implementation of Corporate Social
Responsibility. CRS is oriented towards sustainable development and, as a result,
towards environmental sustainability
‐
Although most of these surveys have similar general objectives and methodologies, they differ
in their specific objectives. Thus, the survey undertaken in The Netherlands focuses on the
implementation of eco-design in SMEs (van Hemel and Cramer, 2002), whereas the study
developed in Finland (Erkko, Melanen and Mickwitz, 2005) analyzes what is done and how ecoefficiency is reported by the companies that elaborated their environmental reports; on the other
hand, the Canadian study (Coté, Booth and Louis, 2006) and the Spanish study of the
Valencian Region (Gómez, 2004) analyze the levels of implementation of eco-efficiency in
SMEs, whereas the survey undertaken in the UK (Netregs, 2005) analyzes environmental
189
Area: Environment
management on a general basis; finally the survey conducted in Latin-America and the
Caribbean (Vives, Corral and Isusi, 2005) studies the corporate social responsibility of the
SMEs in different countries of the area.
Likewise, these works differ in target and scope; for example, the Finish study is not addressed
to SMEs, and the number of companies ranges from 25 in the Canadian survey to 5554 SMEs
in the UK study. Finally, some studies present detailed and, therefore, easily interpreted
descriptions of the findings, whereas other works, like the Canadian survey, present general
results using statements like “most say”, “the most frequent indicators” or “many SMEs do not
perceive…” but they do not provide specific figures or percentages.
Despite these difficulties and the lack of other available sources in the scientific literature, we
made an attempt to measure the level of implementation of eco-efficiency strategies using the
qualitative scale shown in Table 3:
% OF COMPANIES
QUALITATIVE SCALE
80-100%
VERY HIGH
60-79%
HIGH
40-59%
MEDIUM
20-39%
LOW
1-19%
VERY LOW
0%
NOT STUDIED
Table 3: Qualitative scale.
Table 4 shows the comparison analysis of the reviewed research works as well as the study
conducted in Venezuelan SMEs. The issues not addressed in the surveys are indicated under
the heading “NOT STUDIED”. In the last three rows, an X indicates those surveys that identify
saving as a key internal force for the implementation of practices aimed at reducing the use of
energy and materials, those surveys that identify customer demands and environmental
legislation as external driving forces to take into consideration for the implementation of
environmental actions and plans, and those surveys that report a close relationship between
company size and environmental commitment.
3.2. Situation of the SMEs in Venezuela
According to the National Institute of Statistics, in 2003 (latest data available) there were 5970
industries in the country, 1592 (26.67%) of which are medium-sized industries, 3820 (64%) are
small industries and the remaining 9.33% are large companies (Instituto Nacional de
Estadística, 2003).
The total taskforce of Venezuela is 11,172 million workers, 56.5% of whom, i.e. about 6.3
million, work in the formal sector and about 5 million in the informal sector. The informal sector
of the economy refers to workers that offer their products and services in the street and do not
enjoy any labour benefits or social health care. The private business sector employs 81.5% of
the total formal taskforce, i.e. 5,150 million workers, and about 1,167 million work as civil
servants. About 99% of the workers in private companies (5,093 million) work in SMEs (Instituto
Nacional de Estadística, 2007).
190
Area: Environment
Only 2 to 5% SMEs are oriented towards the international market; 80% produce for the local
market, and 15 to 18% for the regional and national markets (Páez, Rodríguez, Itriago and
Cuárez, 2004).
To verify the results of the comparison analysis, a survey on the issues shown in Table 4 was
undertaken in 54 SMEs of Venezuela.
i. Analysis of overall environmental performance by industrial sector
The graph in Figure 1 was designed to better illustrate the environmental performance of the
industrial sectors under study.
COMPARATIVE ANALYSIS BY INDUSTRIAL SECTOR
100,0%
90,0%
80,0%
70,0%
60,0%
50,0%
40,0%
30,0%
20,0%
10,0%
0,0%
FOOD
METALMECH
PAPER
TEXTILE
BUILDING
PLASTIC
CHEMICAL
Consider environment in products/processes
Green marketing of products
Environment Department
Asign budget for pollution reduction
Personel training in environment
Enviromental Management System
Environmental regulations knowledge
Pollution control
Pollution prevention
Product Life Cycle Assessment
Energy consumption reduction
Material consumption reduction
Recuperation and recycling of waste materials
Ecodesign
Packaging and packing management
Environmental efficient product transportation
WOOD
Figure 1: Analysis by industrial sector.
As depicted in the graph, the Food sector is the industrial sector more actively engaged in the
prevention of environmental pollution, closely followed by the Chemical industries; and in
decreasing order, the Textile, Metal-mechanic, Building, Paper, Plastic and Wood industries
(from left to right in the graph).
ii.
Analysis of overall environmental performance by company size
Figure 2 shows the overall environmental performance by company size (micro, small and
medium-size industries). This figure reveals the tendency of the SMEs to improve their
environmental performance depending on company size; micro-companies obtain slightly better
results than small businesses. These results may partly be due to the fact that only a few microcompanies answered the questionnaire, particularly those best managed and more sensitive to
environmental impact. It was difficult to interview micro-companies as they were reluctant to
participate in this kind of surveys.
The results of the survey in the Venezuelan SMEs were added to the comparison analysis of
191
Area: Environment
the other surveys undertaken in different countries and described above. The comparison
analysis can be seen in Table 4.
ENVIRONMENTAL PERFORMANCE BY COMPANY SIZE
100,0%
90,0%
80,0%
70,0%
60,0%
50,0%
40,0%
30,0%
20,0%
10,0%
0,0%
MICRO-BUSINESS
SMALL BUSINESS
MIDDLED-SIZED
Consider environment in products/processes
Green marketing of products
Environment Department
Asign budget for pollution reduction
Personel training in environment
Enviromental Management System
Environmental regulations knowledge
Pollution control
Pollution prevention
Product Life Cycle Assessment
Energy consumption reduction
Material consumption reduction
Recuperation and recycling of waste materials
Ecodesign
Packaging and packing management
Environmental efficient product transportation
Figure 2. Environmental performance by company size.
4. Discussion
The comparison analysis shown in Table 4 indicates that:
‐
‐
‐
‐
‐
‐
‐
The average level of environmental commitment of the SMEs in the different countries
studied is medium-high. By contrast, Venezuelan SMEs generally present a low level of
environmental commitment. This can be explained by the fact that Venezuela, like other
Latin-American countries, has serious social problems which are more important than
sustainable development.
The market, though considered as an influential external driving force, receives little
attention in the industries analyzed in the reviewed surveys. In Venezuela, the market is
even a weaker driving force as the population lack environmental education.
The level of implementation of an environmental management system is low or very low
in the studies of the comparison analysis. The situation is similar for the case of
Venezuela as only a very small number of companies report on the use of
environmental management systems and tools.
The surveys analyzed indicate that SMEs have no environmental department nor
trained personnel in environmental issues. Besides this, the budget allocated to reduce
environmental impacts is small if any.
Surprisingly enough, except for the Dutch case, the rest of surveys report a poor or very
poor knowledge of the environmental legislation. By contrast, Venezuelan SMEs seem
to be better informed.
Most companies do not conduct Product Life Cycle Assessment. This again evidences
that this useful tool is only relevant in academic environments or in large businesses.
Energy and materials reduction, recovery and recycling are the most common
environmental practices mentioned in the studies under analysis, suggesting that such
activities seem to be more motivated by the cost savings they generate than by
environmental concern. The same behavior is observed in the Venezuelan SMEs.
192
Area: Environment
SURVEYS
NETHERLAND
CANADA
VAN HEMEL &
CRAMER
COTÉ et al.
(2002)
(2006)
FINLAND
UNITED
KINGDOM
VALENCIA
LATIN
AMERICA
ERKKO et
al. (2005)
NETREGS
(2005)
SPAIN
Vives et al.
(2005)
ASPECTS
CONSIDER
ENVIRONMENT IN
PRODUCTS/PROCESSES
ENVIRONMENTAL
MARKETING
VENEZUELA
(2003)
NOT STUDIED
HIGH
NOT STUDIED
HIGH
NOT STUDIED NOT STUDIED
HIGH
MEDIUM-HIGH
MEDIUM
LOW
MEDIUM
LOW
NOT STUDIED
VERY LOW
ENVIRONMENTAL
DEPARTMENT
NOT STUDIED
NOT STUDIED NOT STUDIED MEDIUM-LOW
LOW
NOT STUDIED
VERY LOW
BUDGET ASSIGNED TO
ENVIRONMENT
NOT STUDIED
NOT STUDIED NOT STUDIED NOT STUDIED
MEDIUM
NOT STUDIED
LOW
TRAINING IN
ENVIRONMENT
NOT STUDIED
LOW
NOT STUDIED
LOW
ENVIRONMENTAL
MANAGEMENT SYSTEM
NOT STUDIED
LOW
NO ANSWER
VERY LOW
VERY LOW
HIGH
VERY LOW
LOW
NOT STUDIED
MEDIUM
HIGH
VERY LOW
LOW
ENVIRONMENTAL
REGULATION
KNOWLEDGE
POLLUTION CONTROL
NOT STUDIED
NOT STUDIED MEDIUM-LOW NOT STUDIED
POLLUTION PREVENTION
MEDIUM-HIGH
NOT STUDIED
LOW
NOT STUDIED
VERY LOW
NOT STUDIED
LOW
LIFE CYCLE
ASSESSMENT
NOT STUDIED
NOT STUDIED
VERY LOW
NOT STUDIED
VERY LOW
NOT STUDIED
VERY LOW
ENERGY CONSUMPTION
REDUCTION
HIGH
MEDIUM
HIGH
VERY LOW
HIGH
LOW
LOW
MATERIAL
CONSUMPTION
REDUCTION
HIGH
MEDIUM
LOW
NOT STUDIED
HIGH
NOT STUDIED
MEDIUM
RECUPERATION AND
RECYCLING OF WASTE
MATERIALS
HIGH
NOT STUDIED
MEDIUM
MEDIUM
HIGH
LOW
HIGH
ECODESIGN
PACKAGING AND
PACKING MANAGEMENT
VERY LOW
LOW
NOT STUDIED
VERY LOW
MEDIUM-HIGH
HIGH
NOT STUDIED
MEDIUM
X
X
X
X
X
X
X
X
X
X
NOT STUDIED
X
DRIVEN BY SAVINGS
X
RELATION BETWEEN
COMPANY SIZE AND
ENVIRONMENT
NOT STUDIED
DRIVEN BY CONSUMERS
AND LEGISLATION
X
X
X
X
X
Table 4: Comparison analysis of the reviewed surveys and the case study of Venezuela.
193
Area: Environment
‐
‐
Special attention should be paid to the issue of waste material recycling and recovery. In
the highly developed countries, this activity is regulated by law (rather than by
environmental awareness) through authorized entities and systems and compliance
costs require that companies incur in expenditures to adjust themselves to the norms
set in the regulation. By contrast, in Latin-American countries recycling is a profitable
business that generates production cost savings to the companies as it is based on
human labour rather on automatic systems.
The surveys of the analysis also suggest a lack of eco-design strategies in SMEs. Thus,
Ecoefficiency practices are generally addressed towards the production process rather
than to the end product. This trend is changing in Europe with the so called
“Environmental New approach”, but no changes are expected in Latin America or
Venezuela unless there is an increase in the export sales.
‐
Legislation and customer demands are considered powerful external stimuli in most of
the surveys, although a low or very low number of companies report on having a deep
knowledge of the legal regulations and few companies report Green Marketing.
Venezuela shows a similar situation: SMEs declare to have a fair knowledge of the legal
framework and to pay great attention to their customers’ demands, but when asked
whether they emphasize the environmental issues in their commercial actions, very few
companies answer affirmatively.
‐
The studies that include SME size in the analysis show a positive close relationship
between size and environmental commitment. This is also the case in the Venezuelan
SMEs
‐
The analysis of the results by regions reveals three levels of eco-efficiency
implementation in the industries under study. At the highest level of implementation we
find the Dutch companies. At a second level, the companies of Finland, United
Kingdom, Canada and Spain. And at a third level the Latin-American companies. These
implementation levels correlate to the overall perception of the situation of the markets
in these regions and countries. The comparison analysis of the driving forces acting on
each case explains the different levels of ecofficency implementation.
5. Conclusions
The comparison analysis of the surveys shown in Table 4 allows us to draw the following
conclusions on ecoefficiency driving forces:
‐
‐
‐
‐
In all surveys cost saving seems to be the most influential internal driving force that
enhances the implementation of ecoefficient systems, and in the Venezuelan case it is
also a powerful stimulus for reducing the environmental impact of products and
processes
On the other hand, although the issue of product quality is of general concern,
environmental respect is not considered a quality criterion. Similarly, no relationship
between innovation and environmental concern was observed, and although many
companies innovate to keep their competitiveness, they generally do it to improve the
quality of products but not to reduce environmental impacts
Increasing employee motivation and environmental responsibilities does not seem to be
a key issue for SMEs. This is even more remarkable in Venezuela, where there is a
general lack of environmental concern
Management of risks and liabilities only try to comply with the minimum standards set
out in the legislation. In Venezuela, a weak formal enforcement regime allows the
194
Area: Environment
companies to have higher non-compliance rates
‐ Finally, as regards the measures to maintain or increase the market share, SMEs seem
to be very reactive, acting locally to customer demands. In Venezuela, the efforts made
in this direction do not seem to be guided by a reduction of the environmental impacts of
the product
On the other hand, with respect to the external driving forces acting on the environmental
performance of the SMEs, it can be concluded that:
‐ Consumers do not seem to strongly demand for “green” products; in Venezuela, as
mentioned earlier, due to the little environmental concern from consumers, this situation
is more dramatic and the Green Marketing is practically non existent.
‐ No effective influence of the stakeholders on SME’s attitude towards environmental
protection is observed. For example, access to capital is not affected by the adoption of
environmental practices like publishing environmental reports in the companies. civil
associations, educational institutions, national organizations and experts in environment
do not exert any effective pressure to force SMEs to introduce the required changes for
process and product environmental impact minimization
‐ Companies do not seem to address their technical efforts to improve competitiveness
regarding the environment; rather they address them towards the operational
improvement of the product. Compliance with the environmental regulations depends on
the enforcement system. In Venezuela despite the extensive environmental legislation,
compliance rates are low due to their weak regulatory procedures
‐ Although in general SMEs seem very reactive to punctual demands from customers,
they do not seem very sensitive to media campaigns in favour of the preservation of the
environment as SMEs generally think that large companies should take the lead in the
adoption of environmental practices.
The analysis of the surveys and the case study of Venezuela suggests the following causeeffect relationships among the driving forces and the eco-efficiency evidences observed:
‐ Those countries whose SMEs have high export sales, are influenced by the
environmental legislation for product import of the countries they sell their products to.
This is a key external driving force for these companies. As mentioned earlier,
Venezuelan SMEs have low export sales, so that they are little influenced by global
environmental demands
‐ In the countries where taskforce and energy are expensive additional investment is
required for materials and waste recycling. In Venezuela it can be observed a high rate
of materials and waste recovery and recycling motivated by cost savings rather than by
environmental awareness, since both taskforce and energy are very cheap
‐
In countries with a more degraded environment, local and global pressures in favor of
the environment find more support since the environmental situation is more evident to
the community. In Latin-American countries, in particular in Venezuela, which have a
less degraded natural environment, global campaigns of environmental protection do
not make the same effects on people.
‐
The efficiency of the enforcement mechanisms greatly affects environmental
performance in SMEs. In particular, in Venezuela due to the lack of efficient regulatory
procedures, this driving force is not very influential on the firm’s environmental decisions
‐
Incentives for environmental protection that have already been implemented in many
developed countries but which have not yet been adopted in countries such as
Venezuela (e.g. subsidized credits, fiscal incentives and eco-labels) can contribute to
raise environmental awareness in the population, to stimulate Green Marketing and to
facilitate access to capital to those SMEs with proven eco-efficiency practices
‐
On the other hand, it has been observed that the driving forces are closely inter-related
195
Area: Environment
‐
‐
and are synergic in the positive and negative aspects. For example, consumers’
environmental concern compels both firms and regulators towards the adoption of ecoefficiency practices. Regulators, in turn, regulate eco-efficiency norms and also compel
firms towards the adoption of eco-efficiency practices. By contrast, a lack of social
interest makes companies and mass media not care about the environment, and as a
consequence, they do not exert any pressure on public administration for the regulation
and protection of the environment
On the other hand, the companies’ improvement of ecoefficiency generates more ecoefficiency by the effect “example for the competitor”, the effect “consumer awareness
raising” and by the effect “demand for eco-efficiency along the supply or value chain”
Therefore, the countries or regions enter a vicious circle in which there is no offer
because there is no green-product demand nor awareness nor enforcement; there is no
enforcement (by the administration) because there is no demand (by the citizens) and
there is no demand because there is no offer (and the publicity generates other different
shopping needs). However, when eco-efficiency improves, that generates a virtuous
circle as more companies discover the potential market opportunities of eco-efficiency,
or they are compelled to it as the driving forces are strengthened and inter-reinforced.
References
Atlantic Canada Opportunities Agency, 2002. Eco-efficiency fact sheet. Moncton: New
Brunswick. http://www.acoa.ca/e/library/reports/40130205 eco-efficient fact sheet eng.pdf
Coté, R., Booth, A., Louis, B. 2006, Eco-efficiency and SMEs in Nova Scotia, Canada, Journal
of Cleaner Production, vol. 14, pp. 542-550.
Erkko, S., Melanen, M., Mickwitz, P., 2005 Eco-efficiency in the Finnish EMAS reports – a buzz
word?, Journal of Cleaner Production, vol. 13, pp. 799-813.
Gómez, T., 2004. Propuesta metodológica para la mejora de la ecoeficiencia de los productos
industriales a lo largo de su ciclo de vida. Aplicación a las PYME de la Comunidad Valenciana.
Tesis Doctoral. Universidad Politécnica de Valencia. Valencia.
Instituto Nacional de Estadística, 2003 Estadísticas Económicas. Obtenido el 10 de febrero de
2006, de http://www.ine.gob.ve/industria/Indus_Man_2003.htm
Instituto Nacional de Estadística, 2007 Estadísticas Económicas. Obtenido el 12 de marzo de
2007, de
http://www.ine.gob.ve/tripticos/IndicadoresEconomicos/C1IndicEcon.asp?periodo=200701
Netregs, SME-nvironment, 2005 A review of changing environmental attitudes and behaviors
among small and medium sized business in the UK”, 2005, Obtenido el 1 de octubre de 2006
en http://www.netregs.gov.uk/comillonesondata/acrobat/2005_uk_sumillonesary_1197319.pdf
Páez, T., Rodríguez, B., Itriago, D., Cuárez, M., Observatorio PyMEs y nuevas tendencias de la
información y la comunicación, Ceatpro, Caracas, 2004.
UNEP-WBCSD, 1998 Cleaner production and Eco-efficiency: from ideas to action, Geneve:
UNEP.
van Hemel, C., 2002 Cramer, J., Barriers and stimuli for ecodesign in SMEs. Journal of Cleaner
Production, vol. 10, pp. 439-453.
196
Area: Environment
Vives, A., Corral, A. Isusi, I., 2005 Responsabilidad Social de la Empresa en las PyMEs de
Latinoamérica, Subdepartamento de Empresa Privada y Mercados Financieros, Banco
Interamericano de Desarrollo, Washington DC.
María Blanca Fernández
Universidad Metropolitana
Facultad de Ingeniería, Terrazas del Ávila, 1070 Caracas, Venezuela.
Phone: 58-212-240.35.03
Fax 58-212-240.34.96
197
Area: Environment
LAND OCCUPATION AND TRANSFORMATION BY SPANISH ROADS
Garraín, D.
Vidal, R.
Franco, V.
Universitat Jaume I
Abstract
The term ‘land use’ is traditionally used to denote a classification of human activities which
occupy land area. In the field of Life Cycle Assessment (LCA), the terms ‘land use’ or ‘land use
impact’ have been used to designate the environmental impacts related to physical occupation
and transformation of geographic areas. Our study focuses on the calculation of land surface
occupation attributed to Spanish roads, stressing the discrepancy of results determined by the
choice of data source. Occupation values are subsequently used to calculate temporal
occupation and transformation values of soil into roadways. Transformation values can be
incorporated into LCAs of this type of infrastructures, since they can be expressed in terms of
vehicle flow. Finally, the need to consider the potential territorial fragmentation and implications
on land productivity during road planning is highlighted.
Keywords: Land Use, Occupation, Transformation, Roads, Land Fragmentation
1. Introduction
The term ‘land use’ has traditionally been used to denote a classification of human activities
that occupy an area of land. In the field of LCA the term ‘land use’ or ‘land use impact’ is used
to denote environmental impacts relating to the occupation and transformation of physical areas
of terrain (Weidema & Lindeijer, 2001).
In studies to assess the environmental impact of products or processes through the application
of LCA, the impact category for land use is not always considered or may not be appropriate in
the carrying out of these assessments. This circumstance, together with the scarcity of
available data, has meant that the application of land use indicators is still complex today due to
a lack of definition of the parameters to consider and the methodology to use.
Land use, whether for agriculture, forest resources, mining resources, construction of housing,
industry, infrastructure, etc., involves environmental impacts that are often substantial,
especially those affecting biodiversity and its quality.
Various authors have revised the different indicators for the calculation of land use impact.
Initially, assessment was based on the area of terrain occupied. The impact category of land
use was related to the occupied area used in relation to the life cycle of the product or system,
calculated without distinguishing between the different forms of land use and without
198
Area: Environment
considering the original state of the land (Lindeijer, 2000a). Then there was a consensus that
occupation should be combined with the time during which human activities were carried out;
therefore, the period of time when the activity was performed was considered in addition to the
area occupied and the impact was measured as area multiplied by time (Guinée, 2002).
However, this occupation value must be accompanied by a qualitative assessment of changes
undergone by land as a result of its use. It is now internationally accepted that the life cycle
inventory phase must focus on two aspects of land use: the area occupied and the changes
caused to the land quality. Thus two concepts are involved: occupation and transformation of
land (Weidema & Lindeijer, 2001; Antón, 2004).
Later studies have developed methods for presenting indicators that measure the deterioration
of biodiversity and the alterations to the life support functions of the land due to its use.
Among these methods, the most noteworthy are those developed for the analysis of land use
impact on biodiversity (Köller, 2000, 2001, 2007, 2008; Goedkoop & Spriensma, 2001;
Michelsen, 2008; Müller-Wenk, 1998; Vogtländer et al., 2004); those developed to analyse the
impact on life support functions (Cowell, 2000; Milà i Canals, 2003; Milà i Canals et al., 2007);
and those assessing the impacts on biodiversity and on life support functions separately
(Weidema & Lindeijer, 2001; Lindeijer, 2000b; Mattsson et al., 2000).
Furthermore, some authors (Muys & García Quijano, 2002; Wagendorp et al., 2006) have
developed new methods for analysing the impact due to land use based on measuring the
capacity an ecosystem possesses to dissipate exergy. Kloverpris et al. (2008) present an
inventory model of land use resulting from the consumption of crops.
Other authors have performed qualitative comparisons of the methods that take into account
impacts on biodiversity and life support functions (van der Voet, 2001), or they have applied
some of these to particular cases: construction materials (Lindeijer et al., 2002), greenhouses
(Antón et al., 2007) and vegetable oil crops (Mattsson et al., 2000).
This study calculates the values of the land occupied by the Spanish roads and of the
transformation into road surface. Moreover, it considers the potential effect this infrastructure
can cause on the fragmentation of the territory and calculates the ecological footprint of the
CO2 emitted by vehicles using the roads.
2. Occupation of land by Spanish roads
The purpose of this indicator is to provide an estimate of the amount of land occupied by road
infrastructure to enable the subsequent assessment of its environmental impact.
In order to determine the area equivalent to the occupied land, various data sources have been
used.
2.1 Source 1: National data and direct calculations
In this case the data sources were:
199
Area: Environment
•
•
The Ministry of Development (MFOM, 2005) to find the length of the road network for which
the Spanish Government and the autonomous communities are responsible. The average
width estimated for each type of road was 25 m for major trunk roads and motorways; 17 m
for dual carriageway roads; 10 m for roads > 7 m in width; 6 m for roads between 5 and 7 m
in width; and 5 m for roads < 5 m in width.
The current legislation at a Spanish Government and autonomous community level to
obtain data on the affected area that will be provided by the affected width, or easement,
described in Table 1.
Region
Motorways
& major
trunk roads
Easement (m)
Dual
Width
Width
carriag
>7m
5-7 m
eway
Width
<5m
Source
SPAIN
25
8
8
8
8
Andalusia1
Aragon
Asturias
Balearic Islands
Canary Islands
Cantabria
Castile-la Mancha
Castile and León
Catalonia
Extremadura
Galicia
La Rioja
Madrid
Murcia
Navarre
Basque Country
Region of
Valencia2
25
25
25
25
25
25
25
25
25
17
10
50
25
25
8
8
8
18
17
8
8
8
8
2
8
10
30
8
8
8
8
8
8
8
18
8
8
8
8
2
8
10
30
8
8
8
8
8
8
8
14
8
8
8
8
2
8
10
30
8
8
8
8
8
8
8
10
8
8
8
8
2
8
10
30
8
8
RD1812/1994
RD1211/1990
Ley 25/1988
Ley 16/1987
Ley 8/2001
Ley 8/98
Ley 13/86
Ley 5/1990
Ley 9/91
Ley 5/96
Ley 9/90a
Ley 2/90
Ley 7/93
Ley 7/95
Ley 4/94
Ley 2/91
Ley 3/91
Ley 9/90b
Ley Foral 11/86
Ley 2/89
25
8
8
8
8
Ley 6/91
1
Ceuta and Melilla are included here. 2 General values have been assumed for the easement (25 metres for motorways and major trunk roads, and 8
metres for other roads) since there is no distinction between the easement and the protection area.
Table 1: Width of easements according to Spanish and autonomous community legislation.
Considering these sources, the total area occupied by Spanish roads in the year 2004 was
449,313.5 hectares.
2.2 Source 2: European data and indirect calculations
In this case the data sources were Eurostat (2006) for calculating the length of the road network
and the European Environment Agency (EEA, 2000) for obtaining an indirect occupation factor
that takes into account the effects caused by the areas adjoining the roads. Table 2 shows the
results.
200
Area: Environment
Type of road
Motorways and major
trunk roads
Spanish Government
roads
Provincial roads
Community roads
Road network
(km) (Eurostat,
2006)
Occupation factor
(ha/km) (EEA, 2000)
Area occupied
(ha)
9,571
7.5
71,782.5
17,074
6
102,444
69,167
67,987
4.5
2
311,251.5
135,974
Table 2: Length of the road network, indirect occupation factor and area occupied for Spanish roads according to
European data for 2004.
Therefore, the total area occupied by Spanish roads in the year 2001 according to these
sources was 621,452 hectares.
2.3 Source 3: Data from the CORINE Land Cover project
The CORINE Land Cover (CLC) project or programme has the fundamental objective of
capturing numeric and geographic data for the creation of a European database, at a scale of
1:100,000, of the Cover and/or Use of the Territory.
The European Topic Centre on Terrestrial Environment (ETC/TE), accountable to the AEMA, is
responsible for the acquisition and release of data corresponding to distinct European land
uses. One of the land uses that it distinguishes is that corresponding to “road and railway
network and associated land”, determined in hectares. Therefore the value of this land use will
be that corresponding to the indicator.
Table 3 shows the data gathered in the CORINE Land Cover project, provided by the European
Topic Center on Terrestrial Environment (ETC/TE), that correspond to the changes in land use
that occurred in Spain from 1990 to 2000. Therefore, it is only necessary to take into account
the data that correspond to the “road and rail network and associated land” in the year 2000.
The table below shows the data on this land use and from what other type of use the land was
transformed.
201
Area: Environment
Code
Type of use in 1990
Area (ha)
111
Continuous urban fabric
92
112
Discontinuous urban fabric
14
121
Industrial or commercial units
55
122
Road and rail networks and associated land
3.835
131
Mineral extraction sites
32
133
Construction sites
174
211
Non-irrigated arable land
815
212
Permanently irrigated land
132
222
Fruit trees and berry plantations
105
223
Olive groves
81
231
Pastures
386
241
Annual crops associated with permanent crops
89
242
Complex cultivation patterns
575
243
Land principally occupied by agriculture (with areas of natural
vegetation)
108
244
Agro-forestry areas
84
311
Broad-leaved forest
38
312
Coniferous forest
99
313
Mixed forest
33
321
Natural grasslands
296
322
Moors and heathland
28
323
Sclerophyllous vegetation
187
324
Transitional woodland-shrub
96
333
Sparsely vegetated areas
49
Table 3: Area transformed into road and rail network and associated land in 2000 according to the CLC project (CLC,
2000).
So the total area occupied by Spanish roads according to the CLC project for the year 2000
was 7,403 hectares.
2.4 Discussion of the results
A great difference between the values can be observed when comparing the results obtained.
From the point of view of greatest accuracy, the calculation method based on national data is
the most suitable and exact since it takes into account every type of road, every width, and
above all, every piece of legislation applicable to the type of road and autonomous community.
202
Area: Environment
The higher value obtained with European data is mainly due to the higher values for both the
length of the network of Spanish roads and the width in comparison to national data.
The value obtained with the CLC source is in the order of 100 times lower, due mainly to the
low resolution used in the project data capture. The minimum mapping unit for the CLC project
is 25 hectares with a minimum polygon width of 100 metres. Therefore, considering that each
map unit corresponds approximately to a rectangle of 100 x 2,500 metres, it is very difficult for
the satellite to distinguish the area occupied by a normal road, since the majority of transport
infrastructure in Spain is of less than 100 metres in width. However, it does consider the large
motorways, major trunk roads and other important road networks, leading to the very much
lower value obtained by this method. Figure 1 shows an example of the data capture.
Figure 1: CLC interpretation of major road networks (CLC, 2000).
Taking into account the total area of Spain (50,464,500 ha) and using national data, the land
this infrastructure occupies represents almost 0.9% of the total. In comparison with other
countries in the European Union (Belgium 4%, Luxembourg 2.5%, the Netherlands 2.3% and
Germany 2.1%), and considering Europe as a densely populated continent saturated with
transport infrastructures, Spain has one of the lowest percentages of occupation, well below the
average of 1.3% (EEA, 2002). However, the progressive increase of the length of the roads,
particularly motorways and major trunk roads, and the development of new routes mean that
this percentage is increasing, and thus this land use is increasing, resulting in fragmentation of
the territory.
3. Land transformation for Spanish roads
The methodology used for the calculation of land transformation for Spanish roads is based on
studies of the impact on and degradation of biodiversity (Lindeijer, 2000a, 2000b; Spielmann et
al., 2004; Frischknecht, 2002) found in the occupied area per metre of network length and per
year, therefore the functional unit is m2/m·yr.
Data availability means that the temporal reference for the study must be the period of 14 years
from 1990 to 2004 (MFOM, 2005). Taking into account the increase of the road network during
this period, the percentage change was 5.75 % (from 156,172 to 165,152 km).
203
Area: Environment
Applying this value proportionally to the total occupation found above, the occupied area at the
beginning of the study was determined. Taking into account the values for the length of the
road networks and the period of time of the study, the value for transformation was thus
calculated.
Type of infrastructure
National data (m2/m yr)
European data (m2/m yr)
Motorways and major
trunk roads
0.307
0.173
Inter-urban roads
0.098
0.146
Tabla 5: Indicator of land transformation impact caused by Spanish roads per type of road and methodology used.
When performing LCA of the roads, this transformation value can be incorporated into the
environmental effects they cause, since it can be expressed as a function of the flow of vehicles
or unit of transport. Using this value, comparisons can be made in a suitable manner, since the
EEA suggests that environmental results be expressed as a function of this unit, in passengers
per km (pkm) and in tonnes per kilometre (tkm) annually. If we include the product of the
number of vehicles by the distance they travel in a year, we obtain the number of vehicleskilometre per year. Based on this data, the number of passengers-kilometre is the result of
multiplying the vehicles-kilometre travelled by cars, motorcycles and coaches (classified as
light) by the average occupation of these, not counting the driver of the coaches; and the
number of tonnes-kilometre is obtained by multiplying the vehicles-kilometre travelled of heavy
vehicles by their average load.
The impact values assigned for each type of road were based on the passing frequency of each
type of vehicle (light or heavy) and were obtained based on the following data:
• Traffic and transport by road type and vehicle type (MFOM, 2005).
• Total length of each road network (MFOM, 2005).
• Assignment factor of the intensity and passing frequency per type of vehicle (77% light and
23% heavy) (Spielmann et al., 2004).
The assignment factors obtained are shown in Table 5.
Assignment factor
Motorways and major trunk
roads
Inter-urban roads
m yr / pkm
4.44E-05
5.37E-04
m yr / tkm
1.40E-05
2.17E-04
Table 6: Assignment factors of the impact of the intensity and frequency of passing of the Spanish roads, by type of
road, in 2004.
Finally, with the values in tables 4 and 5, the transformation values as a function of vehicle
flows were obtained. Figure 2 shows these results according to the data source.
204
Area: Environment
90
Motorway s and m ajor t runk roads
I nt er-urban roads
80
m/p (t)km
70
60
50
40
30
20
10
0
Nat ional dat a (pk m)
European dat a (pkm)
Nat ional dat a (tkm)
European dat a (tkm )
Figure 2: Transformation of land caused by Spanish roads per type of road, per km and per data source.
3.1 Discussion of the results
Considering the results shown in the above figure, we can state that motorways and major
trunk roads are more sustainable than inter-urban roads. This is due to a much greater
circulation of vehicles and means that substantially less land use is required for the same unit
of transport.
However, comparing these values with those for other transport infrastructure at a European
level, railways require the least land use per unit of transport (e.g. area per pkm): the area
required per pkm for railway lines is approximately 3.5 times less than the area per pkm for
roads (EEA, 2002).
4. Fragmentation of the territory
The construction of transport infrastructure, especially the road network, is one of the main
anthropogenic causes of fragmentation in terrestrial ecosystems, and therefore, it is very
important to consider this when planning the development of a new road.
The impact that infrastructure has on natural ecosystems is not limited to physical occupation or
the direct effects on communities of flora or fauna (animals run over, fleeing, etc.). The
fragmentation of ecosystems caused, above all by the heaviest networks (motorways and major
trunk roads), is a well-documented cause of ecological degradation, directly related to the loss
of biodiversity and productivity. These infrastructures act as true barriers to fauna movements,
reducing the territory for development and genetic exchange of some species to such an extent
that it endangers their reproductive potential and their very survival. The consequences of
fragmentation by roads include the reduction of the net and functional area of the ecosystem,
205
Area: Environment
increased isolation and increased permeability to external impacts due to the proliferation of
marginal or border zones (Forman et al., 2002; PDSTIB, 2002; Delgado et al., 2004).
Another important area for consideration is the visual impact or impact on the countryside that a
road can cause, since its design introduces large elements with an artificial geometry that are
discordant with the natural forms in the territory. Moreover, the presence of areas bereft of
vegetation or the colour of the materials used in the road itself causes great chromatic contrasts
with the surroundings (PDSTIB, 2002).
The United Nations Convention on Biological Diversity considers the fragmentation of territory
as a significant threat to both biodiversity and species, as reflected in the Strategy on Biological
and Landscape Biodiversity, as well as in the European Commission’s Biodiversity Strategy
and in the European Commission’s Habitats Directive. The average size of contiguous units of
terrain not cut through by transport infrastructure in the European Union is 175 km2. In Spain,
this value is around 250 km2 due to low infrastructure density. Other countries have much lower
undivided areas, such as Belgium, Luxembourg and Germany, where the average undivided
area is below 50 km2 (EEA, 2002).
5. The concept of the ecological footprint
A very didactic indicator reflecting the significance of the emissions caused by transport and of
the difficulty of managing them can be obtained through the application of the concept of the
‘ecological footprint’ (Wackernagel & Rees, 1995). The ecological footprint of an activity (in this
case transport) is defined as the area of land necessary to provide the resources that this
activity uses and to assimilate the wastes that it generates. The ecological footprint of transport
would be, in terms of its emissions, the quantity of earth with plant cover needed to absorb all of
the CO2 generated by transport.
In the case of Spain, after prior study of the polluting emissions due to road transport (Garraín
& Vidal, 2007), we obtained an approximate value of 44 million hectares of forest needed to
absorb the more than 96 million tonnes of CO2 emitted during 2004. This area is equivalent to
approximately 40% of the total area of Spain, which means that a territory of almost the same
size as Spain would be required, uninhabited and completely covered in forest, to assimilate
current transport emissions.
6. Conclusions
The outstanding conclusions of this study are as follows:
The method of calculating the occupation of Spanish roads based on national data is the most
suitable and accurate, since it takes into account each type of road, width and above all, each
piece of legislation applied to road type and autonomous community. This value does not
exceed 0.9% of the total area of Spain, and thus Spain occupies one of the lowest rankings on
a European level, much below the average of 1.3%.
In terms of land transformation for roads, motorways and major trunk roads are more
sustainable than inter-urban roads. This is due to the much higher circulation of vehicles,
making the land use required much lower for the same unit of transport.
The fragmentation of land caused by infrastructure is very significant. In Spain the average area
of contiguous units of terrain undivided by transport infrastructure is around 250 km2 and this
206
Area: Environment
does not have as devastating an effect as in other European countries where the undivided
area is less than 50 km2.
The indicator of area of the ecological footprint caused by the CO2 emissions of transport in
Spain has a value equivalent to almost the whole of the national territory.
References
Antón, A., 2004. Utilización del análisis del ciclo de vida en la evaluación del impacto ambiental
del cultivo bajo invernadero mediterráneo, Tesis doctoral, Universitat Politècnica de Catalunya,
Barcelona.
Antón, A., Castells, F., Montero, J.I., 2007. Land use indicators in life cycle assessment. Case
study: The environmental impact of Mediterranean greenhouses, Journal of Cleaner
Production, 15, pp. 432-438.
CLC, 2000. CORINE Land Cover. Part 2: Nomenclature, European Environment Agency.
Cowell, S.J., 1998. Environmental Life Cycle Assessment of agricultural systems: integration
into decision making, PhD dissertation, Centre of Environmental Strategy, University of Surrey,
Guildford, UK.
Cowell, S.J., Clift, R., 2000. A methodology for assessing soil quantity and quality in life cycle
assessment, Journal of Cleaner Production, 8, pp. 321-331.
Delgado, J.D., Arévalo, J.R., Fernández-Palacios, J.M., 2004 Consecuencias de la
fragmentación viaria: efectos de borde de las carreteras en la laurisilva y el pinar de Tenerife,
Fernández-Palacios, J.M. & Morici, C. (Eds), Ecología Insular / Island ecology, Asociación
Española de Ecología Terrestre (AEET) – Cabildo Insular de La Palma, pp. 181-225.
EEA, 2000. Are we moving in the right direction? Indicators on transport and environmental
integration in the EU: TERM 2000, Environmental issue report n. 12, European Environment
Agency, 2000.
EEA, 2002, Paving the way for EU enlargement. Indicators for transport and environment
integration. TERM 2002, Enviromental issue report n. 32, European Environment Agency,
2002.
EUROSTAT, datos para Junio de 2006 (available on: http://epp.eurostat.ec.europa.eu).
Forman, R.T.T., Bissonette, J., Clevenger, A., Cutshall, C., Dale, V., Fahrig, L., Goldman, C.,
Heanue, K., Jones, J., Sperling, D., Swanson, F., Turrentine, T., Winter, T., 2002 “Road
ecology: Science and solutions”, Island Press, Washington, USA.
Frischknecht, R., 2002. Land occupation and land transformation in life cycle inventories,
Project Manager, ECOINVENT 2000, SETAC Europe Annual Meeting, Vienna, Austria.
Garraín, D., Vidal, R., 2007 Efectos nocivos sobre la calidad del aire provocados por las
carreteras españolas, Comunicaciones del XI Congreso Internacional de Ingeniería de
Proyectos, pp. 911-921, Lugo (España).
Goedkoop, M., Spriensma, R. The Eco-indicator 99. 2001 A damage oriented method for Life
Cycle Impact Assessment, Methodology report, Third Edition, PRé Consultants, Amersfoot, The
Netherlands.
207
Area: Environment
Guinée, J.B., 2002. Handbook on Life Cycle Assessment: An operational guide to the ISO
standards, Kluwer Academic Publishers.
Kloverpris, J., Wenzel, H., Nielsen, P.H., 2008. Life cycle inventory modelling of land use
induced by crop consumption. Part 1: Conceptual analysis and methodological proposal,
International Journal of Life Cycle Assessment 13 (1) pp. 13-21.
Köllner, T., 2001. Land use in product life cycles and its consequences for ecosystem quality,
Dissertation PhD 2519, Universität St. Gallen, Switzerland.
Köllner, T. 2000. Species-pool effect potentials (SPEP) as a yardstick to evaluate land-use
impacts on biodiversity, Journal of Cleaner Production, 8, pp. 293-311.
Köllner, T., Scholz, R.W., 2007 Assessment of land use impacts on the natural environment.
Part 1: An analytical framework for pure land occupation and land use change, International
Journal of Life Cycle Assessment 12 (1) pp. 16-23.
Köllner, T., Scholz, R.W., 2008. Assessment of land use impacts on the natural environment.
Part 2: Generic characterization factors for local species diversity in central Europe,
International Journal of Life Cycle Assessment 13 (1) pp. 32-48.
Ley 13/86, de 28 de noviembre, de Ordenación y Defensa de la Carreteras del Principado de
Asturias
Ley 16/1987, de 30 de julio, de Ordenación de los Transportes Terrestres.
Ley 2/89, de 30 de mayo, Reguladora del Plan General de Carreteras del País Vasco.
Ley 2/90, de 16 de marzo, de Carreteras de la Comunidad de Castilla y León.
Ley 2/91, de 7 de marzo, de Carreteras de la Comunidad Autónoma de La Rioja.
Ley 25/1988, del 29 de julio, de carreteras.
Ley 3/91, de 7 de marzo, de Carreteras de la Comunidad de Madrid, y del Reglamento de
Carreteras de la Comunidad de Madrid.
Ley 4/94, de 14 de septiembre, de Carreteras de Galicia.
Ley 5/1990, de 24 de mayo, de Carreteras de la Comunidad Autónoma de la Islas Baleares.
Ley 5/96, de 17 de diciembre, de Carreteras de Cantabria.
Ley 6/91, de 27 de marzo, de Carreteras de la Comunidad Valenciana.
Ley 7/93, de 30 de septiembre, de Carreteras de Cataluña, y del Reglamento de Carreteras de
Cataluña.
Ley 7/95, de 27 de abril, de Carreteras de Extremadura.
Ley 8/2001, de 12 de julio, de Carreteras de Andalucía.
Ley 8/98, de 17 de diciembre, de Carreteras de Aragón, y del Reglamento de Carreteras de
Aragón.
Ley 9/90a, de 28 de diciembre, de Carreteras y Caminos de Castilla la Mancha.
Ley 9/90b, de 27 de agosto, de Carreteras de la Región de Murcia.
Ley 9/91, de 8 de mayo, de Carreteras de Canarias, y del Reglamento de Carreteras de
Canarias.
208
Area: Environment
Ley Foral 11/86, de 10 de octubre, de Defensa de las Carreteras de Navarra.
Lindeijer, E., 2000a. Review of land use methodologies, Journal of Cleaner Production 8
(2000a), pp. 273-281.
Lindeijer, E., 2000b “Biodiversity and life support impacts of land use in LCA”, Journal of
Cleaner Production, 8, pp. 313-319.
Lindeijer, E., Kok, I., Eggels, P., Alfers, A., 2002 “Improving and testing a land use methodology
in LCA. Including case-studies on bricks, concrete and wood”, Ed: Dutch Ministry of Transport,
Public Works and Water Management (RWS DWW), The Netherlands.
Mattsson, B., Cedeberg, C., Blix, L., 2000 “Agricultural land use in life cycle assessment (LCA):
case studies of three vegetable oil crops”, Journal of Cleaner Production, 8, pp. 283-292.
MFOM, 2005 Ministerio de Fomento. “Anuario 2004”, Ministerio de Fomento de España.
Michelsen, O., 2008. Assessment of land use impact on biodiversity, International Journal of
Life Cycle Assessment 13 (1) pp. 22-31.
Milà i Canals, L., 2003. Contributions to LCA methodology for agricultural systems. Sitedependency and soil degradation impact assessment, PhD dissertation, Universitat Autònoma
de Barcelona, Spain.
Milà i Canals, L. Romanyà, J., Cowell, S.J., 2007 “Method for assessing impacts on life support
functions (LSF) related to the use of ‘fertile land’ in Life Cycle Assessment (LCA)”, Journal of
Cleaner Production 15 (2007) pp. 1426-1440.
Müller-Wenk, R., 1998. Land use - The main threat to species. How to include land use in LCA,
IWÖ-Diskussionbeitrag n.64, IWÖ, Universität St. Gallen, Switzerland, 1998.
Muys, B., García Quijano, J., 2002. A new method for Land Use Impact Assessment in LCA
based on the ecosystem exergy concept, Internal report, Laboratory for Forest, Nature and
Landscape Research, KU Leuven, The Netherlands.
PDSTIB, 2002. Recuperar el transporte público, mejor para todos, Gobierno de las Islas
Baleares, Consejería de Obras Públicas, Vivienda y Transportes, Plan Director Sectorial de
Transportes de la Islas Baleares, Palma de Mallorca.
RD1211/1990, Real Decreto, de 28 de septiembre, por el cual se aprueba el Reglamento de la
Ley de Ordenación de los Transportes Terrestres.
RD1812/1994, Real Decreto de 2 de septiembre, por el cual se aprueba el Reglamento
General de Carreteras.
Spielmann, M., Kägi, T., Tietje, O., 2004. Life cycle inventories of transport services, Final
report ecoinvent n. 14, Swiss Centre for Life Cycle Inventories, Duebendorf, Switzerland.
Van der Voet, E., 2001. Land use in LCA, CML-SSP Working Paper, Centre of Environmental
Science, Leiden University, The Netherlands, 2001.
Vogtländer, J.G., Lindeijer, E., Witte, J.P.M., Hendriks, C., 2004. Characterizing the change of
land-use based on flora: application for EIA and LCA, Journal of Cleaner Production, 12, pp.4757.
Wackernagel, M., Rees, W., 1995. Our Ecological Footprint: Reducing Human Impact on the
Earth, New Society Publishers, Gabriola Island, BC & Philadelphia, PA, USA.
209
Area: Environment
Wagendorp, T., Gulinck, H., Coppin, P., Muys, B., 2006. Land use impact evaluation in life
cycle assessment based on ecosystem thermodynamics, Energy, 31, pp. 112-125.
Weidema, B., Lindeijer, E., 2001. Physical impacts of land use in product life cycle assessment.
Final report of the EURENVIRON-LCAGAPS sub-project on land use, Department of
Manufacturing Engineering and Management, Technical University of Denmark.
Acknowledgements
This study has been carried out as part of the project “Indicadores de impacto y vulnerabilidad
de las infraestructuras de transporte” (Impact and vulnerability indicators for transport
infrastructure) funded by the Spanish Ministry of Public Works in the 24th Public Call for
Research Grant Applications, 2005, for infrastructure subjects (Order FOM/2376/2005).
Daniel Garraín
GID – Engineering Design Group.
Dept Mechanical Engineering & Construction, Jaume I University .
Av. Sos Baynat, s/n. E-12071 Castellón (Spain)
Tel. +34 964729252 Fax +34 964728106
e-mail: [email protected]
URL: http://www.gid.uji.es
Rosario Vidal
Dept Mechanical Engineering & Construction, Jaume I University.
Vicente Franco
Dept Mechanical Engineering & Construction, Jaume I University.
210
Area: Environment
THE NOISE IMPACT CATEGORY IN LIFE CYCLE ASSESSMENT
Garraín, D.
Franco, V.
Vidal, R.
Moliner, E.
Casanova, S.
Universitat Jaume I
Abstract
The Life Cycle Assessment (LCA) methodology is used to evaluate environmental impacts by
grouping the negative effects upon the environment of a given product or process into a
reduced set of impact categories. Global warming, ozone layer depletion, fossil fuel depletion
and acidification are the most typical of these categories. Unfortunately, reliable methodologies
are lacking for the assessment of some categories, as is the case of noise. Considering that
transport and housing are cornerstones of the world production system, traffic noise is one of
the categories that is likely to gain relevance in the near future, given its effects upon human
health. In our work, several studies about noise in the LCA methodology are analysed. Also, the
guidelines to include noise in the environmental assessment of products and processes within
the LCA methodology are presented. Finally, the DALY (Disability Adjusted Life Year) is
supported as the best unit to measure the negative impacts of noise upon human health.
Keywords: Noise, Life Cycle Assessment, Impact Category, DALY
1. Introduction
1.1 LCA and impact categories
In recent decades awareness of environmental issues has increased among the population and
this has led to the generation of strategies and methods for evaluating the impact on the
environment so that levels of pollutants can then be lowered.
One of the tools that is most widely accepted by the scientific community for evaluating
environmental impact is Life Cycle Analysis (LCA), an analytical procedure that assesses the
entire life cycle of a process or activity. According to standard UNE-EN ISO 14040:2006, LCA
“addresses environmental aspects and potential environmental impacts (such as the use of
resources and the environmental consequences of emissions) throughout the whole life cycle of
a product from the acquisition of the raw materials to production, use, final treatment and
recycling and finally disposal (that is to say, from the cradle to the grave)”. One clear advantage
of the methodology is that it makes it possible to detect situations in which one particular
system seems cleaner than another simply because it shifts the environmental loads to other
211
Area: Environment
processes or to a different geographic region, with no real improvement from a global point of
view (a phenomenon known as “problem shifting”) (Iglesias, 2005).
This method allows the composition and the amounts of the pollutants that are generated and
the resources that are consumed to be evaluated in terms of their impacts on the environment
by grouping them in a small number of environmental categories. The impact categories that
are most commonly considered in LCA of processes or products are the greenhouse effect, the
thinning of the ozone layer, the depletion of fossil fuels, acidification, eutrophication, human and
environmental ecotoxicity, tropospheric ozone precursors or emissions of heavy metals.
Unfortunately, to date, no reliable methods have been developed to analyse some categories,
such as the impact on land use, the visual impact or impact on the landscape, or the impact of
smells or noise. These last impact categories are not always taken into account or are simply
not really suitable for environmental impact assessments. If this is added to the scarcity of
available data, we find ourselves with a situation in which the application of indicators for these
categories is still a time-consuming, complex task due to the lack of agreement as to which
parameters are to be considered and the methodology to be followed.
Bearing in mind that transport or housing are cornerstones of the world productive system, one
of the categories that is likely to become one of the most significant in the future (given its
effects upon human health) will be noise from road traffic.
1.2 Effects of noise on human beings
Noise has become one of the major issues affecting people’s quality of life, especially in city
centres and in suburban areas that lie close to main roads, where the noise generated by
vehicles makes the problem even more important. In fact, it has been estimated that 80% of the
noise produced in cities can be attributed to motor vehicles. In our part of the world, the issue is
particularly serious because, in 1986, an OECD (Organisation for Cooperation and Economic
Development) report ranked Spain as the second noisiest country in the developed world after
Japan.
According to the WHO (World Health Organisation), noise can have negative effects on human
health when the equivalent levels exceed 65 dB(A) during the day and 55 dB(A) at night. High
levels of noise can have many physical and mental side effects on human beings, including
impaired oral communication, sleep disorders, increased levels of stress, damage to the
circulatory system and effects on balance, apart from obviously giving rise to hearing disorders.
It can also have negative effects on relationships with family and neighbours lower the selling
price of housing or affect people's fundamental rights to their own privacy or that of their family,
as well as the inviolability of the home. This situation has led to the introduction of strict
regulations in several European countries to limit the amount of noise the population is exposed
to, depending on the activity that is being performed and the time of day.
2. Noise in LCA
Most studies on environmental noise that have been conducted around the world focus on
quantifying or predicting it, on estimating the percentage of the population exposed to different
levels, or on describing its effects on people. Very few, however, attempt to establish a relation
between the emission of a particular type of noise and its real, measurable impact on human
beings.
212
Area: Environment
The Swiss professor Rudolf Müller-Wenk stands out as a reference for his studies on the
impact produced by noise from road traffic. His work (Müller-Wenk, 1999, 2002, 2004) has
enabled him to develop a methodology for quantifying the effect of noise on health, using the
DALY (Disability-Adjusted Life Years) as the unit of measurement, and to incorporate it into
LCA. Nevertheless, other authors, such as Doka (2003) or the Danish researchers Nielsen &
Laursen (2003), have developed other methodologies for assessing the real impact on health.
The main features of these methods are outlined in the following.
2.1 The Müller-Wenk methodology
The method developed by Müller-Wenk (1999, 2002, 2004) is based on the cause-effect chain.
This methodology consists in analysing any modification undergone by a variable (with a direct
effect on a pollutant) that is registered in the Life Cycle Inventory (LCI) and affects human
health. The procedure for creating this chain is made up of the following stages.
• The fate analysis, which describes the increase in concentration of the pollutant, in this
case acoustic (noise level), caused by changes in some variable registered in the LCI.
• The exposure analysis, which shows how many people are affected by such changes and
to what extent.
• The effect analysis, which describes the incremental effect on health that would occur if
human beings are exposed to a certain increase in the concentration of the pollutant (noise)
over a certain period of time.
• The damage analysis, which describes the total extent of the damage to human health that
is represented by the above-mentioned effects on health.
The road traffic noise model that is applied in the method is the one developed by SAEFL
(Swiss Agency for Environment, Forest and Landscape) (Balzari et al., 1998). This is a simple
model in which the noise emission from a road (LAeq) are determined by the noise from cars
(LE1) and from lorries (LE2) that, in turn, depends on the volume of traffic (N1 and N2), the
average speed (V1 and V2) and the slope of the road (i), in accordance with the following
equations.
(
LAeq = 10 ⋅ log 100.1⋅LE1 + 10 0.1⋅LE 2
)
(1)
where:
LE1 = E1 + 10 ⋅ log N1
(2)
LE 2 = E 2 + 10 ⋅ log N 2
(3)
E1 = max[{12.8 + 19.5 ⋅ log V 1}, {45 + 0.8 ⋅ (0.5 ⋅ i − 2 )}]
(4)
E 2 = max[{34 + 13.3 ⋅ log V 2}, {56 + 0.6 ⋅ (0.5 ⋅ i − 1.5)}]
(5)
Once the result has been obtained for the overall level of emission, the first step in the chain
consists in recalculating the previous value with the addition of, in this case, an increase that is
proportional to the initial value for the flow of vehicles. The difference between the two values of
overall levels of emission is ∆LAeq, which indicates the noise that is produced by adding a
proportional increase in the number of vehicles.
213
Area: Environment
The transport that is to be evaluated is not considered as a single isolated event, but rather as
a small part of the annual increase in traffic density on the whole network of roads within a
region or country. According to this author, there are statistics to show that the annual increase
in traffic on the different routes is, as a preliminary approximation, proportional to the level of
traffic from the previous year. The calculations, together with the theoretical considerations,
show that the value of ∆LAeq is more or less constant on all segments of the road network, with
small differences that can be attributed to different speeds of different vehicles and to the
properties of the road surface. In fact, the ∆LAeq due to an increase of one vehicle per hour is
more or less proportional to the first derivative of the logarithm of the number of vehicles (N),
which is inversely proportional (1/N). But if the increase in road traffic in each segment is
proportional to N, instead of a constant corresponding to a higher number of vehicles, the ∆LAeq
value is proportional to N multiplied by its reciprocal [N (1/N)]. It thus remains independent of N
and the same value is considered for roads with high and low volumes of traffic.
The second step consists in calculating the number of people exposed to excessive noise
levels. This was carried out by using a computer model to find the data on exposure to road
noise for the Swiss canton of Zurich (which accounts for approximately a sixth of the total
population of Switzerland). The findings for this area were then extrapolated to obtain the
exposure data for the whole Swiss population.
Later, subjective disturbance values were collected, in this case using surveys answered by
people exposed to traffic noise. In these questionnaires respondents were asked about the
extent to which they considered they were disturbed by noise and whether it impaired both
sleep (in order to measure night-time effects) and oral communication (in order to measure
daytime effects). Müller-Wenk comes to the conclusion that the approximate percentage of
persons who report that they suffer from sleep impairments increases linearly by 1.7% per dB,
starting at a night-time outdoor level of 46 bB. The conclusion drawn from the daytime
disturbance curve is that the approximate percentage of persons who report that they suffer
from communication impairments increases linearly by 2.5% per dB, starting at a daytime
outdoor level of 55 dB.
The last steps consist in quantifying the effect and the damage caused to human health. The
foregoing data on exposure and disturbance were used together with the so-called ‘disability
weights’ (DW) to obtain values for health damage in DALY units, in other words, the number of
years spent adapting to a disability (see point 4).
2.2 The Doka methodology
According to the Swiss author Gabor Doka (2003), no linear relationship exists between the
value of a noise in decibels and its effects on human health. The decibel is a logarithmic
measure of acoustic energy. As such, there is no single characterisation factor in LCA that can
be multiplied readily by a value in decibels to give a DALY.
The methodology proposed by this author has managed to adapt Müller-Wenk’s concept (1999,
2002, 2004) so as to be able to calculate the DALY resulting from noise generated by different
models of cars in Switzerland. To achieve this, reasoned approximations were taken to arrive at
a simplified formula to measure the damage, in DALY, per vehicle-kilometre (vkm, which is
obtained by multiplying the number of vehicles on the road network under consideration by the
distance travelled in a certain amount of time) depending on the emission noise measured in
decibels, according to equation 6.
214
Area: Environment
(a⋅L +b )
⎡ DALY ⎤
Damage ⎢
= K ⋅10 p
⎥
⎣ vkm ⎦
(6)
The terms are defined as follows:
•
•
Lp is the standard unit of measurement of noise, measured in dB.
a, b and K are regression parameters depending on the time of day in which the journey is
undertaken:
Parameter
Unit
a
b
K
1/dB
dimensionless
DALY
Average journey
(7% of vkm at
night)
0.099962
-6.243371
1.23406E-07
Daytime
journey
Night-time
journey
0.09998766
-6.3738654
7.60872E-07
0.999043
-5.5943622
2.30486E-07
Table 1: Values of the regression parameters of Doka's formula (Doka, 2003).
Different emission noise values can therefore be used to calculate the value of the DALY per
vkm, resulting in graphs like the one in Figure 1.
Figure 1: DALY per vkm depending on the decibels caused by rolling traffic on an average journey, during the day
and during the night (Doka, 2003).
2.3 The Nielsen and Laursen methodology
These Danish authors have focused their study (Nielsen & Laursen, 2003) exclusively on noise
that disturbs human beings while goods are being transported. The effects in areas that are
potentially more sensitive to noise (such as natural parks or recreational zones) and on animals
have not be taken into account to allow a clearer, simpler model to be developed. They also
considered other simplifications with regard to the distribution of noise and the quantification of
the extent to which it disturbs people.
215
Area: Environment
The noise model proposed by these authors can be represented diagrammatically as in Figure
2, which shows the isophones around a source, as well as the number of persons in each one.
The noise level decreases as the distance increases, due to the attenuation caused by the
divergence of the sound waves and absorption by the atmosphere. This reduction may also be
influenced by several factors such as the topography and the acoustic properties around the
source, the presence of walls or buildings, wind speed and direction, relative humidity,
temperature gradient, noise directionality, position of the source with respect to the recipients,
and so forth.
Figure 2: Model of the distribution of the population (x) in isophones at a distance d around a source of noise (o)
(Nielsen & Laursen, 2003).
The isophones in the previous figure only appear in flat, open landscapes when the atmosphere
is homogeneous. Moreover, in many situations the isophones are not circular and are shaped
by the different conditions of the moment. For the sake of simplicity, they assumed a circular
isophone model and a noise level given by simple mathematical formulae.
Noise Nuisance (NNd) at a specific distance (d) from a source-point can be defined in terms of
person-hours, according to equation 7.
NN d = Pd ⋅ T proc ⋅ NNFL p
(7)
The terms are defined as follows:
• Pd is the number of persons within a distance d from the source (this can be counted or
estimated).
• Tproc is the duration of the noisy process (in hours), that is to say, the time usually required
to produce a product or service unit, depending on the functional unit. It may be determined
by direct measurements or by calculations of the average.
• NNFLp is the specific noise nuisance factor for the current noise level, Lp being the relative
background noise (dimensionless). It represents the inconvenience caused by noise in
human beings and is a subjective parameter that is determined by aspects such as the
noise level, the composition of the frequency of the noise, the background noise and the
qualities and characteristics of each person, and so forth.
The following equation shows the relationship between noise and the specific nuisance
factor, in which the exponential factor expresses the part of the noise that exceeds the
background noise.
NNFL p = 0.01 ⋅ 4.22
(
0.1 L p − K
)
(8)
216
Area: Environment
The previous terms are defined as follows:
•
•
Lp is the noise level, which can be measured or calculated (dB).
K is the background noise relative to 20 µPa, also measured in dB.
The total noise nuisance caused by a specific process (NNproc) can be determined by the sum
of the nuisances for all the persons in each isophone, in accordance with equation 9.
NN proc = Tproc ⋅ ∑ Pd ⋅ 0.01 ⋅ 4.22
(
0.1 L p ( d ) − K
)
(9)
In this equation Lp(d) is the noise level at a certain distance d from the source.
Finally, the total nuisance (NNprod) is determined by summing up all the previous processes, in
accordance with the following equation.
NN prod = ∑ NN proc
(10)
This method can be used to calculate the noise due to goods transportation by road and rail.
With a series of modifications, it can also be employed to calculate the nuisance caused by
noise from other sources such as industry, loading, building works and sea or air transport.
2.4 Discussion of the methods
Of the methods described above, the one proposed by Müller-Wenk (1999, 2002, 2004) is the
most commonly cited. Despite apparently being difficult to use at first, in fact it greatly simplifies
the tasks of determining the increase in cases affected by a rise in initial traffic flow and
calculating the DALY, thanks to the constants that have been determined. This method is very
useful for obtaining generic overall impacts of noise, regardless of the route followed, that can
be applied to large areas such as an entire country. Nevertheless, the method has several
aspects that need to be improved, such as the noise emission model, which is a little obsolete,
and it should also take into account other effects on human health.
Doka's method (Doka, 2003) is quite practical because, by obtaining adjustment parameters, it
establishes a direct relation between the harmful effects on health and the noise produced by
traffic. Drawbacks of the method include the fact that, as it is based on the one by Müller-Wenk
(1999, 2002, 2004), it possesses the same features that are in need of improvement, as well as
being applicable over a very restricted range because it only considers the Swiss population.
Nielsen & Laursen (2003) also provide a method that is simple to apply to vehicle traffic
because it takes into account only the population density, the distance to the centre point, noise
emission and the process time, and from these data it calculates the number of persons
affected. This method does not consider the subjectivity of potentially affected persons, since it
does not take into account the degree to which individuals are disturbed.
3. Guidelines for incorporating the effects of noise into LCA
The purpose of studies aimed at incorporating the category ‘noise’ into LCA must be to analyse
the disturbance caused from the product-oriented point of view. This will allow the noise
nuisance to be taken into consideration as an environmental aspect in the development of
products. The environmental behaviour of products and services could also be compared when
conducting an LCA. Thus, in the future, noise will be assessed and taken into account on the
same level as any other impact category.
217
Area: Environment
We therefore believe it is wise to consider the cause-effect chain as the basis for incorporating
this category into LCA. To achieve this, it is necessary to start out with the data on noise
emission available for different types of vehicles. This is used to model a flow of vehicles that
simulates the entire vehicle fleet and a virtual network of roads with a virtual population
distributed around the roads. The model must be up-to-date and has to estimate the noise of a
flow made up of different types of vehicles (mopeds, motorcycles, lorries, vans, buses, cars,
etc.), in terms of their speed and the flow of traffic. Furthermore, other adjustment parameters
must also be considered, such as the environmental temperature, acceleration and
deceleration, the slope of the road, humidity, type of road surface, type of tyres, type of engine,
the use of studs, illegal exhaust systems, and so forth. Introducing these adjustment
parameters makes it possible to model the virtual road network and its virtual vehicles with the
same characteristics as the roads under study, thereby simulating the vehicle fleet. One of the
most widely accepted models in Europe for this purpose is that of the IMAGINE project
(IMAGINE, 2007).
In a later stage a second flow of vehicles has to be modelled with a small increase (compared
to the initial level) and the noise levels are recalculated without varying the initial conditions.
This operation has to be repeated for each of the stretches of virtual road, which results in an
overall increase in noise that can be attributed to the increase in traffic. In this way the
differences between levels can be used to quantify the effect of this increase. It must be noted
that, because it is an incremental model, the conditions under which the noise is transmitted are
no longer a problem because they are the same in both the initial and the final situation.
In order to determine the impact of the increase in noise on human health, it is necessary to
have access to data about the population frequency distribution with respect to the different
levels of noise. The distribution of this frequency can be determined, in principle, by using a
combination of strategic noise maps from roads and geographic data about the population
density within the areas under study.
Once the population exposed to excessive levels of noise has been quantified, all that remains
to be done is to calculate how this exposure affects those who experience it. This must be
achieved by means of surveys and population studies that establish a relation between the
number of persons who report that the exposure is an “important disturbance” and a particular
level of noise. The graphs that represent this relation allow psychological aspects of noise to be
introduced. They also have the advantage that the more they tend to follow a straight line the
easier it is to determine high-disturbance traffic noise.
Finally, the DALY, which depends on the above-mentioned DW, is the unit that must be used to
quantify the negative effects of noise. This unit of measurement is internationally recognised
and recommended by the WHO, and its chief characteristics are defined in the next section.
4. The DALY
In the early 90s, the concept of the unit known as the DALY started to be developed in
opposition to the QALY (Quality Adjusted Life Years). After long reviews and discussions at an
international level and based on the findings of a study aimed at quantifying the overall burden
of a disease on human life, Murray & López (1996) published “The global burden of disease”,
where they laid down the foundations for defining and calculating this unit.
A DALY can be defined as a variant of the QALY that expresses the number of life-years lost
due to premature death and the number of years spent with a disability that has a specific
218
Area: Environment
severity and duration. A DALY is therefore a healthy life-year that has been lost (Seuc et al.
2000). This unit is the one chosen and recommended by the WHO to quantify the weight of
diseases and their sequelae in human populations.
Years lost due to premature death are calculated by subtracting the age of death from the life
expectancy. Years spent with a disability are calculated from the moment the disease begins
until it ends, using the conversion factors known as DW.
DW are listed by the WHO for each disease category depending on how severe the associated
damage is and they act as factors that compare the weight of any disability with death. Thus,
they are measured on a scale that goes from zero to one, where zero means optimal health
and one means death. They are recorded in tables drawn up by the WHO for each type of
disease. Table 2 shows some examples of diseases and the corresponding range of DW.
DW range
Diseases as indicators of conditions
0.00 – 0.02
Mild obesity, facial marks
0.02 – 0.12
Diarrhoea, anaemia, bad sore throat
0.12 – 0.24
Fracture of the radius, infertility, tonsillitis, erectile dysfunction
0.24 – 0.36
Dumbness, below-the-knee amputation
0.36 – 0.50
Down’s syndrome, mild mental retardation, rectovaginal fistula
0.50 – 0.70
Severe depression, blindness, paraplegia
0.70 – 1.00
Lung cancer, active psychosis, dementia, severe migraine
Table 2: Evaluation of the disability (DW) resulting from some diseases (Murray & López, 1996).
The WHO database does not include DW data produced by noise and, hence, several authors
have conducted studies and surveys to obtain coherent data quantifying the damage produced
by noise in human beings:
• Müller-Wenk (1999, 2002, 2004) conducted a survey involving 41 physicians and
psychologists. His findings (obtained by statistical processes and interpolations) provided
DW values for the most significant effects of noise, which were found to be the disturbance
caused by impaired sleep (night-time) and impaired communication (daytime), with values
of 0.055 and 0.033 respectively.
• Within the LCA framework, Meijer (2006) carried out a study on improvements to the quality
of buildings by using materials with better soundproofing specifications. To be able to
perform comparisons on the effects of noise in human beings, this author uses the DW
provided by Müller-Wenk (1999, 2002, 2004).
• Westerberg & Glaumann (2002) conducted an analysis of health risks in buildings and
outdoors and used their findings to draw up a table of values; one of these problems
affecting comfort was outdoor noise, which ranged from 0.01 to 0.05.
5. Conclusions
The most notable conclusions of this study are as follows:
219
Area: Environment
Because transportation now plays a key role in the worldwide system of production and is
continually expanding, the category of noise has to be included in assessments of the
consequences of vehicles on the environment and their effects on people’s health.
The cause-effect methodology is the ideal procedure to be able to include the effects of
noise on people when evaluating environmental impact with methods like LCA.
The DALY is the unit of measurement that offers the best characteristics for quantifying the
effects of noise on health because, in addition to being recommended by the WHO, it is also
simple to calculate. Furthermore, despite the complexity involved in attempting to measure
the state of health in a population, it is also an indicator that is easy to interpret. The DW
related to the effects of noise, however, have yet to be tabulated, but can easily be predicted
either by conducting comprehensive surveys with experts or by making use of previous
studies.
References
Balzari & Schudel, Grolimund & Peterman, 1998. Modèle de bruit du traffic routier dans les
zones habiteés, CDE report n. 15, SAEFL, Berne, Switzerland.
Doka, G., 2003. Ergaenzung der Gewichtungsmethode für Oekobilanzen Umweltbelastungspunkte'97 zu Mobilitaets - UBP'97, Doka LCA, comisionada por ASTRA, Switzerland,
(download from http://www.doka.ch/mubp/AstraMUBP97DOKAv2.pdf).
Iglesias, D.H., 2005. Relevamiento exploratorio del análisis del ciclo de vida de productos y su
aplicación en el sistema agroalimentario, Contribuciones a la Economía, (Full text at
http://www.eumed.net/ce -- last accessed November 2007).
IMAGINE., 2007. The noise emission model for European road traffic, Deliverable n. 11 of the
IMAGINE (Improved Methods for the Assessment of Generic Impact of Noise in the
Environment) Project, (available at http://www.imagine-project.org).
Meijer, A., 2006. Improvement of the life cycle assessment methodology for dwellings, Thesis
Universiteit van Amsterdam, The Netherlands, ISBN 1-58603-690-4.
Müller-Wenk, R., 2004. A method to include LCA road traffic noise and its health effects,
International Journal of Life Cycle Assessment 9 (2), 76-85.
Müller-Wenk, R., 2002. Attribution to road traffic of the impact of noise on health, Environmental
series n. 339, SAEFL, Berne, Switzerland.
Müller-Wenk, R., 1999. Life-cycle impact assessment of road transport noise, IWÖ, Universität
St. Gallen, Switzerland.
Murray, C.J.L., López, A.D., 1996. The global burden of disease. A comprehensive assessment
of mortality and disability of diseases, injuries and risk factors in 1990 and protected to 2020,
Harvard University Press, Cambridge, UK.
Nielsen, P.H., Laursen, J.E., 2003. Background for spatial differentiation in LCA impact
assessment - The EDIP2003 methodology – Part 9: Integration of external noise nuisance from
road and rail transportation in life-cycle assessment, Danish Environmental Protection Agency,
Denmark.
220
Area: Environment
Seuc, A.H., Domínguez, E., Díaz-Díaz, O., 2000. Introducción a los DALYs, Rev Cubana Hig
Epidemiol ; 38 (2): 92-101.
UNE-EN ISO 14040:2006. Gestión ambiental – Análisis de ciclo de vida – Principios y marco.
Westerberg, U., Glaumann, M., 2002. Weighting health risks in buildings and outdoor
environment, Sustainable Building 2002, Oslo, Norway.
Acknowledgements
This study was conducted as part of project C46/2006, entitled “Desarrollo de categorías de
impacto aplicadas a materiales cerámicos usando la metodología del análisis del ciclo de vida”,
funded by the Spanish Ministry of Public Works and Transport.
Daniel Garraín
GID – Engineering Design Group
Dpt. Mechanical Engineering &Construction, Universitat Jaume I
Av. Sos Baynat, s/n. E-12071 Castellón (Spain)
Tel. +34 964729252 Fax +34 964728106
Vicente Franco
Rosario Vidal
Enrique Moliner
Sonia Casanova
221
Area: Environment
CREATION OF OZONE MAPS TO REGISTER THE EVOLUTION OF
THIS POLLUTANT USING GEOSTATISTICAL TECHNIQUES
López-Rodríguez, F.
Moral, F.J.
Valiente, P.
Pinilla, E.
Abstract
Ground-level tropospheric ozone is one of the air pollutants of most concern. It is mainly
produced by photochemical processes involving nitrogen oxides and volatile organic
compounds in the lower parts of the atmosphere. Ozone levels become particularly high in
regions close to high ozone precursor emissions and during summer, when stagnant
meteorological conditions with high insolation and high temperatures are common.
Tropospheric ozone levels in Europe continue to exceed both target values and the long-term
objectives established in EU legislation to protect human health and prevent damage to
ecosystems, agricultural crops and materials. In this work, some results of urban ozone
distribution patterns in the city of Badajoz , which is the largest (140.000 inhabitants) and most
industrialized city in Extremadura region (southwest Spain) are shown.
Keywords: Ozone, Geostatistics, Geographic Information System, Variogram, Kriging, Map.
Introduction
The environmental policy is an issue which attracts an important attention in the European
Union and, particularly, in Spain, due to the increasing alarm that economic development
causes on human health and security, and the worrying events as the Chernobyl disaster, acid
rain, and at worldwide level, greenhouse effect or destruction of the ozone layer.
Today´s society worry for nature and its progressive degradation, due to pollution, has as a
consequence that people are demanding a less aggressive way of life for the environment,
claiming clean industries, ecological produces, etc. Citizens also demand to their governing
class different measures and facts to benefit the environment where they live, favoring a better
life quality. Moreover, from a planning point of view, future proceedings in urban areas should
consider distribution patterns of pollution.
For the particular case in the city of Badajoz, southwestern Spain, the incidence and the spatial
distribution of the atmospheric pollution was studied. Therefore, some geostatistical techniques
were used to properly analyze and characterize the spatial distribution of the tropospheric
ozone, which is taken as indicative of the pollution level, that is, an excessive level of ozone is
222
Area: Environment
indicative of high pollution. Later, this information was incorporated in a geographical
information system (GIS) to produce accurate ozone maps, where a continuous graphic
representation of the pollutant is shown.
1. Materials and Methods
The first stage of the work was to take samples in the city. Thus, 138 urban locations were
chosen as sample points, covering the majority part of the city and taking into account its
different characteristic, as inhabitants density, type of streets or roads, etc. Sampling interval
was not uniform, ranging from, approximately, 75 to 500 m. This favours the geostatistical study
and, in consequence, the interpolation and mapping process. An automatic portable analyzer,
based on UV absorption, was used to obtain air ozone concentration, in parts per billion by
volume (ppbV).
14 sampling campaigns, one per month (two in August), were carried out between May 2007
and August 2007. All noise measurements were made on working days and under suitable
meteorological conditions (no cloudy days). Ground ozone levels were measured between 4
p.m. and 10 p.m. for all sampling campaigns, because during that time span maximum ozone
concentrations occur.
Therefore, the final data set consists of ground-level ozone measurements from 138 locations
situated throughout Badajoz. For each location, its geographic coordinates were ascertained
using a GPS device. From this information, using the GIS software ArcGISG (version 9.2) and
its extension Geostatistical AnalystG, the geostatistical study was carried out and different
maps were generated to visualize the spatial distribution of the variable in the experimental
area, that is, the city of Badajoz.
2. Geostatistical Estimates
Geostatistics can be defined as the set of tools and techniques to analyze the spatial patterns
and predict at unsampled locations the values of a continuous variable distributed in space or in
time. It is also denominated spatial statistics, due to its direct application to GIS.
All geostatistical study has to fulfill 3 stages (e.g. Isaaks y Srivastava, 1989):
1. Exploratory analysis of data. Geographical distribution of data is not taking into account.
Statistics was applied to check data consistency, removing outliers and identifying
statistical distribution where data came from.
2. Structural analysis of data. Spatial distribution of the variable was analyzed. Spatial
correlation or dependence can be quantified with semivariograms (or variograms).
These function relate the semivariance, half the expected squared difference between
paired data values Z(xi) and Z(xi+h), to the lag distance, h, by which sample points are
separated. When a experimental variogram is defined, i.e. some points of a variogram
plot are determined by calculating variogram at different lags, a model (theorical
variogram) should be fitted to the points.
3. Predictions. The main objective of a geostatistical study is to get estimates of values of
the studied variable at unsampled locations, considering the spatial distribution pattern
and integrating information from sample points and observed or known trends, if they
223
Area: Environment
exist. Simulations can also be carried out, taking into account the chosen spatial
patterns.
Usually, the variogram is the function used to model the spatial variability. For discrete
sampling locations, the function is estimated as:
1 N(h )
{Z( x i ) − Z( x i + h )}2
γ (h ) =
2 N ( h ) i =1
(1)
∑
where γ(h) is the experimental semivariance value at distance interval h, Z(xi) are the measured
sample values at sample points xi , in which there are data at xi and xi+h; N(h) is the total
number of sample pairs within the distance interval h. For irregular sampling, h is represented
by a distance band because the distance between the sample pairs to be exactly equal to h is
very rare.
Geostatistics offers a great variety of methods that provide estimates for unsampled locations.
These methods are known as kriging, in honor of Danie Krige, who first formulated this form of
interpolation in 1951. Kriging is regarded as the best linear unbiased estimator (BLUE), which is
a process of a theoretical weighted moving average:
n
Ẑ( x 0 ) = ∑ λ i Z( x i )
i =1
(2)
Ẑ(x )
0 is the value to be estimated at the location x , Z(x ) is the known value at the
where
0
i
sampling place xi, n is the number of the closest samples used for estimation, and the weights
for sample values, λi, are calculated based on the parameters of the variogram model. The sum
of all weights must be one due to the necessity for ensuring that estimates are unbiased.
The main characteristics that make kriging superior to other traditional interpolation techniques,
as inverse distance weighting, triangulation, etc., have been described elsewhere (e.g.,
Goovaerts, 1997; Moral, 2003).
The more important idea of kriging is based on the theory of regionalized variables (see, for
instance, Isaaks y Srivastava, 1989), which show spatial autocorrelation such that samples
close together in space are more alike than those that are further apart. All different types of
kriging are distinguished depending on the chosen model for the trend of the random function
(e.g. Goovaerts, 1997). In this work, the geostatistical interpolation method known as ordinary
kriging was used. This procedure considers that the mean fluctuates locally; thus, stationarity is
limited to local areas. Deutsch and Journel (1992) described ordinary kriging as the anchor
algorithm of geostatistics because of its robustness under different conditions.
3. Results
From measured values of tropospheric ozone, for all sampling locations (Figure 3), a map,
where estimated values of that pollutant are shown everywhere in the experimental area with
an adequate accuracy, should be generated. A complete geostatistical study was performed to
achieve the objective. Thus, during the phase of exploratory analysis of data, histograms and
224
Area: Environment
normal QQplots for each sampling campaign indicate normality: the shape of the histograms
looks bell shaped and the points of these plots are located close the 45º line (Figure 1 contains
the histograms and QQplots for one sampling campaign, which is similar for all others).
Moreover, descriptive statistics (Table 1) confirm the normality of the data distribution. The fact
that mean and median are very similar and the skewness values are near zero is also indicative
of data coming from a normal distribution
Frequency
28
22,4
16,8
11,2
5,6
0
2,52
2,68
2,84
3
3,16
3,32
3,48
Data 10
3,64
3,8
3,96
4,12
-1
Figure 1: Histogram (left) and QQplot (right) for data, ground-level ozone (ppbV), corresponding to the 17th-19th of
July 2007 sampling campaign.
Table 1: Statistics of the ground-level ozone measurements made in 138 points of the city and for all
sampling campaigns.
May 21-24
June 25-28
July 17-19
August 8-10 August 18-20
Mínimum (ppbv)
25.90
26.20
25.20
35.00
27.00
Maximum (ppbv)
43.20
45.70
40.70
47.00
42.00
Mean (ppbv)
33.13
36.21
33.21
39.73
36.41
Median (ppbv)
33.10
36.70
33.10
39.00
37.00
Standard Deviation (ppbv)
4.25
4.32
3.30
2.57
2.93
Skewness
0.04
-0.42
-0.04
0.62
-0.67
Kurtosis
2.02
2.62
2.65
3.21
3.39
Experimental variograms were determined assuming isotropy conditions because there were
no reasons to justify the consideration of anisotropy and, what is more important, with 138
sample points, the influence of different directions in space had supposed the impossibility to
define acceptable directional variograms. Therefore, spatial correlation does not depend on
directions and experimental variograms were calculated with a directional tolerance of 360º
(omnidirectional). When the experimental variogram was calculated, a theoretical variogram
was fitted to their points. It is known how the choice of a particular variogram model implies a
belief in a certain kind of spatial variability. Possibly, a variable like ground–level ozone is not
225
Area: Environment
evenly distributed in reduced distances. In these cases, exponential and spherical models are
the most suitable (e,g. Isaaks y Srivastava, 1989); the spherical ones were finally chosen.
Finally, during the most important stage of a geostatistical work, the estimation, the ordinary
kriging method was used (e.g. Moral, 2003). A grid, constituted of 40 m side square cells, was
designed and superimposed on the city, and estimates were conducted at center of the cells,
i.e., ozone concentrations were estimated at a spatial resolution of 40 m throughout Badajoz.
The number of observations (neighbours) that were used to estimate the value at each location
is at least the closer 15 sample points. From the estimated values, the distribution of ozone
levels in the city of Badajoz can be mapped (Figure 3).
Figure 2: Experimental variograms (red points) and theoretical spherical variograms (yellow lines) for data
th
th
th
corresponding to the 25 -28 of June (left) and 8th-10 of August (right) 2007 sampling campaigns.
One important advantage of the geostatistical interpolation techniques with respect to other
estimation methods is the possibility of obtaining the reliability of estimates. Thus, together with
the estimated value, another output of kriging can be obtained for each location: the kriging
variance, or its square root, the kriging standard deviation (KSD). KSD can be mapped similarly
to estimates, giving an idea of the quality of the estimates at different places. However, these
maps should be used with caution because the reliability of kriging depends on how accurately
the variation is represented by the chosen spatial model. Thus, if the nugget effect is
overestimated, our estimates could be more reliable than they appear. In general, areas with
many sample points or areas where data were sparse but evenly distributed had the most
reliable estimates.
Another interesting application of geostatistics related to ground-level ozone studies is the
generation of probability maps (Goovaerts, 1997), which are based on the combination of
kriging map and KSD map. For example, if ozone levels higher than 42 ppbV are not
considered optimum, according to the regional directive, areas which are likely to surpass that
threshold can be delimited. Thus, Figure 4 shows the probability maps corresponding to the
25th-28th of June and 8th-10th of August 2007 sampling campaigns, in which areas with high risk
of ground-level ozone exceeding the proposed limit are represented, with the probabilities
providing a measurement of confidence for hazard assessment of ozone concentration. Areas
with low probabilities, for example <25%, could be regarded as “clean” zones where the ozone
level is unlikely to be higher than 42 ppbV and, on the other hand, areas with high probabilities,
for example >50%, could be regarded as “dangerous” zones where the noise level is very likely
to be higher than 42 ppbV.
226
Area: Environment
4. Conclusions
In this work, using geostatistical techniques, the double objective of characterizing the spatial
variability of ground-level ozone and including this information in the interpolation algorithm is
achieved; it would not have been possible if other interpolation methods had been applied.
The proposed techniques provide some reliable surfaces at enough spatial resolution to
correctly visualize the spatial patterns of this pollutant. Considering different sampling
campaigns, the temporal evolution of tropospheric ozone can be analyzed.
Polluted areas in the city have to be delimited. Future actions against ozone should be
particularly aimed at reducing the high levels in these zones. Consequently, the ozone maps
can influence decisions concerning air-quality policy.
227
Area: Environment
Figure 3: Kriged maps of ground-level ozone in Badajoz city for the 25th-28th of June (up) and 8th-10th of August
(down) 2007 sampling campaigns. Scale unity is ppbV.
Green points correspond to sampling locations.
228
Area: Environment
Figure 4: Probability maps of ground-level ozone higher than 42 ppbV in Badajoz city, for the 25th-28th of June (up)
th
and 8th-10 of August (down) 2007 sampling campaigns. Scale unity is probability.
229
Area: Environment
References
Deutsch, C.V., y Journel, A.G. 1992. Gslib: Geostatistical Software Library and user’s guide,
Oxford Univ. Press, New York.
Goovaerts, P., 1997. Geostatistics for Natural Resources Evaluation, Oxford Univ. Press, New
York. 1997.
Isaaks E.H., and Srivastava, R.M. 1989. An Introduction to Applied Geostatistics, Oxford Univ.
Press, New York.
Moral, F.J. 2003. La Representación Gráfica de las Variables Regionalizadas. Geoestadística
lineal. Servicio de Publicaciones de la Universidad de Extremadura, Cáceres.
Fernando López Rodríguez
Phone: (+34) 649833585
E-mail de contacto: [email protected]
Empresa / Institución: Universidad de Extremadura
230
Area: Environment
SPATIAL MULTISCALE MODEL FOR THE FOREST AREAS OF
PLANNING: AN EXAMPLE IN THE REGION OF EASTERN MARINE
(GALICIA NE)
Díaz Varela, E. R.
Rodríguez Vicente, V.
Marey Pérez, M. F.
Abstract
Development in recent years of new trends in forestry planning - in the framework of
Sustainable Forestry Planning - has acknowledged the need to widen the scope of action to
one with multiple scales or functional levels in forestry areas. It also required from the
compatibilization of different objectives of forestry production, corresponding with the multifunctional possibilities of the forest, oriented to the economical, social-cultural and ecological
sustainability of forestry systems. In this work an application of spatial strategies at multiple
scales is proposed, with the goal of establish a coherent compartimentation of the territory for
forestry activities. With this objective, successful examples of spatial planning are analyzed, in
order to identify the key spatial concepts that ensure their successfulness. Next, a multi-scale
structure is developed responding to the dynamics and spatial arrangement of land uses in a
shire in NE Galicia (Spain). As a result, a reference model is obtained for forestry planning,
adapted to the characteristics and needs of land use in the study area, and which interconnects three spatial levels. micro-, meso- and macroscale.
Keywords: Sustainable Forestry Planning; multi-scale; spatial concept; land use; land use
planning.
1. Introduction
The sustainable forest management seeks the multifunctionality of forest ecosystems and
landscapes, assuming that they can assume a variety of functions: to the basic productive
function (wood and wood products, plus other alternative uses) is joined to the carbon
sequestration, protection aquifer, habitat for plant and animal species, biodiversity, recreation,
social uses, cultural identity, etc. (Von Gadow et al, 2000). In order to maintain or increase
such multifunctionality, several factors must be taken into account. Among them, the search for
spatial complexity of forest landscapes has been identified as a critical factor that can have
positive effects on the ecological character of the forest areas (ecosystem stability, diversity of
species ...), as in the productive (increased production, resistance to pests ...). Consequently,
the inclusion in forest planning strategies to increase the spatial complexity is of great
importance. However, it is precisely the definition of two components: first, a spatial oriented
231
Area: Environment
design to optimize the different functions of the forest; on the other, the ability of actual
implementation of normative instruments and management.
The development of an optimal spatial design can take the form of a spatial concept (Van Lier,
1998) which orientates in a strategic and precise way the actions to specific points in the
territorial system. Such concepts should not only serve as static references for planning, but
allow the development of strategies that allow the development of agroforestry activities, while
avoiding the removal of elements of ecological importance with the evolution in land uses. But
for the application in the real world of a spatial concept is successful, it must include the
distinction between different levels or spatial scales with different functional characteristics,
depending on the objectives of the planning, and development to provide tailored solutions to
the specific properties of such level. This utility to consider several hierarchy levels in the
planning was highlighted in the specialized literature (Dramstad et al, 1996; Mander et al,
2003).
2. Objectives
This work seeks the development of spatial strategies to multiple scales with the goal of
establishing consistent different territorial areas in forest planning. The basis for this
development is, on the one hand, the analysis of successful examples in spatial planning, to
obtain spatial concepts of general interest that can be combined in a multiscale strategy of
planning. Moreover, the differentiation of scales or levels of planning from the structural
analysis of the landscape. The subsequent analysis of the dynamic of land uses in the
considered area, will allow determining the necessary concepts for planning at each level.
3. Methodology and/or Case of study
3.1. Study area
The study area is the region of Eastern Marine in the northeastern part of the Autonomous
Community of Galicia (Spain, EU). This region presents a great diversity in their geographical
characteristics due to a marked gradient interior-coast, as well as imbalances between
agricultural and forest areas with different intensities of production. This diversity is reflected in
the landscape pattern, providing an interesting source of study for planning at different levels in
different areas.
Figure 1: Location of the study
area (Region of Eastern MarineGalicia).
232
Area: Environment
3.2. Identification of appropriate scales for planning
The identification of reference scales for planning in the study area of previous works (Díaz
Varela, 2005; Díaz Varela & Crecente, 2007), through an analysis of spatial heterogeneity done
on digital mapping of land cover in raster format. Such mapping has been obtained from
interpretation of aerial photographs 1:18.000 scale and post field tests (Marey, 2007), and then
processed with software ArcView 3.3 ®. The pixel size is 10x10m, and the Minimum Area
Mapping of 400 m2. On the same, the Shannon-Wiener index was applied (Shannon & Weaver,
1949) by mobile windows of different sizes, allowing generating maps of spatial distribution of
landscape heterogeneity in the study area at different scales. For this, it has used the software
FRAGSTATS (Mcgarigal et al, 2002). As a result, it is identified three scales of interest:
"microscale", consisting of areas of change in the heterogeneity of the landscape in small areas
(between 4 and 40 ha), inserted into larger units, which define a "mesoscale" and that divided
the study area in different areas for its landscape heterogeneity (see Figure 2). A third level,
"macroscale", defines the regional context of the study area, which can be defined by working
on such levels. The multiscale model of planning will answer, therefore, these three levels.
Figure 2: Landscape differentiated units to macro-, meso and microscale. The units to mesoscale are divided into
"A", homogeneous mosaic. "B", agroforestry homogeneous mosaic. "C" agroforestry heterogeneous mosaic. "D",
large seminatural stain. E1 and E2, marshes. "b" and "c" correspond, respectively, to small heterogeneous and
homogeneous areas in mosaic the contrary trend in terms of heterogeneity.
233
Area: Environment
3.3. Correspondence between scales of planning and forest planning instruments
The units and scales identified in the different levels of planning must have a Correspondence
with planning instruments currently in the study area. In Galicia, and with special emphasis on
forest planning, can perform the following identification of instruments:

Microscale: Units of forest management (UXFOR) / Forest Management Plan

Mesoscale: Plan Management of Forest Resources (PORF)
The instruments to macroscale must be dependent of determinations of a larger spatial ambit,
such as Territorial Development Guidelines, which are not currently approved in Galicia.
The UXFOR, established by autonomic law 3/ 2007, "Prevention and Protection against Forest
Fire of Galicia", are defined as forest management units requested by any number of forest
owners in excess of 50% of the area of forest contiguous plots with a minimum extension of 15
ha. This type of groupings has by objectives to avoid the process of leaving the forest property,
the creation forest economies with a minimum scale to return profitable the forestry activity and
fire-fighting and finally the promotion and multifunctional enhancement of forest.
Management Forest Plans under the Forestry Law 43/2003 with the structure established in the
General Guidelines for Management of Forest Trees (IGOMA) adopted in 1970 which is the
main instrument for sustainable forest intrinsic management to the forest Spanish and thus
Galician forest that achieving an adequate scale that in any case exceed 50 ha.
The PORF are defined in state Law 10/2006, of 28 April, amending the Law 43/2003 of
November 21, of forests. This law established the need for forest planning on a global scale,
creating the existence of the Spanish Forest Strategy and Spanish forest Plan. In this area, the
most important innovation of the law is the plans for the management of forest resources
(PORF). They are configured as forest planning tools of regional space integrated in the
framework for regional planning, so that the forest management and planning are connected
with the crucial area of territorial management. In Galicia in 2002 took out the first proposal of
PORF in Spain (Marey et al, 2004).
3.4. Generating strategies spatial of planning
Identification of spatial concepts of interest
A detailed analysis of spatial concepts such as ecological networks (Cook Van Lier,1994;
Jongman, 1995; Jongman et al, 1995), the territorial systems of ecological stability (Petch &
Kolegka, 1995; Miklös, 1996; Sustek, 1998), the concept framework (Van Langevelde, 1994;
Vrijlandt & Kerkstra, 1994; Van Lier, 1998), or the spatial solution (Forman, 1995a; Forman,
1995b; Forman & Collinge, 1996; Forman & Collinge, 1997) can distinguish a generic set of
common general principles, that underlie the basis for application. These could be summarized
as follows:

Ecological Stabilization: Protection of the ecological function in natural and seminatural
ecosystems, protection areas of water resources, or areas with little intensive dynamic
in terms of land use, training areas of compensation against other land uses
(agriculture, forestry, urban ) more intensive.

Spatial strategical disposition: Using the spatial configuration of the elements in the
territory to promote the multifunctionality in it.
234
Area: Environment

Connectivity: Ensuring the continuity of processes across geographical areas with
similar functions (e.g. between areas of high ecological value, productive forest areas,
etc.).

Segregation: Separation of the territory in different areas according to their different
characteristics, which will protect through administrative intervention.

Protection: Elaboration of spatial strategies for the protection of elements of high quality
or fragility.

Divergence of scales: Alternance of elements or areas of fine and thick grain in the
mosaic of landscape.
The use of these principles in the generation of spatial design allows the adaptation them to the
characteristics, needs and opportunities in each hierarchical level of planning.
Dynamic classification of the territory
In order to facilitate the process of design of the spatial concept, it has classified the different
land cover in terms of the dynamic of uses that has generated it. This will allow differentiating
areas according to their dynamic, optimizing the location of the elements of the spatial concept
according to their Correspondence to areas with a high dynamics of land use, or more stable
trend. Five classes have been distinguished to a better adaptation to the characteristics of the
study area, and the evaluated criteria for application in the spatial design. These five categories
are:

Metastable: the dynamic is based in natural biogeochemical cycles, and in disruption of
non-human origin. The intervention of man, although existing, is weak and hardly affects
the development of the ecosystem (seminatural deciduous forests, vegetation of
marshes, etc.).

Transition: associated coverages with neglect or lack of management in land uses. They
are constituted mainly of scrub trees with seminatural species, forest mixed of
seminatural species and intensive production.

Low intensity: Insensitive dynamic but with some length of time periods (12-50 years)
between cycles of production. It is formed, in general, for activities of forest production.

High intensity: Intensive dynamic with short time periods between production cycles (0-3
years). Agricultural areas of forage production, horticulture, fruit, etc.
Artificial: Urban areas and infrastructures.
Once defined this basic legend, the map of land cover of the original ground is reclassified
obtaining a map of the intensity in land uses. This map represents the spatial distribution of
areas based on their dynamic of use. Based on, the location of the components of spatial
concepts can be done precisely according to the spatial detected dynamic.
4. Results
Spatial concepts applied to microscale
Centers of ecological stabilization (1). Spots of major stability, large size and spatial continuity,
a high percentage of interior area (to minimize edge effects), and an edge adapted to the
235
Area: Environment
characteristics of the surrounding land uses. Two types are differed: centers of native
deciduous forests (1a) and centers of thicket (1b).
Areas of stabilization (2). Grouping of spots of seminatural ecosystems whit adjacent areas, not
seminatural, which was intended to be in a metastable state by a decrease in the dynamics of
land use.
Transition zones (3).- They are abundant areas in spots of transitional coverage, although not
directly adjacent to seminatural areas, its position become them in strategic to give continuity to
the spatial concept. This is essentially a connective element, based in spots of coverage the
use of which shows a decrease in the dynamics of land use.
Corridors of connection (4). Elements used for the increase of the connective capacity between
constituent elements of the spatial concept.
Areas of forest intensive production (5). Broad areas in which, by the current existence of forest
crops, or their potential for production, are designated for carrying out an intensive forest
production, with high homogeneity in species (usually Eucalyptus globulus and Pinus radiata)
Areas of agricultural intensive production (6). Broad areas in which, by the current existence
agricultural or fodder crops, livestock activity, or their potential for agricultural production, are
designated for the realization of agricultural production more intensive.
Urban zones (7). Populated areas, usually embedded in areas of intensive agricultural
production.
Figure 3: Elements of the spatial concept to microscale (fragment). The numbering corresponds to that adopted in
the text.
Spatial concepts applied to mesoscale
The elements described above are combined spatially depending on the characteristics of the
territorial structure and each planning unit identified to mesoscale. The combination of elements
leads to three basic structures:
236
Area: Environment
A matrix of forest uses (a) and agricultural (b) of medium intensity, which is alternated
agricultural land for the production of forage and different types of crops, and forest areas with
plantations of eucalyptus and pine.
A network of stabilization (c) (in areas of low dynamic) and/or protection (d) (in areas of
dynamic algae), that surrounding the agroforest matrix and establishes a structure in which the
forest uses are low dinamic. They are composed of cores and areas of stabilization and
transition to microscale, and by connecting corridors. Usually they are areas of deciduous
native vegetation, corridors of vegetation of riverside and abandoned areas that seem to advise
a transition to low land use dynamic.
Cores of stabilization (e). Structures formed by a set of centers and areas of stabilization,
whose main function is segregation and preservation of areas of high metastabilization which
occupy a high extension.
Figure 4: Elements of spatial concept to mesoscale (fragment). The numbering corresponds to that adopted in the
text.
Spatial concepts applied to macroscale
There are three major territorial units, functioning as:
Units of intensive land use.- They are consisted by areas with an productive outstanding
vocation, both agricultural and forest. Consequently, they are areas with great development of
urbanization and infrastructures. Its main function is the agricultural and forest production.
Units of transition .- Units in which are compatible the intensive use of land (agriculture, forest
activities, urbanization) with extensive uses (traditional and ecological agriculture, etc.) and
conservation of natural resources, through the spatial concepts developed in lower levels. The
result is a zone of transition between units of stabilization and other intensive land uses.
237
Area: Environment
Units of stabilization.- The elemental function is the ecological stabilization, on the basis of
structures that promote for the whole unity of the low dynamic in land use, with human
intervention affecting natural processes minimally.
Figure 5: Spatial concept to macroscale. a, units of intensive land use; b, units of transition; c, units of stabilization; d,
elements of the network Natura 2000 are shown for comparison of the area of action.
Integration into a multiscale model
The integration of spatial concepts developed for each different level or differenciated scale
initially is essential for the validity of the model. The structuring of the model about spatial
coordinated strategy with the dynamics of land use at various scales allows the consistency of
the same in different areas of intervention through planning. Also, once defined the spatial
model, it must develop its implementation through policy current instruments, previously
described.
5. Conclusions
The differentiation of units of planning depending of the variation of its heterogeneity is one of
the basis of the exposed methodology, and has allowed the identification of relevant scales in
the planning of the study area (Diaz Varela, 2005). However, this study has helped to articulate
the descriptive aspect about the hierarchical organization of agroforestry landscape, in a
planning methodology at multiple scales. Microscale design focuses on the individual properties
of territorial elements. To mesoscale, is the spatial structure which harbors the properties of
interest, once formed by elements the lower level, which combined effect leads to a
functionality adapted to local conditions of land use in the area. To macroscale, the structures
reflect regional dynamics, both socioeconomic as the geographical and climate conditions
belong to this level. The obtained result is a multilevel structure adapted to the local
238
Area: Environment
peculiarities of the forest production and with ability to coordinate with other spatial strategies of
planning in its different administrative and geographical levels.
The ultimate objective of this structure is to facilitate, through a spatial strategy disposition, the
realization of a consonant planning with ecological integrity of the landscape system, as with
the multifunctional use of natural resources, through the simultaneous combination of an
analytical perspective for each levels and synthetic for the system in set. The combination of
areas of different intensities of land use allows the coexistence of productive activities in the
forest and agricultural fields, with others ecological, cultural and social functions of the mosaic
agroforest.
The implementation of the spatial presented model precise of the development, updating and
implementation of current instruments of forest planning. The UXFOR, to microscale, and the
PORF and PORN to mesoscale, and appropriate guidelines of orientation seem, in a first
analysis, the most appropriate instruments to capture in the territory a multilevel strategy that
allows the multifunctional use of agroforest mosaic.
References
Cook, E.A.; Van Lier, H.N., 1994, Landscape Planning and Ecological Networks. Elsevier
Science B. V. Amsterdam.
Díaz Varela, E., 2005: Ecological Planning multiscale with sytems of geographic information for
sustainable landscapes: application to the region of Eastern Marine (NE of Galicia, Spain). PhD
thesis. University of Santiago of Compostela, Service of Publications and Scientific Exchange.
Santiago of Compostela.
Díaz-Varela, E.; Crecente, R., 2007: Delineation of landscape units at diverse scales using
moving windows for heterogeneity analysis. In R.G.H. Bunce, R.H.G. Jongman, S. Weel, L.
Hojas: 25 Years of Landscape Ecology: Scientific Principles in Practice. Proceedings of the 7th
IALE World Congress. Wageningen, Holland.
Dramstad, W.; Olson, J.D.; Forman, R.T.T., 1996: Landscape Ecology Principles in Landscape
Architecture and Land-Use Planning. Island Press, Washington.
Forman, R.T.T., 1995a. Land mosaics: The ecology of landscapes and regions. Cambridge
University Press, Cambridge.
Forman, R.T.T., 1995b. Some general principles of landscape and regional ecology. Landscape
Ecology, 10 (3): 133-142.
Forman, R.T.T.; Collinge, S.K., 1996. The “Spatial Solution” to Conserving Biodiversity in
Landscapes and Regions. In R.M. DeGraaf; R.I. Miller (Eds.): Conservation of Faunal Diversity
in Forested Landscapes. Chapman & Hall, London, pp. 537-568.
Forman, R.T.T.; Collinge, S.K., 1997. Nature conserved in changing landscapes with and
without spatial planning. Landscape and Urban Planning, 37: 129-135.
Jongman, R.H.G., 1995. Nature conservation planning in Europe: developing ecological
networks. Landscape and Urban Planning, 32: 169-163.
JongmaN, R.H.G.; Külvik, M.; Kristiansen, I., 2004. European ecological networks and
greenways. Landscape and Urban Planning, 68:305-319.
239
Area: Environment
Mander, Ü.; Külvik, M.; Jongman, R.H.G., 2003. Scaling in territorial ecological networks.
Landschap, 20(2): 113-127.
Marey-Pérez, M.F., 2003. Land tenure in Galicia. Model for the characterization of forest
owners. PhD thesis. Digital Publishing (CD). Publications Service of the University of Santiago
de Compostela.
Marey-Pérez, M. F., Fernández-Alonso, S., Crecente, R., Aboal-Viñas, J., 2004. Manual for the
realization of PORF in Galicia, Xunta de Galicia.
Mcgarigal, K.; Cushman, S.A.; Neel, M.C.; Ene, E., 2002: FRAGSTATS: Spatial Pattern
Analysis
Program
for
Categorical
Maps.
Disponible
en
internet,
URL:
www.umass.edu/landeco/research/fragstats/fragstats.html.
MIKLÖS, L., 1996. The concept of the territorial system of ecological stability in Slovakia. In:
R.H.G. Jongman (Ed.): Ecological and landscape consequences of land use change in Europe.
Proceedings of the first ECNC seminar on land use change and its ecological consequences,
Tilburg 16-18 February 1995. ECNC publication series on Man and Nature, vol. 2. European
Centre for Nature Conservation, Tilburg, pp. 385-406.
Petch, J.R.; Kolejka, J., 1993: The tradition of landscape ecology in Czechoslovakia. In R.
Haines-Young; D.R. Green; S.H. Cousins: Landscape Ecology and GIS. CRC Press, Boca
Raton-Florida, pp. 39-56.
Shannon, C. E.; Weaver, W., 1949. The mathematical theory of communication. University of
Illinois Press, Urbana, Illinois.
SUSTEK, Z., 1998. Biocorridors. Theory and practice. In J.W. Dover; R.G.H. Bunce (Eds.): Key
concepts in landscape ecology. IALE (UK), Preston, pp. 281-296.
Van Langevelde, F., 1994. Conceptual integration of landscape planning and landscape
ecology, with a focus on the Netherlands. In E.A. Cook; H.N. van Lier: Landscape Planning and
Ecological Networks. Elsevier, Amsterdam, pp. 27-69.
Van Lier, H.N., 1998. The role of land use planning in sustainable rural systems. Landscape
and Urban Planning, 41:83-91.
Von Gadow, K., Pukkala, T. and Tome, M., 2000. Sustainable Forest Management. Kluwer
Academic Publishers, Dordrecht, 356 p.
VrijlandT, P.; Kerkstra, K., 1994. A strategy for ecological and urban development. In T. Cook;
H.N. van Lier (Eds.): Landscape Planning and Ecological Networks. Elsevier, Amsterdam, pp.
71-88.
Emilio Díaz Varela.
University of Santiago de Compostela. Agricultural Engineering Department..
Polytechnic University. University Campus. 27002 Lugo. Spain.
Phone: +34 982252231
Fax: +34 982241835
240
Area: Environment
ANALYSIS OF PACKAGING WASTE GENERATION AND
MANAGEMENT IN SPAIN
Paneque, A.
Cháfer, C.
Pacheco, B.
Capuz-Rizo, S.F.
Hortal, M.
Instituto Tecnológico del Embalaje, Transporte y Logística (ITENE)
Abstract
The objectives for packaging waste valorization set out by the European Union for 2008 have
mostly been achieved in Spain. However, generation of packaging waste continues to rise
following economic growth very closely. This paper presents an analysis of the current situation
of packaging waste production in Spain compared to other types of waste and its evolution in
the last years indicating some factors which may have affected waste streams. In addition,
some actions and policies that are being taken are also described: prevention in packaging
waste generation and waste material and energy recovery.
Keywords: Packaging waste, waste management
1. Introduction
Packaging waste is any package or packaging material thrown away by users. Packaging
materials soon become packaging waste, which generate significant amounts of waste that
enter the waste disposal, recycling, and recovery streams, contributing to the environmental
impacts brought about by these activities.
The characteristics of packaging waste vary depending on the type of material of which the
packaging is composed. Household waste comes from households and it may have a primary
(sales packaging) and secondary (group packaging) sources from products purchased by the
end consumer in shops. Household waste is recovered by selective collection, or by defect, by
mixed USW (urban solid waste) collection. On the other hand, industrial or commercial
packaging waste is waste from packaging whose use and consumption takes place exclusively
in industries, shops, services or farms (BOE, 1997). Tertiary (transportation packaging) and
secondary packaging that are used to group quantities of packaged goods for distribution and
for display in shops and thus are not acquired by the end consumers make up the third type of
packaging waste (Aucejo et al., 2007). In this case, the sales point should provide proper waste
disposal and management systems, generally through agreements with authorized waste
241
Area: Environment
managers. Industrial packaging waste is more easily recoverable than household waste due to
the great quantity, concentration, homogeneity and quality of the waste.
Regarding waste composition, packaging waste can be composed of different materials, such
as paper, cardboard, glass, wood, plastic or metal either individually or mixed (complex
packages). For example, wood is widely used in the manufacture of pallets; plastic in food
packaging, and glass in beverage packaging.
2. Packaging waste generation in Spain
Packaging waste generation has steadily increased between 1997 and 2005, rising by about
33.66% as shown in Figure 1. The reason for the apparent reduction in the amount of waste
generated in the years 2001 and 2002 is actually due to the lack of available data about wood
waste and the so called “other materials” waste. There is also no accurate data available on
wood waste generation in 1997, but it was included in the “other materials” waste category. In
2005 a total amount of 177 kg packaging waste per capita was generated in Spain.
(http://ec.europa.eu/environment/waste/packaging/data.htm).
Figure 1 was elaborated from the data of the European Commission about the different EU
member states with the aim of monitoring the compliance of the EU’s directive. In turn, each
member state collects the data supplied by the entities and organizations in charge of
packaging waste management. Some of such Spanish entities are Ecoembes, Ecovidrio,
Recipap/Aspapel, Cicloplast, Ecoacero, Arpal or Fedemco (MMA, 2007).
Thousand of tones
8.000
7.000
Other
6.000
Plastic
5.000
Paper/cardboard
Wood
4.000
Glass
3.000
Metal
2.000
1.000
0
1997
1998
1999
2000
2001
2002
2003
2004
2005
Figure 1: Packaging waste generation in Spain (1997 – 2005). Source: Elaborated by the authors based on the data
of the European Commission (http://ec.europa.eu/environment/waste/packaging/data.htm).
Figure 1 shows the distribution of the amount of packaging waste by material. Paper and
cardboard are the most widely used packaging materials in terms of weight, followed by glass
and plastic with similar amounts. In 2005, paper and cardboard accounted for 40.28% by
242
Area: Environment
weight of all packaging, glass 21.47%, plastic 20.12%, wood 12.12%, metals 6.02% and other
materials 0.25%. Observe that, although glass and plastic present similar percentages, the
number of disposed individual plastic containers is higher than that of glass.
Plastic is by far the material that has increased the most over this period. Plastic packaging
waste increased by 51.68% in detriment of glass waste, which only increased by 19.96%, thus
becoming the packaging waste material with the least growth. A moderate increase can be
observed for metals (+37.47%), wood (+36.52% between 1998 and 2005) and paper-cardboard
(+30.02%).
Generation of packaging waste is closely related to current production and consumption trends
in our society, so that it follows the growth of GDP very closely, i.e. the wealthier a country
becomes the more it consumes and the more waste it generates. For this reason, the big
challenge of waste management policies is the “de-coupling” of waste generation from
economic growth (AEMA, 2004).
Figure 2: Evolution of packaging waste (PW) generation vs GDP in Spain (1997-2005), taking 100 in 1997 as base
year. Source: Elaborated by the authors based on the data of the European Commission
(http://ec.europa.eu/environment/waste/packaging/data.htm) and the INE (www.ine.es).
Figure 2 represents the evolution of GDP vs the generation of metal, plastic, cardboard and
glass packaging waste, expressed in percentage taking 1997 as base year. Wood packaging
waste has been excluded from the graph because of the lack of data available for the years
1997, 2001 and 2202, as mentioned earlier. The growth of the Gross Domestic Product in the
period between 1997 and 2005 has been closely followed by an increase in the amount of
packaging waste generated.
243
Area: Environment
Figure 3: Evolution of packaging waste generation vs GDP in Spain (1997-2005), taking 1997 as base year. Source:
Elaborated by the authors based on the data of the European Commission
(http://ec.europa.eu/environment/waste/packaging/data.htm) and the INE (www.ine.es).
Figure 3 presents the rates of packaging waste generated by material (expressed in percentage
for the four materials under analysis) vs the GDP.
Observe that plastic packaging waste is the type of waste that grew the most, far beyond the
GDP, whereas glass packaging waste shows a lower increase. Regarding metal packaging
waste, it grew faster than the GDP until 2002, but since then it has experienced a slight
reduction. This is due to the evolution of packaging technology and the changes in the
consumption habits of Spanish consumers.
Among the factors that affect this growing trend of packaging waste generation we can
mention:
• The reduction in the size of family homes brings about a higher consumption of
individual packages to avoid wasting products (out of date products), which in turn
results in an increase in the amount of individual packaging (Ecoembes, 2003).
Figure 4: Waste generation per capita according to home size (Ecoembes, 2003).
•
•
The consumption of pre-cooked food, of an increasing demand in developed countries,
is closely linked to the generation of packaging waste. The lack of time for cooking or
shopping leads to an increased consumption of packaged and pre-cooked food (AEMA,
2004).
A higher purchasing power allows consumers to buy customized products, that is,
adapted to the consumers’ needs, instead of a single larger product for all the family
244
Area: Environment
•
members. Food, hygiene and beauty industries are the most strongly affected by this
reduction in package size (Ecoembes, 2003).
Market globalization means that the goods manufactured in one part of the world must
arrive in perfect conditions at the end consumer thousands of kilometres away, which
results in more quantity of packaging per product.
3. Packaging waste management
The prevention of packaging waste generation is a key factor of any waste management
strategic policy and system. Reducing the amount of packaging waste and the toxic
components in the waste greatly simplify the tasks of waste management. This reduction
should take place during the process of packaging design. Packaging design must ensure
packaging functionality as well as reduce the environmental impacts associated with the
package life cycle. Some strategic actions for environmentally-friendly packaging design are the
selection of low-impact or “clean” materials, limited use of materials, optimization of production
techniques, optimization of product distribution systems and optimization of the product life
cycle (Capuz et al, 2004).
Royal Decree 782/1998 of 30 April sets down the regulations for the development and
implementation of Law 11/1997 on Packaging and Packaging Waste. One of the key
mechanisms set in the Law to meet the waste prevention and reduction objectives and targets
are the Business Plans for Waste Prevention.
The Business Plans for Waste Prevention use indicators such as reducing the weight of the
materials used in packaging, increasing the number of reusable packages over one-way
systems and recyclable over non-recyclable packaging, limiting unnecessary packaging or
using secondary raw materials obtained from packaging waste recycling.
Packaging companies putting on the market,
during a year, packages that can generate more
than:
250 t, only glass
50 t, only steel
30 t, only aluminium
21 t, only plastic
16 t, only wood
14 t, only cardboard or compound materials
350 t, several materials when any of them
exceed individually the previous quantities
BUSINESS PLANS FOR WASTE PREVENTION
They have
the
obligation to
elaborate
Content:
-Quantified targets of prevention
-Planned measures to achieve the objectives
-Control mechanism to assess its compliance
The BPWP can be elaborated by the Management
Integrated System in which the packaging company
participate.
They must be approved by the Autonomous Region
They have a triennial periodicity
Figure 5: Commitment of developing a Business Plan for Waste Prevention. Source: Elaborated by the authors
based on Hortal et al., 2006.
Despite these prevention measures, as mentioned earlier, the quantity of packaging waste
continues to grow. Any end-of-life treatment of a packaging waste product causes
environmental impacts. According to the binding hierarchy of disposal options set out in Law
10/1998 on Waste Disposal the different possible alternatives of waste management are
prioritized according to their associate environmental impact from low to high as follows:
Prevention, re-use, recycling, energy valorization and disposal (landfill and incineration with no
or low energy recovery).
245
Area: Environment
100,0%
90,0%
80,0%
70,0%
60,0%
50,0%
40,0%
Landfilling
Energy valorization
Recycling
30,0%
20,0%
10,0%
0,0%
1997 1998 1999 2000 2001 2002 2003 2004 2005
Figure 6: End-of-life treatments of packaging waste in Spain. Source: Elaborated by the authors based on the data of
the European Commission (http://ec.europa.eu/environment/waste/packaging/data.htm).
Figure 6 presents the evolution rates of waste recycling, energy valorization and landfilling. Till
2004 the amount of dumped packaging waste was higher than that of valorized waste. In 2005,
a total amount of 56.1% was valorized, 54.4% by the recycling method and 5.7% by energy
valorization.
Recycling is the product end-of-life treatment that has changed the most since 1997, increasing
by 50.45%. Energy valorization has increased by 42.5% and landfilling has decreased by
29.76%.
The comparison of the evolution rates of the end-of-life treatment methods in Spain and the
European Union reveals that until the most recent data available for 2004, the recycling and
energy valorization rates in Spain are lower than the European average rates. The greatest
difference can be observed in the energy valorization rates as the European rate in 2004 was
about twice as much as in Spain (11% in the EU vs 5.6% in Spain).
However, with these overall valorization and recycling rates, Spain approaches the targets set
out for 31 December 2008 by Directive 2004/12/CE (which amends Directive 94/62/CE).
Figure 7: Evolution of the overall rates towards the targets set out for 2008. Source: Elaborated by the authors based
on the data of the European Commission (http://ec.europa.eu/environment/waste/packaging/data.htm).
246
Area: Environment
Packaging end-of-life treatments depend on the type of material of which the packaging is
composed, according to whether it can be easily recycled or used as fuel.
Metals have the second lowest recycling rate after paper-cardboard. Metal recycling brings
about substantial water and energy savings as compared to metal extraction from the ore.
Ferrous metals are easily recoverable by means of electromagnets; by contrast, non-ferrous
metals require more sophisticated recovery techniques. In 2005 a total amount of 59.7% metal
packaging waste was recycled, 1.4% was valorized and 38.9% was dumped. The European
target of 50% metal recycling has been exceeded since 2004.
Glass is a material easy to identify and sort out for further recycling and, among other
advantages, the re-use of glass containers as raw material for the manufacture of new glass
containers saves substantial amounts of energy. During 2005 a total amount of 44.4% glass
waste was recycled and the rest (55.6%) was dumped. Due to its properties, glass cannot be
energetically valorized. The 60% target for glass recycling has not been met yet as there is still
15.6 percentage points missed to meet the objective set in Directive 2004/12/CE.
247
Area: Environment
Landfilling
Energy valorization
Recycling
*Unavailable data.
Figure 8: Packaging end-of-life treatments by material. Source: Elaborated by the authors based on the data of the
European Commission (http://ec.europa.eu/environment/waste/packaging/data.htm).
Paper and cardboard possess high calorific value so that they can be incinerated and the
energy recovered. After plastic, paper and cardboard wastes present the highest energy
valorization. However, their high value as raw material for the paper industry prevents the use
of natural cellulose from woods; this along with its facility to be recycled causes this type of
packaging waste to present the highest recycling rates. In 2005 a total rate of 69.2% paper
waste was recycled and 24.3% was dumped. In 2004 the objective of 60% paper and
cardboard recycling for 2008 was exceeded.
248
Area: Environment
Plastic is the most energetically valorized material although it is also the most frequently
dumped to landfills. Plastic packaging waste possesses a high calorific value so that it is used
as fuel in solid urban waste incineration plants. Additionally, there is a great variety of plastics
that have to be previously sorted out for further recycling, thus increasing the cost of recycling.
According to the 2005 data, a total amount of 20.7% plastic packaging waste was recycled,
13.4% was energetically valorized and the remaining 65.9% was dumped. With these rates,
plastic lies 1.8 percentage points below the EC target of 22.5%.
Finally, wood is a material with a high potential to be energetically valorized due to its
combustible capacity, but wood packaging waste is also used in the manufacture of
conglomerate boards. In addition, some wood packages, e.g. pallets, are frequently re-used
either by the company itself or supplied by recyclers (second-hand pallets). In 2005 a total
amount of 43.9% wood packaging waste was recycled, 2.6% was energetically valorized and
53.4% was dumped. Therefore, the European objective of 15% has been significantly
exceeded.
Figure 9: Packaging end-of-life treatments by material (2005). Source: Elaborated by the authors based on the data
of the European Commission (http://ec.europa.eu/environment/waste/packaging/data.htm).
Figure 9 shows the comparison of the end-of-life treatments of the different packaging wastes
by material using the 2005 data. As can be observed:
•
Plastic is the most frequently dumped, less commonly recycled and most energetically
valorized material
•
Paper and cardboard are the materials with the highest recycling rate, the second in
energy valorization and the least frequently dumped to landfills
•
Glass and wood present similar recycling rates; but due to the fact that glass, unlike
wood, is energetically little valorized, the volume of dumped glass waste is slightly
greater
•
The rates of dumped metals are only slightly higher than those of paper or cardboard
thanks to the great efforts made in metal recovery and recycling because of their high
economic value
249
Area: Environment
4. Conclusions
European current trends tend to reduce the amount of packaging waste that ends its life in
open dumps or landfills (AEMA 2007). For this end, the solution is to recycle or energetically
valorize larger amounts of waste. As mentioned earlier in the paper, energy valorization is more
or less feasible depending on the type of waste material.
In the case of plastic packaging waste, the trend is to increase the amount of energetically
valorized plastic either by the incineration method with energy recovery or by alternative
processes like gasification (Cicloplast 2005).
As regards paper and cardboard, paper industries bet for raising the recycling tax through the
improvement of the paper waste collection systems and using the incineration method only
when the paper conditions prevent it from being recycled.
With respect to glass and metals, the trend is to increase their recycling rates and reduce the
amount of dumped waste.
Regarding wood packaging waste, in addition to increasing its recycling rate, the trend is to
significantly increase its energy valorization rate and take advantage of the combustible
properties of this material.
As for the reduction in waste generation by means of prevention measures, it is a task that
concerns all stakeholders involved in the process (Public Administration, materials
manufacturers, packaging industries, suppliers and consumers); this means to apply a general
approach that takes into consideration the overall life cycle of the packaging product thus
avoiding the transfer of environmental impacts from one stage of the process to the next
(Ecoembes 2003).
Finally, a comment on the problem of data availability. Although the data on packaging waste
generation and management are easily available as it is mandatory for the EU member states
to submit fact sheets and reports to the European Commissions for waste monitoring, the data
about the different materials are collected by different entities and each of such entities uses a
different data collection method. Thus, data availability depends on the type of material and the
waste collection method.
References
Aucejo S., Capuz S. et al., 2007 Estudio comparativo de las características técnicas y de
tratamiento y disposición final del cartón frente al film retráctil como materiales de envase y
embalaje, Ed. ITENE, Valencia.
BOE (Boletín Oficial del Estado), Ley 11/1997, de 24 de abril, de Envases y Residuos de
Envases, BOE nº 99, april the 25 th. Madrid. 1997.
Capuz S., Gómez T., Vivancos J.L., Viñoles R., Ferrer P., López R. y Bastante M.J., 2004,
Ecodiseño. Ingeniería del ciclo de vida para el desarrollo de productos sostenibles, Ed.
Universidad Politécnica de Valencia, Valencia.
Cicloplast, 2005. Estadísticas de consumo, generación, reciclado, valorización y mercados de
los plásticos, España.
250
Area: Environment
EC
(Comisión
Europea).
Web
consulted
http://ec.europa.eu/environment/waste/packaging/data.htm
in
february
2008,
Ecoembes, 2003. Catálogo para la prevención de residuos de envases, Madrid.
EEA (European Environment Agency), 2004. European packaging waste trends and the role of
economic instruments, Brussels.
EEA (European Environment Agency), 2007. The road from landfilling to recycling: common
destination, different routes.
Hortal, M.; Vivancos, J.L.; Aucejo, S. 2008. How to combine environmental tools to minimize
packaging environmental impacts. Proceedings 15th IAPRI World Conference on Packaging.
Tokyo, Japan. 2006.
INE. (Spanish Statistics National Office). Web consulted in march 2008, http://www.ine.es
MMA (Spanish Ministry for the Environment), 2007. Versión Preliminar del II Plan Nacional
Integrado de Residuos 2008-2015, Madrid.
Acknowledgements
The findings reported in this paper are part of a more extensive work carried out within the
general project “Identification and Analysis of the Critical Points in Waste Management
Systems for Plastic and Cardboard Packaging from food industry, through Environmental
Indicators - ECOWASTECH”, financed by the Spanish Ministry for the Environment within the
2007 R&D&I Project Call
Ángela Paneque de la Torre
Universidad Politécnica de Valencia. Departamento de Proyectos de Ingeniería.
Camino de Vera s/n. 42006. Valencia, Spain.
Phone: +34 96 387 7000 (ext. 75688)
URL : http://www.dpi.upv.es/id&ea/
251
Area: Energy Efficiency and Renewable Energies
TOPIC AREA: ENERGY EFFICIENCY
AND RENEWABLE ENERGIES
253
URBAN TRANSPORT PROJECT BY HYDROGEN TECHNOLOGIES
FOR INTERNATIONAL EXHIBITION EXPOZARAGOZA 2008
Montaner, P.
IDOM Engineering
Abstract
This summer, 2008, Zaragoza hosted an international Exhibition with the theme of Water and
Sustainable Development. With this baseline EXPOAGUA, the managing company, decided to
carry out an urban transport project with hydrogen technologies. The project began with the
opening of the Exhibition in June 2008, and it continues alive in the city today. The project,
pioneered in Spain, developed a completely sustainable model. The vehicles selected for the
project consume only hydrogen as energy for their engines and they generate only water steam
as waste. The provision of hydrogen for these vehicles is done at a new hydrogen filling station
with hydrogen generated on-site by electrolysis.
Keywords: Hydrogen, Electrolyser, Fuel cell, Project management
1. Introduction
This summer an International Exhibition was held in Zaragoza (Spain) with the theme of “Water
and Sustainable Development”. In line with this marvellous philosophy, EXPOAGUA, the
management company of the exhibition, invested firmly on developing a green urban
transportation project with vehicles powered solely by hydrogen. The aim of the project was not
only to be ready for the opening of the exhibition but also to continue this as a lifelong project in
the city, as a reminder of the sustainability philosophy.
This pioneering project is aimed at the future. Taking advantage of the success of an
International Exposition, it aims to familiarize people with hydrogen technologies and to make
them aware of the need to avoid using contaminant fuels, in an effort to reduce the harmful
effects that contaminating fuels have on the environment. In the same way, the idea of
transforming Zaragoza into a strategic point of the future Hydrogen Highway, with a hydrogen
supply network for national transport as the European Community already foresees, is also
important.
The project was composed of two main areas. On the one hand to build a hydrogen supply
infrastructure, and on the other to get a fleet of hydrogen vehicles, as various and versatile as
possible, with no emissions.
According to the idea of a completely sustainable project, all the electricity consumed by the
exhibition facilities was obtained from renewable resources. For this purpose EXPOAGUA
installed two wind farms and two solar gardens, where it generated the energy required for all
of EXPO’s needs, including the hydrogen station. Through this route, the circle of sustainability
was closed, showing the technical viability and the possible establishment of this kind of
technology.
255
Figure 1: Energetic circle of the project.
2. Hydrogen Filling Station
“Hidrogenera EXPO” was the name we familiarly called the hydrogen filling station. This station,
located in a new neighbourhood in the south of the city, was the first non-private installation of
hydrogen, and the only one in operation in Spain during the summer of 2008. Furthermore, it
had the peculiarity of being able to produce its own hydrogen on-site, meaning that is was not
necessary to take the fuel by road, as is the case with traditional filling stations for conventional
vehicles.
The operation of the hydrogen station can be split into several clearly differentiated phases. In
the first phase, the production of hydrogen is carried out by a water electrolysis process. The
system takes water from the city network and demineralises it with an inverse osmosis filter.
The electrolyser takes this water and with the application of an electric current produces the
disassociation of the water molecule into its two components, hydrogen and oxygen, according
to the following chemical reactions:
Catod (−) : 2 H 2 O + 2e − → 2OH − + H 2
Anode(+ ) : 2OH − → H 2 O + 1 O2 + 2e −
2
Figure 2: Diagram of alkaline electrolyser.
256
The alkaline electrolyser uses as an electrolyte, a dissolution of potassium (KOH), and is
capable of producing a maximum of 12 normal cubic meters of hydrogen with a purity of over
99.5% with an outset pressure of 8 bar. The system is designed for continuous operation, so
the maximum capacity of daily generation is close to 25 kg of hydrogen. The use of hydrogen
in fuel cells requires a very high purity of gas, more than the hydrogen that is obtained directly
from the electrolyser. This means that it is necessary to use a purifier unit, so that we can get a
high purity of hydrogen (99,9995%) with a maximum waste of 5 ppm of oxygen.
In case the hydrogen production is not enough to satisfy the demand or in case of maintenance
or failure of the electrolyser, the station has the ability to get an external supply of hydrogen
from 200 bar hydrogen platforms.
The hydrogen produced in the electrolysis process is compressed by a two-phase membrane
compressor, increasing the outset pressure from the electrolyser to the storage pressure
designed at 420 bar.The compressor connects automatically when the pressure of the storage
system decreases below the key point and it stops when the storage is full. The gas is stored
in high pressure bottles with a total maximum capacity of storage at over 56kg of hydrogen.
The supply to vehicles is done through a special hydrogen dispenser which is capable of
refilling at two different storage pressures: 200 bar and 350 bar. This special condition exists
because of the need to be able to refill the different kinds of vehicles that have been used
during the International Exhibition.
The whole system has been measured to supply hydrogen to a standard fleet of three to four
buses, so that they can be refilled daily according to the operating schedules of the charge and
discharge of storage bottles.
257
Electrolyser
BackBack-up
Storage
Compressor
Control Unit
Figure 3: Diagram of components.
Each one of the phases mentioned is carried out within a new facility which has the most up to
date control systems and the most secure facilities ensuring correct operation.
3. Vehicle fleet
With respect to vehicles, the project had two main goals. On the one hand, to take advantage of
the Exhibition to combine a fleet of various kinds of hydrogen vehicles, as extensive and
various as possible, showing their functionality and working at the same level as internal
combustion engine vehicles. On the other hand, it aimed to keep a small fleet of these vehicles
in operation in the city once the exhibition finished, using the new hydrogen station for refilling.
The current reality of hydrogen vehicles is a little virtual. Though many possibilities can be
found at an international level, few of them are available or developed enough. Most of the
small number of vehicles on offer are simply prototypes or are in test phase, so there is no real
possibility of purchasing one. The small number of options with any real possibility of being
purchased, are very costly. So it is very difficult create a fleet with enough units for a new
transport system. Working within this scenario our project chose to acquire some of the
vehicles and to borrow or rent others from covenant firms with a view of returning them once
the exposition finished.
The project’s own fleet included three minibuses with a capacity of 22 people and range of over
200 km. They are small sized vehicles, less than 6 meters long, so they are very adapted to a
258
city centre or areas with restricted mobility. During the international Exposition they were doing
special shuttle services inside EXPO for diplomatic staffs, but they will be used in city standard
transport service now that the exposition has finished.
Apart from these minibuses, there were also a number of other vehicles. The primary one used
was a large capacity bus, capable of transporting more than 100 people and with range of over
400 km. There were several single-seater vehicles too, such as scooters and bicycles, also
powered by hydrogen. Between them all, a distance of more than 19,000 km was covered in
the three months of the exhibition. And they transported over 60,000 people.
Photograph 1: EXPO hydrogen vehicle fleet.
Although fuel cells are not the only technology available for use in hydrogen vehicles, all the
vehicles in the EXPO project, were fed from fuel cells. The process for obtaining electricity
from a fuel cell is the opposite to that used in the electrolysis process. In this process,
hydrogen stored in tanks and the oxygen taken from the atmosphere are mixed to get a water
molecule which generates an electric current that is used to feed the electric engines.
This process, for a PEM fuel cell, follows the following chemical reactions:
Anode(+) : H 2 → 2 H + + 2e −
Catode(+) : 2 H + + 1 O2 → 2 H 2 O
2
With this technology, the only waste produced by the vehicles is the water steam of the
chemical conversion. There is no other possible type of contamination. If we can obtain the
hydrogen from water electrolysis with electricity from renewable sources, as we currently do,
we are able to assure that the total process is completely free of any contaminant emissions.
4. Project challenges and conclusions
The management of a project of these characteristics, with such a so innovative technical
development, and with a very tight schedule, supposed a big challenge just only for the
management of all the works and stakeholders.
Some of the key challenges for the project were issues related to authorisation and licensing,
as well as standards and regulations for these types of installations. In fact there are no
specific directives in Spain for hydrogen stations, so it was necessary to work together with the
Municipal and Regional Governments to agree the applicable regulations and security
procedures to improve at the current installation.
259
The solution to this problem was to apply all the regulations referring to any facility related to
the present installation, in all or in part, with specific consideration of the most restrictive
conditions for each of them. Because of this, the hydrogen station complied with all the
requirements in the main industrial regulations including the Fire Protection Rules, Chemical
Products Storage Rules, Electrical Standards, and so on. The application of all these
regulations made the station one of the most secure industrial installations.
Another area that really created a big challenge was the project schedule. Though the project
began some months before, the first important milestone of the project was the contracting of
the supply and construction of the hydrogen station, which was signed on August 2007. So, it
was necessary to develop the main part of the project described above in less than one year.
The authorisation projects were finished on November 2007 and the building and activity
licences were organised at the beginning of 2008. The work of building and urbanizing the
station was finalized in April in order to start the assembly of the hydrogen facilities. The testing
phases were done by the first half of May, only one month before the opening of the Exhibition.
As a conclusion of the project it is necessary to say that although it has been shown that
hydrogen technologies are theoretically viable, the actual reality of hydrogen is not as advanced
as that of currently operating technologies.
It is obvious that the future establishment of hydrogen as a replacement for petroleum fuels
should begin with the hard support of government and with all the initiatives that prompt these
independent projects.
References
Bossel, U. 2006. Does a Hydrogen Economy Make Sense?. Proceedings of the IEEE, Volume
94, Issue 10, Page(s):1826 - 1837 Gobierno de Aragón. 2007. Plan Director del Hidrogeno en
Aragón,
Fuel Cells Handbook, 7th edition, Department of Energy, DOE, US. 2004
Ochoa, J.R., 2008 Normativa y Reglamentación de seguridad en materia de Hidrógeno
Diploma de Especialización en Tecnologías de Hidrógeno, Univ. de Zaragoza.
Ponencias, 2007. 3er Encuentro Sectorial del Hidrógeno y las Pilas de Combustible. Santiago
de Compostela. www.pmH2.com
Pool, R.; 2005 Pump up the gas [hydrogen storage for transport applications]. Power Engineer.
Volume 19, Issue 1, Page(s):18 - 21
T-Raissi, A.; Block, D.L.; 2004. Hydrogen: automotive fuel of the future. Power and Energy
Magazine, IEEE Volume 2, Issue 6, Page(s):40 - 45
Zabalza, I. Valero, A. Scarpellini, S., 2005 Hidrógeno y Pilas de Combustible, Fundación para
el Desarrollo de las Nuevas Tecnologías del Hidrógeno en Aragón. 2005.
Pedro Montaner Izcue.
IDOM
Phone: +34 976 561 536
260
AN EXAMPLE OF BIOCLIMATIC CONSTRUCTION.THE BUILDING
PETER3
López-Rodríguez, F.
Segador, C.
Cuadros, F.
Mena, A.
Universidad de Huelva
Abstract
The principal objective of this communication is to show an example of bioclimatic building. It is
about the construction of an Intelligent "Zero Energy Building" (ZEB) of 1.000 m2 in the Campus
of the University of Extremadura in Badajoz. The concepts of energy efficiency, power storage,
environmental impacts, etc., in construction will be put in practice. A comparative analysis
between the different conventional technologies and the renewable ones will be carried out.
Between the used technologies are those of passive and active solar heating, photovoltaic
power, ventilated façade, Trombe walls, use of an hybrid solar-biomass system to feed an
absorption engine for refrigeration, etc. Finally, a complete control and analysis of the more
important confort variables will be made in real time.
Keywords: Zero Energy Building (ZEB), Energy Efficiency in Edification, Bioclimatic Building
1. Introduction
Energy consumption in the housing and service sectors has great relevance in the ambit of the
European Union, provided that it exceeds 40% of the total. All European countries have taken
steps to restrict the energy demand (like for instance the new Royal Decree for the Technical
Building Code –TBC- in Spain). Therefore, the application of sustainability criteria to building
activity in order to favour energy collection is a priority task for public administrations.
Spain is the most energy consuming country in the European Union if referring to the tertiary
sector. It has also the greatest potential for harnessing renewable energy sources.
3
PETER. Parque Experimental Transfronterizo de energías renovables, (Experimental Transborder Park
on Renewable Energies)
261
The use of bioclimatic architecture is far from being a modern technique: caves and most
traditional constructions are based on bioclimatic principles. However, bioclimatic buildings are
an unusual type of housing at present. Although there is no specific official registry, it is
estimated that there are between 5000 and 10000 buildings of this type in Spain, according to
the Centre for Energetic, Environmental and Technological Research (CIEMAT).
Unfortunately, many housing projects are promoted nowadays as bioclimatic, although they
should not be taken as such. To what extent should a building be considered to be bioclimatic?
Such a denomination is wrongly being assigned to buildings that have a single eave as passive
shading element and that integrate certain renewable energy devices.
It is important to note that the concept of bioclimatic construction might be seen as new from a
textual point of view, but it is not as the constructive concept itself. Yet the Romans designed
ventilated covers, and the Arabs invented the tile, the courtyards with ponds and convective
loops, and inside gardened zones. In this sense, let us cite some examples like the ventilated
adobe walls in centenary constructions of Marrakech, or the efficient cow-dung huts of the
Nubis people in the High Nile.
There are some projects being undertaken both inside and outside Spain which are similar to
that proposed in the present communication. Some of the most relevant are the following:
Sustenergy (Sustenergy, 2007). It was supported by the INTERREG III C Program, and
ended in 2007. Six different institutions took part in this project, and was led by the Centre
of Environmental Resources of Navarra (CRAN). The main goal was the development of
common technologies in order to achieve energy saving and efficiency strategies in
construction. One of the actions promoted was the set up of thirty pilot experiences, among
which the Program for Efficiency in Construction was the most relevant.
PSE-ARFRISOL(http:/www.energiasrenovables.ciemat.es
/suplementos/arfrisol/
psearfrisol.htm). Bioclimatic architecture and solar cooling. It is a Singular and Strategic project
included in the National Plan of MEC and led by the CIEMAT. It is developed by twelve
institutions, like solar energy technology companies and research groups from different
universities and from the CIEMAT itself.
Project CONAMA VII. It was presented in the Sustainable Development Conference held in
November 25th 2004, in connection with the topic of Ecoefficiency in Construction. This
project allowed to carry out detailed studies on the evaluation of the environmental
behaviour of several buildings in Spain.
Contest of ideas for the construction of a rural house in Chile, according to the principles of
sustainability and energy saving Cities for a More Sustainable Future, which can be looked up at the website of the
Department of Urban and Landscape Planning in the Superior Technical School of
Architecture (Polytechnic University, Madrid).
Agenda for Sustainable Construction (Sustainability and construction, 2000) , a website
designed by the Official College of Master Builders and Technical Architects of Barcelona,
the Superior Technical School of Architecture of Vallés, the Association for Geobiological
Studies and the Cerdá Institute.
New York’s Department of Design and Construction. Office for Sustainable Design. This
website is specialised in resources for sustainable construction, and is organised in six
categories: projects, local legislation, reports and manuals, specification documents, pilot
projects and reference examples.
Website
of
the
Official
College
of
Architects
of
Cataluña
(http://www.coac.net/mediambient/), devoted to the Environment in the ambit of
sustainable construction.
262
CIEMAT, a technology centre linked to the Ministry of Education and Science devoted to
promote Research-and-Development projects in the fields of energy (energy use of
conventional and alternative sources), of environment (impact of energy on population and
environment), of radiation protection (control and surveillance of ionizing radiation
generated at CIEMAT), of magnetic confinement fusion (study of the device Stellerator
Heliac flexible TJ-II) and also on basic research.
Project LIFE-EcoValle. Bioclimatic conditioning of the road C-91 of UE-1 in the outskirts of
Vallecas, in collaboration with CIEMAT.
Project Regen-Link, for the restoration of buildings and commercial lots in San Cristóbal,
Madrid. In collaboration with CIEMAT.
Ecocity in Sarriguren, promoted by the Department of Environment, Landscape Planning
and Housing of the Government of Navarra. The actions for planning and management are
being held by the public company Navarra de Suelo Residencial (NASURSA).
Forum Barcelona, a recent project based on the recycling of urban land.
The present communication describes the development of a project for the construction of an
intelligent zero-conventional-energy building (“bioclimatic”) in the Campus of the University of
Extremadura. It is linked to the development of the Project PETER (Experimental Transborder
Park on Renewable Energies), which involves the Spanish institutions University of
Extremadura, Province Deputation of Badajoz, Government of Extremadura (General Direction
of University and Technology, Regional Ministry of Economy, Commerce and Innovation), IDEA
and CIEMAT, and the Portuguese institutions University of Évora, INETI and ADRAL. The
project is being carried out under the auspices of the Program INTERREG III of the European
Union.
This proposal corresponds to what is referred to as sustainable edification, which includes
actions in three different aspects: economic, environmental and social, and can be regarded as
the continuation of other previous works performed by the research group DTERMA, from the
University of Extremadura (Cuadros et al., 2007; Lopez-Rodríguez et al., 2006; García SanzCalcedo et al., 2008)
The main goal of the project described here are:
- To put in practice the concepts of energy saving and efficiency for the energy transformation
devices relating different sources, energy storage, environmental impact, etc.
- To carry out a comparative empirical analysis among different energy sources, in order to
emphasise the advantages of renewable versus conventional energy sources.
2. Description and technologies
The main features and technologies used in the construction of the bioclimatic building are
listed below. The building is projected to have two floors with approximately 400 m2 each, and a
300 m2 cellar. Laboratories and office rooms will be oriented to the south, whereas rooms of
general use (like bathrooms, meeting rooms, conference room...) will face North. The building
shape is rectangular and its main facade will be oriented to the South.
Figures 1 and 2 show some views of the building as it will definitively look
263
Figure1: Views of PETER building.
Figure 2: Views of the North and South facades of PETER building.
The concrete technologies involved are the following:
•
Construction according to the local climate parameters of Badajoz.
•
Monitoring of energy fluxes and of wind, as well as installation of human presence
detectors, in order to achieve an automatic operation to control such fluxes.
•
Application of the specific concepts concerning passive solar heating. Use of Trombe
walls. Ventilation of South facade by installing photovoltaic panels. Avoid direct solar
radiation into the building during the summer (and vice versa during the winter).
•
Efficient thermal insulation, avoiding thermal bridging.
•
Window and cover shadowing.
•
Natural lighting, combined with high efficiency artificial support lighting.
•
Mixed solar-biomass acclimatisation. Installation of approximately 70 m2 high efficiency
solar surface collectors using a biomass boiler (pellets), which will serve as energy
supply to a cooling absorption engine.
•
Full monitoring of the building.
•
Real time data transfer to the internet.
3. Techniques
The most innovative techniques will be applied in order to ensure the bioclimatic behaviour of
the building, the installation of efficient insulation and the use of renewable energy sources. The
specific actions concerning each of the technical aspects are described in next subsections:
3.1. Use of solar radiation
The position of the sun has been simulated along the whole year in order to evaluate its
influence and project actions to avoid heating during the summer and to favour solar exposure
during the winter. This will be of special interest for rooms of general used, which will thus be
facing South. Figure 3 shows the trajectory of the sun for all months, referred to the concrete
location of the building. The aim is to avoid shadowing by surrounding buildings.
264
Figure 3: Position of the sun referred to the concrete location of PETER building.
3.2. Insulation
The process of thermal energy transmission in a building of these features takes place mainly
by radiation process (75%), but also involves residual conduction and convection (25%).
Therefore, any type of building insulation project should be designed mainly to impede the
radiative thermal energy flux, but it should also account for residual conductive processes. This
way, reflective insulation appears as the most adequate and efficient technique, provided very
high output due to high reflective power and to the specific structure for the retention of air
bubbles.
The results achieved by reflective insulation (versus traditional methods) are quite significant.
Figure 4 shows the thermal resistance for several types of insulation (values of thickness
between 1-10 cm, with 5 mm step interval).
Reflective insulation presents a flat curve from a concrete value of the air chamber thickness,
and efficiency is not improved for higher values. However, thermal resistance is significantly
higher than that of a conventional insulating material. In sum, this feature favours the
optimisation of useful interior space.
265
Figure 4: Comparative plot of thermal resistance of a wall as a function of thickness for several types of insulating
materials.
Let us describe as an example the insulation pattern projected for both South and West
facades of PETER building, following the actual order of the insulating coatings:
•
Marble shield (model Frontek de Venatto), 2 cm thick.
•
Low-emissive air chamber (4 cm).
•
4 mm reflective insulator, with intrinsic resistance 0.11 m2 K W-1.
•
Structural panel (12 mm).
•
Natural fibre insulator (25 mm).
•
Low-emissive air chamber (2 cm) (ε = 0.03).
•
4 mm Superpolynum reflective insulator, with intrinsic resistance 0.11 m2 K W-1.
•
Low-emissive air chamber (4 cm) (ε = 0.03).
•
Laminated plaster sheet (15 mm).
The total transmission coefficient (the U-value) of this type of wall is U-value=0.31 W m-2 K-1.
The next step is the comparison of the thermal conductivity parameters with those of the TBC,
DB-HE 1, which indicates the maximum limitation of the energy demand of a building,
considering maximum required transmittance according to the type of walled enclosure and to
the climatic zone of the location of the building. Taking this aspect into account, a detailed
analysis of the energy demand shows energy savings over 69% in winter and over 86% in
summer with respect to the values stated by the TBC (see Table 1).
266
Winter values
Summer values
Walls
TBC
Project
Reduction
TBC
Project
Reduction
N
4 294.0
1 259-50
71%
4 294.00
557.30
87%
E
1 073.50
354.08
67%
1 073.50
156.67
85%
W
1 073.50
354.08
67%
1 073.50
156.67
85%
S
2 147.00
708.17
67%
2 147.00
313.35
85%
Winter values
Summer values
SE
0.00
0.00
0.00
0.00
SW
0.00
0.00
0.00
0.00
Σ=
8 588.00
2 675.84
8 588.00
1 184-00
69%
86%
Table 1. Values for the energy efficient of the envelope of PETER building, compared with those stated by the TBC.
Such a reduction of the energy demand will lead to lower power in the heating and the
acclimatisation systems. Besides, if renewable energy devices were installed, then fossil fuels
or conventional energy sources would not be required. The use of those devices would not only
result in environmental benefits, but also in notable economic savings.
3.3. Heating system
The heating system of the building is designed as the combination of the following installations:
•
Radiant floor warming.
•
Solar thermal panels.
•
Biomass boiler support.
•
Hot air support for rooms of general use from photovoltaic panels installed in walls.
The installation of 25 solar thermal panels (50 m2 effective surface, approximately) will provide
56% of the energy demand of the building. A biomass boiler will work as energy support
equipment. Besides, the air flow along the back side of the photovoltaic panels installed as a
chimney on the South facade will serve as supplementary heating system in winter. On the
other hand, the thermal energy obtained from the photovoltaic panels will be transferred to the
surroundings, according to the principles of Trombe walls.
Thermal energy supply to the building will be made by radiant floor warming, which is currently
considered to best approach the ideal heating profile, since it provides very high efficiency and
maximum thermal comfort. Figure 5 shows values for energy supply of photovoltaic panels.
267
NEEDS AND SAVINGS
SAVINGS
NEEDS
Figure 5: Energy supplied by the solar thermal panels heating system.
3.4. Cooling system
Some of the installations projected will also be used for the cooling system of the building.
•
Upper air grilles.
•
During night hours, use of grilles to collect cool air flow generated at an artificial lake
located in front of the South facade.
•
Absorption engine, supplied by solar thermal panels and a biomass boiler.
•
Automatic natural ventilation support at nights.
Figure 6 describes the cooling circuit of the building. The analysis of data from the 30 solar
thermal panels shows that the absorption cooling engine supplies 88% of the acclimatisation
needs of the building. The remaining 12% will be supplied by the supplementary biomass
boiler.
The absorption cooling system, together with the solar thermal panels, provides a high
efficiency since the greater the energy demand for cooling is, the higher will also solar radiation
be. This will make the photovoltaic panels work at maximum capacity.
268
Figure 6: Working scheme of the absorption cooling engine.
3.5. Electric system
A set of photovoltaic solar modules are projected to be installed on the South facade of the
building in order to work as electric energy supply. Part of this energy will be used to design a
high output installation for lighting. The remaining energy will be transferred to the electricity
grid, which will allow the perception of the bonus for photovoltaic solar energy production,
according to the Spanish legislation.
The photovoltaic panels are designed to be installed at an angle of 85º with the horizontal.
Despite such value is not optimum (i.e. it does not maximise the production of electricity), this
configuration favours the integration of the panels to the design of the building, which in the end
should also be a priority goal.
3.6. Rest of installations
An installation for the management of water is also projected. For this purpose, the following
actions are considered:
•
Collection of rainwater and storage in an artificial lake.
•
Use of water for cooling of air and radiating floor.
•
Recycling of wastewater for WC and for garden irrigation.
Moreover, all the systems are designed to be fully integrated and monitored by a control unit,
so that a global quantification of systems and resources can be carried out at any moment. The
main actions regarding this are:
269
•
Lighting control.
•
Temperature control for the solar collection system, the photovoltaic installation, the
biomass boiler and the adsorption engine.
•
Control of the photovoltaic installation connection to the electricity grid.
•
Control of blinds at night.
•
Control of blind slats according to solar radiation.
•
Control of the natural cooling system at night.
•
Control of the heating of common use rooms with the air from the chambers between
the photovoltaic panels and the facade of the building.
4. Conclusion
The aim of the construction of a building with these particular features is to put in practice the
most innovative technologies in order to minimise the energy demand. The basic line of work is
the detailed analysis of all equipment, as well as the focusing on the thermal insulation of the
building so that energy supply –always from renewable sources- is reduced as much as
possible.
Following the principles of sustainable construction, the project described in the present
communication is intended to show that renewable energy technologies can be satisfactorily
integrated in construction, with similar efficiency to conventional ones. It is also relevant to note
that important economic savings and high profitability will be achieved during the life of the
building.
The PETER building is projected to work as a laboratory that will allow the comparison of
results obtained in real situations with simulation data generated in the present study. In sum,
our efforts are focused towards the industrialisation of construction, and we believe this project
to provide an important “know how” reference in such direction.
References
Cities for a more sustainable future. Website of the Department of Urban and Landscape
Planning in the Superior Technical School of Architecture (Polytechnic University, Madrid) and
Housing Ministry. http:// habitat....
Contest of ideas: energy efficiency development for rural social housing. Chile.
http://www.plataformaarquitectura.cl/2006/09/23/concurso-de-ideas-desarrollo
-de-eficienciaenergetica-para-vivienda-social-rural/
Cuadros F., Lopez-Rodríguez F., Segador C., Marcos A., 2007. A simple procedure to size
active solar heating schemes for low-energy building design, Energy and Building. Vol 39, pp.
96-104.
García Sanz-Calcedo J., López-Rodríguez F., 2008. Cuadros Blázquez F., Moura Joyce A.,
Energy Management in Health Centres, Energy Conversion and Management. Sent for
publication.
Lopez-Rodríguez F., Segador C., Marcos A., and Cuadro F., 2006. Cálculo y comparación de
rendimientos para distintas aplicaciones, Era Solar, Vol 131, 2006, pp. 73-77.
270
PSE-ARFRISOL. Bioclimatic architecture and solar cooling. CIEMAT.
http:/www.energiasrenovables.ciemat.es/suplementos/arfrisol/pse-arfrisol.htm
Sustainability and construction: applications and lines of work. Gerona: Department of
Architecture and Civil Engineering of the University of Girona. D.L.2000.
Sustenergy. Project financed by Program INTERREG III C, 2007.. http://www.sustenergy.com
Website of the Official College of Architects of Cataluña.
http://www.coac.net/mediambient/
Acknowledgements
The authors wish to express their gratitude to Program INTERREG III of the European Union –
as to the rest of cofinancing institutions- for financial aid for PETER Project. Cooperation
among all members of the Project is also acknowledged.
271
Area: Information and Communication Technologies and Software Engineering
TOPIC AREA: INFORMATION AND
COMMUNICATION TECHNOLOGIES
AND SOFTWARE ENGINEERING
273
SOFTWARE FOR DETERMINING DRIPLINE IRRIGATION PIPES
USING A POCKET PC
Molina Martínez, J.M.
Ruiz Canales, A.
Universidad Miguel Hernández de Elche
Abstract
This communication presents the development of a freeware for mobile devices (Smartphone
and Pocket PC). It was developed with a LabVIEW program, and enables engineers and
installers to calculate commercial diameters to be used in laterals of drip irrigation, without the
need for being at the personal computer. Input data required: drippers flow rates, number of
drippers, spacing between the drippers, medium pressure in the lateral and pressure tolerance.
As results, it shows commercial diameters that can be used with diode LED, maximum and
minimum pressures for each lateral, and a graphic that shows, on a comparative, wether it
exceeds each lateral of the pressure tolerance or not.
Keywords: Software, Irrigation, PocketPC, LabVIEW
1. Introduction
Requirements in agricultural production, environmental protection and water resources
optimisation have been made farmers to modernize irrigation systems. One of these
modernisations is the installation of drip irrigation systems (López, 1996; Valiantzas, 2003).
Drip irrigations system is characterized by high water use efficiency. Another advantage of this
irrigation system is the precision in water and fertilizers application under adequate design
conditions (Pedras and Pereira, 2001; Holzapfel et al., 2001).
Hydraulic design of drip irrigation subunits is integrated in the drip line calculation. The
calculation of the drip line is carried out after the agronomic design. This design is based on
several data (characterization of choose emitter, field topography, etc). In order to design an
irrigation subunit is necessary to combine the hydraulic calculation (flow, diameters and
pressure of drip line and sub main pipes) with irrigation net distribution plane.
Specific agronomic features (plant frame, crop water demand…) are used in drip irrigation
systems design. The results of the design are the number and the distribution of the emitters
(Medina, 1997). As the crop is growing it is used to add new emitters in order to supply the
increasing crop water needs. In some cases, when the farmer wants to increase the plant
density or he decides to change the crop and to maintain the existing irrigation system he have
275
to adapt to the new situation. This is to change the number of emitters in drip line to the new
water requirements. On this case, if flow and pressure required at the beginning of drip line are
known, it is very useful to provide software that let it to know if drip line is ready to these
changes. It is necessary that this software must be quickly and precise. This software would
indicate immediately to the farmer what decision he would take.
Some software under Windows but not in mobile devices is used for agronomic and hydraulic
design in drip irrigation (Rodrigo and Cordero, 2003). One adequate solution is the installation
of developed software for mobile devices as Smartphone or PocketPC.
In the last years there is a lot of programming languages used in mobile devices. One of this is
LabVIEW, that it is a revolutionary system of graphics programming used for applications that
includes acquisition, control analysis and data presentation (Lajara and Pelegri, 2007). This
software is imposing in engineer applications because its great versatility and simplicity to use.
This paper shows developed software for this kind of Mobile devices using LabVIEW PDA as
programming language. This is a freeware, for this reason its use is limited to fed drip lines from
the extreme, with no slope terrain. This software could down from the next link:
http://decibel.ni.com/content/docs/DOC-1385.
2. Basis of calculation
Theoretical base of used formulas in the hydraulic design of drip irrigation is considering that
flow distribution in a drip line is coming close to a continue distribution. A description of the
calculation procedure is described in the next lines.
Head losses
hL (m) in a pipe of length L (m) has been calculated with the next equation:
ΔhL = F ·J*· L
(1)
Where,
F is the Christiansen’s reduction factor and J* (m. m-1) is the head losses coefficient that
includes lineal and singular losses.
The formula used to estimate the Christiansen’s reduction factor is:
F=
β −1
1
1
+
+
1+ β
2n
6 n2
(2)
Where the value of β is 1,75 in Blasius formula for polythene pipes and n is the number of
emitters in the dripline.
Unit head losses J*, have been determined using an empirical formula that includes the
singular head losses of the emitters by means the next expression:
J *= J
e + fe
e
(3)
Where e (m) is the distance between emitters and y fe (m) is the equivalent length of the emitter
that depends on the kind of the emitter and the diameter of the drip line.
Linear head losses, J (m·m-1), has been determined using the Blasius’s formula in polythene
pipes:
276
J = 0,473 ⋅
Q 1,75
D 4,75
(4)
where D (mm) is the pipe diameter and Q (l·h-1) is the flow.
In order to estimate the singular head losses, fe (m), has been considered table 1.
∅ interior (mm)
fe (m)
Big
Standard
Little
10,3
0,32
0,24
0,18
13,2
0,20
0,15
0,11
16,0
0,14
0,11
0,08
18,0
0,11
0,08
0,06
20,4
0,09
0,07
0,05
28,0
0,05
0,04
0,03
Table 1: fe values for upperline conexions
The pressure distribution in a drip line is showed in figure 1.
hmáx
ΔhL = J*·F·L
hmín = hu
Q
Figure 1: Pressure distribution in a drip line fed by the extreme in a plane zone.
The maximum of pressure coincides with the pressure in the begining of the drip line, hmax (m),
which is determined by means the next expression:
hmax = ha + 0.75 ΔhL
(2)
ha (m) is the average pressure in the drip line.
The minimum pressure, hmin (m), is the same pressure in the last emitter, hu (m), which is
determined by:
hmin =hu = hmax - Δh
(3)
If the difference of pressure hmax-hmin is under the value Δhp (m), the commercial diameters are
adequates.
277
3. Software
The graphic interface of the developed software consists of three flaps:
a) In the flap “Inicio” is introduced, at the top of the screen, the data for calculation and at the
lower part valid commercial diameters are showed the valid commercial diameters by means
illuminated leads (Figure 2). Required data are: nominal flow of the emitter qe (l·h-1), number of
emitters Ne, drip line length L (m), distance between emitters Ee (m), average pressure in the
drip line ha (m) and pressure tolerance hp (m). After the data introduction the button named
“Cálculos” have to be pressed in order to show the results in the frame.
Figure 2: Initial flap.
b) The flap named “Resultados” shows, in every commercial diameter, the head losses in the
drip line Dh (m), maximum pressure hmax (m) and minimum pressure hmin (m). At the lower part,
is showed a comparative graphics that indicates if the allowed pressure tolerance hp (m) is
exceed or not, for every drip line (Fig 3.).
Figure 3: Flap of results.
c) The last flap “Ayuda”, shows the contacting data for the authors.
278
The block diagram of this software is developed under while loop structure that allows ending
the software execution by means the button “Salir” (Figure 2). This button determines the true
and false values of a case structure included in the while loop. Pushing this button the case
structure takes the true value and the internal programming is carried out. In figure 4, is showed
the block diagram used to determinate if the pipe with a diameter of 13.2 mm could be used in
the drip line considering general data entry.
Figure 4: Block diagram to determinate the use of a pipe with a diameter of 13,2 mm in the drip line.
A similar block diagram (Fig. 3) has been developed for every commercial diameter (10.3; 13.2;
16.0; 18.0; 20.4 and 28.0). This diagram determines what diameters of commercial pipes could
be used for the drip line and the results could be compared.
After calculations, this could be showed in the console.
4. Conclusions
Software has been developed with LabWIEW PDA. This software allows installers and
engineers of drip irrigation companies to calculate in the field commercial diameters used in
drip lines.
This is an accessible freeware, contacting
http://decibel.ni.com/content/docs/DOC-1385
with
authors
or
from
the
link
This software version is limited to drip lines fed by the extreme in a horizontal land
References
Holzapfel, E. A. and Jara, J.; Matta, R., 2001. Nivel de agua aplicado y fertirrigación
bajo riego por goteo en cítricos, Agro-Ciencia. Chillan, Vol.17, pp.20-31.
Lajara Vizcaino, J.R. and Pelegri Sebastia, J., 2007. LabVIEW 8.20 Entorno Gráfico de
Programción, Marcombo S.A., Barcelona.
Lopez, J. R., 1996. Riego localizado II. Programas informáticos, Mundi-Prensa, Madrid.
Medina San Juan, J.A., 1997. Riego por Goteo, Mundiprensa (4ª ed), Madrid.
279
Pedras, C. M. G. and Pereira, L. S., 2001. A simulation model for design and evaluation
of micro-irrigation systems, Journal of Irrigation and Drainage, ASCE, New York, Vol.
50, n.4, pp.323-334.
Rodrigo López, J. and Cordero Ordóñez, L., 2003. Riego Localizado. Programas
Informáticos para Windows, Mundi-Prensa, Madrid.
Valiantzas, J. D., 2003. Explicit hydraulic design of microirrigation submain units with
tapered tanifold manifold and laterals, Journal of Irrigation and Drainage, ASCE, New
York, Vol.129, n.4, pp.227-236.
Acknowledgments
The authors would like to express their gratitude to National Instruments Spain, S.L. for its help
spreading the developed software in its website. (http://sine.ni.com/cs/app/doc/p/id/cs-11093).
José Miguel Molina Martínez.
Departamento de Ingeniería de los Alimentos y del Equipamiento Agrícola. Área de Ingeniería
Agroforestal, Universidad Politécnica de Cartagena, Paseo Alfonso XIII, 48, 30203. Cartagena. (Murcia)
España.
Phone: +34 968 32 59 29
Fax: + 34 968 32 57 32
Antonio Ruiz Canales
Departamento de Ingeniería. Área de Ingeniería Agroforestal. Escuela Politécnica Superior de Orihuela,
Universidad Miguel Hernández, Crtra. de Beniel, km 3,2, 03312 Orihuela (Alicante), España.
Phone: +34 966 74 96 39
Fax: + 34 966 74 97 18
280
ONTOLOGIES AND FBS FRAMEWORK
Cebrián-Tarrasón, D.
Vidal, R.
Universitat Jaume I
Abstract
In the past years there have been great advances in the development of ontologies applied to
the field of engineering design, essentially in functional and structural models. A great diversity
of methodologies for the analysis of the design process in relation to the FBS (Function Behaviour - Structure) framework have been modelled, thereby attaining a profound knowledge
of the subject. Hence, it is possible to carry this knowledge to the industrial domain through the
use of ontologies. In the article, an in-depth analysis of the existing methodologies based on
the FBS framework is carried out. Later, the focus is shifted onto the ontologies that were
based upon the aforementioned framework. From this review, a discussion of the ontologies is
made, pointing out their main virtues. The aim is to insist upon the benefits of ontologies and to
encourage their wider industrial application.
Keywords: ontology, conceptual design, FBS
1. Introduction
The end goal of Engineering Design is the conceptual creation of an object, product, system or
process that meets functional requirements in order to fulfill the customer’s needs in a
workable, economical, ecological and manufacturable way. The development of technologies
based on artificial intelligence and CAE (Computer Assisted Engineering) has facilitated access
to information related to the structure and form of objects, although design know-how used in
the conceptual design phase remains hidden due to its subjective nature and implicitness
(Kitamura and Mizoguchi, 2004).
The development of KBE (Knowledge-Based Engineering) systems is aimed at improving this
aspect (Cebrián-Tarrasón, Muñoz et al., 2007; Chulvi, Sancho et al., 2007). Nonetheless, the
full use of these technologies is limited by the impossibility of fully reusing and sharing
knowledge in KBE systems, along with the lack of common knowledge from which to create a
knowledge base and the limited success of methodologies for the extraction of knowledge
(Mizoguchi 2003; Baxter, Gao et al. 2007). In the field of engineering design, more and more
attention is being focused on the development of ontologies as a possible solution of the
aforementioned deficiencies of KBE systems (Ahmed and Wallace 2004; Ahmed and Storga,
2007). An ontology can be described as an explicit specification of a shared conceptualization,
which can be taxonomically or axiomatically based (Gruber, 1993). Ontologies can be based
around a single taxonomy or several taxonomies and their relationships. Taxonomies consist of
concepts and relationships that are organised hierarchically and whose concepts can be
281
arranged as classes with sub-classes (Cottam, Milton et al., 1998; Gómez-Pérez, FernándezLópez et al., 2004).
Abbreviation
Source
Description
A-Design
Campbell et al. (Campbell,
Cagan et al., 2003)
Agent-Based Approach to
Conceptual Design in a Dynamic
Environment
Axiomatic Design
Suh (Suh, 1990)
Axiomatic theory of design
FBRL
Sasajima et al. (Sasajima,
Kitamura et al., 1995)
Function and behaviour
representation language
FBS
Umeda et al. (Umeda,
Takeda et al., 1990)
Theory of objects described in
terms of function, behaviour and
structure
FBS
Gero et al. (Gero, 1990)
Theory of objects described in
terms of function, behaviour and
structure
FEBS
Deng et al. (Deng, 1999)
Function-Environment-BehaviourStructure modelling
FPPT
Klein Meyer et al. (Klein
Meyer, Cabannes et al.,
2007)
Function, Physical Principle and
Technology
Formal theory of objects and object
functionalities
FR
Chandrasekaran et al.
(Chandrasekaran 1990;
Chandrasekaran, Goel et
al., 1993; Chandrasekaran,
2005)
Reconciled Functional
Basis
Hirtz et al. (Hirtz, Stone et
al., 2002)
Taxonomy of flows and functions
HSA
Zhang et al. (Zhang, Tor et
al., 2002)
Heuristic State-Space Approach
NIST Design Repository
Szykman et al. (Szykman,
Racz et al., 1999)
Formal model of objects from
object-oriented perspective
FB
Pahl and Beitz (Pahl and
Beitz, 1984)
General theory of design activity
PFM
Roy et al. (Roy and
Bharadwaj, 2002)
General theory of objects
Theory of Technical
Systems
Hubka and Eder (Hubka
and Eder, 1988)
General non-formal theory of
objects
Table 1: Investigations related to the FBS Framework.
282
The structure of an ontology should be based on a taxonomy that allows for the modelling of a
system based one certain functional descriptions (Garbacz, 2006). In this way, a great diversity
of methodologies for analysis of design process based on the FBS framework have been
modelled (Umeda, Takeda et al. 1990; Hirtz, Stone et al., 2002; Umeda, Kondoh et al. 2005;
Borgo, Carrara et al., 2006).
Notwithstanding, Suh (Suh 1990) describes design as a zigzag between functional
requirements and design parameters and Veyrat et al. (Veyrat, Blanco et al. 2007) questions
whether design is actually a direct mapping between functions and structure. In order to resolve
this question, Cebrián-Tarrasón (Cebrian-Tarrason, Lopez-Montero et al., 2008) combines the
concept of the OntoRFB (ontologically clean Functional Basis), Garbacz (Garbacz, 2006) based
on DOLCE (Descriptive Ontology for Linguistic and Cognitive Engineering) (Masolo, Borgo et
al., 2003) and the model B-FES (Behaviour-driven Function-Environment-Structure) (Zhang,
Tor et al., 2003; Zhang, Tor et al., 2005) and create a new ontology with special emphasis on
the importance of behaviour as the link in the FBS framework.
The purpose of this paper is to show what the reasons are which a better application in the
industrial scope of the existent ontologies in the field of engineering design would allow. For
that, an updated analysis of the existent methodologies in the field of engineering design
referred to the FBS framework is carried out. Next how those methodologies have been related
to the field of ontologies is pointed out. From this cited analysis, a discussion about the different
improvements in the subject is carried out.
2. FBS framework
Though the terms function, behaviour and structure had been used before, it wasn’t until 1990
that they were clarified and used to define a framework to model and represent system
functionality (Gero 1990; Umeda, Takeda et al., 1990). In the FBS framework (FunctionBehaviour-Structure), function represents the functions that the system performs; structure
represents the physical elements of the solution and behaviour acts as the relationship between
F and S. In design synthesis, behaviour is derived from an intended functionality in order to
reach a solution from it. When a solution is defined, its behaviour is deduced from it to evaluate
if the solution reaches the intended functionality. Then, behaviour is related to the design
physical state, which may be static or variable with time.
The FBS framework can also be used as a methodology for analysing the design process,
through the representation of the evolution of design state from the analysis of design protocols
(Takeda, 1994). From a review of literature on design theories related to this topic, an updated
list of investigations carried out in Table 1 is formed (Garbacz, 2004; Camelo, 2007).
Roughly speaking, two approaches to FBS can be distinguished (Chandrasekaran, 2005). In
the first, one relates functions to behaviours of an element, and then relates these behaviours
to structural-physical descriptions of the elements. It was developed by Gero (Gero, 1990), who
proposed a Function-Behaviour-Structure model of designing, and by Umeda et al. (Umeda,
Takeda et al., 1990), who proposed a Function-Behaviour-State model of designing.
The first approach considers behaviour as a key concept that suggests a clear ontological
ordering: technical objects have their physical structure. This structure, in interaction with a
physical environment, gives way to the objects’ behaviours; and these behaviours then
determine in some way the objects’ functions (Borgo, Carrara et al., 2006).
Several investigations have been developed with respect to this approach. Mizoguchi et
al.(Mizoguchi 2003) use the FBRL (Function and Behaviour Representation Language) model
based in the work of Sasajima et al. (Sasajima, Kitamura et al., 1995) and expresses behaviour
as a conceptualization of the change of attribute values in the spatio-temporal space over time.
Also, they consider that function is a teleological interpretation of behaviour under a given goal.
283
Klein Meyer et al. (Klein Meyer, Cabannes et al., 2007) proposed the Function, Physical
Principle and Technology (FPPT) theory which states, with respect to functional analysis and
technology selection, that modelling concepts are based on FBS approach with the objective of
allowing behavioural analyses at the beginning of the design process, when no geometry is
available.
The possibilities of search, exploration, combination and selection systems based on FBS
representation have increased thanks to the B-FES model proposed by Tor et al. (Tor 2000;
Tor, Britton et al., 2002; Zhang, Tor et al., 2005). This model is an extension and a refinement
of the dual-step function environment-behaviour-structure (FEBS) modelling framework by
Deng et al. (Deng, 1999).
In the second approach, one models functions of objects in terms of inputs and outputs, and
then relates these functions directly to structural-physical descriptions of objects (Szykman,
Racz et al., 1999). It is also known as Functional Modelling because it considers behaviour as a
mathematical representation of the states of a mechanism (Pahl and Beitz, 1986; Szykman,
Racz et al., 1999; Hirtz, Stone et al., 2002).
Functional Modelling is created with the aim of reducing the existent ambiguity at the level of
modelling of an object. For that, some discrimination between the meaning of function and flow
is carried out. Flows are divided in three types mainly: material, energy and signal. That
definition comes from the previous work of the Value Analysis and later from Pahl and Beitz
(Pahl and Beitz, 1986) because they inspire the classification of functions in classes. Thus the
joint of functions and flows is called functional basis. Later it was improved with the name of
Reconciled Functional Basis (RFB) (Szykman, Racz et al. 1999; Hirtz, Stone et al., 2002).
From that revision, a priori great amount of works developed in the field of engineering design
were applied to the FBS framework. Likewise the diversification of the grasp between the
several researches in the matter is perceived.
3. Ontologies in Engineering Design
In this section, we briefly discuss ontologies in the domain of engineering design. Generally,
ontologies can be categorized by the subject of the conceptualization (Gómez-Pérez,
Fernández-López et al., 2004), among others:

Top-level ontologies or Upper-level ontologies describe very general concepts (e.g.
substance, tangible, intangible) and provide general notions under which all root terms
in existing ontologies should be linked. The aim is to get a great number of ontologies
based in an upper-level ontology.

Domain ontologies are reusable in a given specific domain (engineering, manufacturing,
design, etc). These ontologies provide vocabularies about concepts within a domain and
their relationships, the activities taking place in that domain, and the theories and
elementary principles governing that domain.

Task ontologies describe the vocabulary related to a generic task or activity (e.g.
diagnosing, scheduling). They provide a systematic vocabulary of the terms used to
solve problems associated with tasks that may or may not belong to the same domain.

Application ontologies are application-dependent. They contain all the definitions
needed to model the knowledge required for a particular application. Application
ontologies often extend and specialise the vocabulary of the domain and of task
ontologies for a given application.
There is a great diversity of languages for ontology representation, each one with several
capacities of knowledge representation and reasoning mechanisms, although the most
284
frequently used is OWL (Ontology Web Language). Nonetheless, the capacity of the restriction
representation in OWL is not fitted for the implementation of relation in the field of design (Kim,
Manley et al., 2006). With that aim SWRL (Semantic Web Reasoning Language) is used which
is a language based in a combination of OWL sublanguages with the rules language RML (Rule
Markup Language). The tool which is more used for its development is Protégé (Protégé3.4,
2008), an environment for the development of ontologies, based in Java and open source,
created by Standford University.
In the domain of engineering design, such ontologies can be categorized into task ontologies
and domain ontologies (Cebrian-Tarrason, Lopez-Montero et al., 2008). Table 2 contains a
review of engineering design literature related to ontologies, carried out by Kitamura (Kitamura
and Mizoguchi, 2004), in order to relate it to the field of FBS framework and indicating the type
of ontology, where a description of the more relevant ontologies is carried out, pointing out the
type of ontology and its language.
Abbreviation
Source
SOFAST®
Kitamura
et
al.
(Kitamura, Sano et al.,
2002;
Kitamura
and
Mizoguchi,
2003;
Kitamura and Mizoguchi,
2004; Kitamura, Washio
et al., 2006)
Masolo et al. (Masolo, Descriptive
Borgo et al., 2003)
Ontology
for
Linguistic
and
Cognitive
Engineering
Kim et al. (Kim, Manley Assembly Design
et al., 2006)
Ontology
Garbacz
(Garbacz, Taxonomy
of
2006)
artefact functions
Ahmed et al. (Ahmed, Engineering
2006)
Design Integrated
Taxonomy
Storga et al. (Storga, Design Ontology
Andreasen et al., 2005)
Van Rennsen et al. (Van A taxonomy of
Rennsen, Vermaas et functions
on
al., 2007)
Gellish English
Liu et Bason (Liu and Distributed
Bason, 2007)
Design Assistant
Bryant Arnold et al. A Function-Based
(Bryant-Arnold, Stone et Component
al., 2007)
Ontology
For
Systems Design
DOLCE
AsD
OntoRFB
EDIT
DO
Gellish
English
DiDeas II
FB Ontology
Description
Type
of Ontology
ontology
language
Domain
Non-formal
ontology
of ontology
objects in the
Hozo
FBRL language
Upper-level
ontology
OWL
Task
ontology
Domain
ontology
Domain
ontology
Domain
ontology
Task
ontology
Domain
ontology
Domain
ontology
OWL/SWRL
OWL
OWL
OWL
OWL
OWL
OWL
Table 2: Ontologies in the domain of engineering design.
As one of the principal ontologies of the Semantic Web, DOLCE provides a common reference
framework for ontologies in order to facilitate sharing of information among them. In its
representation, DOLCE aims at capturing ontological categories underlying natural language
and human common-sense (Masolo, Borgo et al., 2003).
285
Gero, who improved his former study of FBS representation, establishes the basis for the
computational modelling of process to support the design process based on the FBS
representation (Gero and Kannengiesser, 2002; Gero and Kannengiesser, 2004). He also
developed the term “FBS Ontology” to refer to his model.
Engineering Design Integrated Taxonomy (EDIT) (Ahmed 2006) consists of several taxonomies
and their relations. As the integrated taxonomy is populated with instances, the relationships
between concepts (or multiple concepts) are captured and the ontology emerges. EDIT
ontology was developed for the purpose of indexing, searching and retrieving design
knowledge (Ahmed, 2006).
The DO (Design Ontology) is related to the before mentioned work (Ahmed and Storga 2007).
Storga et al. (Storga, Andreasen et al., 2005) created the DO as a potential formal description
of the shared engineering knowledge in design domain. Along these lines, Ahmed and DO
(Ahmed and Storga, 2007) created a comparison between both, where the DO is described as
ontology conceived to describe design as a product. On the other hand, the EDIT was
established design as an activity, incorporating the process as well as the product.
Eventually, several works of special relevance are pointed out. The first is based in a
exhaustive study of the different design theories, where Garbacz (Garbacz, 2006) shows
OntoRFB, a taxonomy of artifact functions, based in DOLCE and RFB and analyzed from a
perspective of philosophical logic. On the other hand, AsD ontology (Kim, Manley et al., 2006)
is the one which uses OWL as well as SWRL applied to the field of manufacturing. Finally,
applied to the industrial field, the ontology developed by Kitamura et al. (Kitamura, Washio et
al., 2006) is the only one that manages to develop a domain ontology and apply it to a practical
example by means of the SOFAST® program.
4. Discussion
Through the ontology discussion, a set of improvements in the developed works can be
observed in order to achieve a better application of ontologies to the industrial field.
In FBS framework, the first possible improvement comes from the existence of two different
approaches differentiated in its first definition. This situation has allowed advances in the
researches carried out in this field and also allows a great margin of progress. Specifically, the
concept of behaviour could be studied in depth in the approach related to the functional
modelling in order to achieve some efficient links between those structures whose behaviour is
necessary to relate to the functional area (Vermaas and Dorst, 2007). This progress would
allow a stable conceptual scheme since it boosts the connexion with the definition of the
function, behaviour and structure areas (Chandrasekaran, 2005; Borgo, Carrara et al., 2006;
Vermaas and Dorst, 2007)
As known solutions in this field and due to the great interest which is being developed in
ontologies, several authors, who come from the branch of philosophy, are planning formal
schemes with the aim of proposing solutions. Those works provide a new point of view since
their studies are based in logical formalism (Garbacz, 2004; Borgo, Carrara et al., 2006;
Garbacz, 2006; Van Rennsen, Vermaas et al., 2007; Vermaas and Dorst, 2007).
This possibility of improvement increases even more when an ontology is set out, based in FBS
framework. As it has been pointed out, the definition of an ontology involves the conciseness
and explicitness of all the selected knowledge, which a precise formalization implies. The
mentioned formalization constitutes a task where a lot of work lies ahead. Its consecution would
allow the optimization of the industrial design process since all the knowledge about a product
would be available. This fact would suppose a considerable reduction of time and costs(Ahmed,
2006; Vermaas and Dorst, 2007). One example is the work of Kitamura (Kitamura, Washio et
286
al., 2006), who could apply his complex model in a production chain of a Japanese company
with great success.
Other element that has allowed the advance of the application of the ontologies in the field of
engineering design, has been the progressive adoption of the definition of ontology considered
nowadays in computer science (Storga, Andreasen et al., 2005; Garbacz, 2006; Ahmed and
Storga, 2007; Liu and Bason, 2007; Van Rennsen, Vermaas et al., 2007; Vermaas and Dorst,
2007). It has to be considered that some years ago, Gero coined the term "FBS Ontology" in
order to refer to his model (Gero and Kannengiesser, 2004), without developing a taxonomy
based in that.
There is criteria unification in the level of representation of ontologies because the majority of
ontologies developed in this field (Table 2) make use of OWL language. This standard
language allows for the use of logic which allows the inference of new knowledge by means of
queries based on reasoning (Antoniou, Bikakis et al., 2008). However, only the AsD ontology
(Kim, Manley et al., 2006) makes use of this application to the area of manufacturing, and is
also the only one to link the results of the ontology with a CAD program.
In conclusion, in a short-term the field of ontologies applied to engineering design could be
considered with great possibilities of progress and capacity of application in the industrial field.
The increase of applications based in ontologies related to FBS framework, practical examples
in industry (Vermaas and Dorst, 2007), a better homogeneity or depth of the abilities of an
ontology are aspects which would allow the success (Cebrian-Tarrason, Lopez-Montero et al.,
2008).
5. Conclusion
This work has been presented with the aim of showing some means in order to incentive the
industrial application of the ontologies developed in the field of engineering design. This study
analyses several existing schemas around FBS framework as well as ontologies based in that
mentioned framework.
The study of FBS framework shows the great advance achieved in the investigations carried
out in this field although an improvement in the connection in the definition of the sections of
function, behaviour and structure. Respect to the methodologies in the field of engineering
design, the inference of new knowledge is observed as the great advantage of the application
of the ontologies.
Other remarkable fact in this analysis is the great possibility of expansion of ontologies in the
field of engineering design, waiting a great peak in its application in the industrial level of this
technology in next future.
The achieved results of this study establish the basis for the building of an ontology which could
combine advantages of the researches discussed previously. With that aim, OntoFaBeS
(Cebrian-Tarrason, Lopez-Montero et al., 2008), is being developed for its application to the
industrial level.
In next researches is expected to develop OntoFaBeS through other areas of engineering
design as environment field (Garraín, Vidal et al., 2007), evaluation criteria (Justel, Vidal et al.,
2007) or creative techniques (Chulvi and Vidal, 2007).
287
References
Ahmed S. and Wallace K., 2004. Reusing design knowledge, 14th CIRP Desing Seminar,
Cairo.
Ahmed S.,2006. A methodology for creating ontologies for engineering design. Journal of
computing and Information science in engineering, Vol.7 (2), pp.132-140.
Ahmed S. and Storga M., 2007. Engineering design ontologies - contrasting an empirical and a
theoretical approach, International Conference on Engineering Design, ICED’07, Cite des
Sciences et de l'Industrie, Paris, France.
Antoniou G., Bikakis A., Dimaresis N., Genetzakis M., Georgalis G., Governatori G., Karouzaki
E., Kazepis N., Kosmadakis D., Kritsotakis M., Lilis G., Papadogiannakis A., Pediaditis P.,
Terzakis C., Theodosaki R. and Zeginis D., 2008. Proof explanation for a nonmonotonic
semantic web rules language, Data & Knowledge Engineering, Vol.64, pp.662-687.
Baxter D., Gao J., Case K., Harding J., Young B., Cochrane S. and Dani S., 2007. An
engineering design knowledge reuse methodology using process modelling, Research in
engineering design, Vol.18, pp.37-48.
Borgo S., Carrara M., Vermaas P.E. and Garbacz P., 2006. Behaviour of a technical artifact: an
ontological perspective in engineering, Frontiers in Artificial Intelligence and Applications,
Vol.150 (Formal Ontology in Information Systems), pp. 214–225.
Bryant-Arnold C.R., Stone R.B., Greer J.L., Mcadams D.A., Kurtoglu T. and Campbell3 M.I.,
2007. A function-based component ontology for systems design, 16th International Conference
on Engineering Design, Paris, France.
Camelo D., 2007. Modelado y desarrollo de un modelo computacional de síntesis interactivo y
multirrelacional para guiar la actividad de diseño en la fase conceptual, Dpto. de Engeniería
Mecánica y Construcción, Universitat Jaume I, Castellón, Spain.
Campbell M., Cagan J. and Kotovsky K., 2003. The a-design approach to managing automated
design synthesis, Research in Engineering Design, Vol.14, pp.12 - 24.
Cebrián-Tarrasón D., Muñoz C., Chulvi V. and Vidal R., 2007. Nuevo enfoque en el diseño
inteligente de implantes craneales personalizados a través de KBE, XI Congreso Internacional
de Ingenieria de Proyectos (AEIPRO), Lugo, pp.668-677.
Cebrian-Tarrason D., Lopez-Montero J.A. and Vidal R., 2008. OntoFaBeS: Ontology design
based in FBS framework, CIRP Design Conference 2008, University of Twente.
Chandrasekaran B., 1990. Design problem solving: A task analysis, AI Magazine, Vol.11, pp.
59-70.
Chandrasekaran B., 2005. Representing function: relating functional representation and
functional modeling research streams, AIEDAM Artificial Intellig. for Engineering Design, Vol.19
(2), pp.65-74.
Chandrasekaran B., Goel A. and Iwasaki Y., 1993. Functional representation as design
rationale.
Chulvi V. and Vidal R., 2007. Vinculación de TRIZ con el diseño funcional, XI Congreso
Internacional de Ingeniería de Proyectos (AEIPRO), Lugo, pp.613-621.
Chulvi V., Sancho A., Cebrian D., Gimenez R., Muñoz C. and Vidal R., 2007. KnowledgeBased Engineering in cranioplasty implant design, 16th International Conference on
Engineering Design, Paris, France.
288
Cottam H., Milton N. and Shadbolt N., 1998. The use of ontologies in a decision support system
for business process re-engineering, Proceedings of IT & KNOWS Conference of the 15th IFIP
World Computer Congress, OCG, Vienna, Autriche, Vienna, Budapest, pp. 76-89.
Deng Y.M., 1999. A computerized design environment for functional modeling of mechanical
products, fifth ACM symposium on Solid modeling and applications, ACM Press, Ann Arbor,
Michigan, pp.1-12.
Garbacz P., 2006. Towards a standard taxonomy of artifact functions, Applied Ontology, Vol.1
(3), pp.221-236.
Garbacz P., 2004. Subsumption and relative identity, Axiomathes, Vol.14 (4), pp.341-360.
Garraín D., Vidal R., Martinez P., Franco V. and Cebrian-Tarrason D., 2007. LCA of
biodegradable multilayer film from biopolymers, 3rd International Conference on Life Cycle
Management, Zurich.
Gero J.S., 1990. Design prototypes: a knowledge representation schema for design, AI
magazine, Vol.11 (4), pp.26 - 36.
Gero J.S. and Kannengiesser U., 2002. The situated Function-Behaviour-Structure-Framework,
Artificial Intelligence in Design'02, pp.89-104.
Gero J.S. and Kannengiesser U., 2004. Modelling expertise of temporary design teams, Journal
of Design Research, Vol.4 (3).
Gero J.S. and Kannengiesser U., 2004. The situated Function-Behaviour-Structure Framework,
Design Studies, Vol.25 (4), pp.373-391.
Gómez-Pérez A., Fernández-López M. and Corcho O., 2004. Theoretical foundations of
ontologies - Chapter 1, Springer, ed., Ontological Engineering with Examples from the Areas of
Knowledge Management, E-Commerce and the Semantic Web, London, 1-44.
Gruber T.R., 1993. Toward principles for the design of ontologies used for knowledge sharing,
Knowledge Systems Laboratory, Stanford University., Padova, Italy.
Hirtz J., Stone R., Mcadams D., Szykman S. and Wood K., 2002 A functional basis for
engineering design: reconciling and evolving previous efforts, Research in Engineering Design,
Vol.13, pp. 65-82
Hubka V. and Eder W.E., 1988. Theory of technical systems, Springer-Verlag, Berlín.
Justel D., Vidal R., Arriaga E., Franco V. and Val-Jauregi E., 2007. Evaluation method for
selecting innovate product concepts with greater potential marketing success, 16th International
Conference on Engineering Design Paris, France.
Kim K.-Y., Manley D.G. and Yang H., 2006. Ontology-based assembly design and information
sharing for collaborative product development, Computer-Aided Design, Vol.38 (12), pp.12331250.
Kitamura Y., Sano T., Namba K. and Mizoguchi R., 2002. A functional concept ontology and its
application to automatic identification of functional structures, Advanced Engineering
Informatics, Vol.16 (2), pp.145-163.
Kitamura Y. and Mizoguchi R., 2003. Ontology-based description of functional design
knowledge and its use in a functional way server, Expert Systems with Applications, Vol.24 (2),
pp.153-166.
Kitamura Y., Washio N., Koji Y., Sasajima M., Takafuji S. and Mizoguchi R., 2006. An
Ontology-based annotation framework for representing the functionality of engineering devices,
ASME 2006 International Design Engineering Technical Conferences & Computers and
Information in Engineering Conference, Pennsylvania, USA.
289
Kitamura Y. and Mizoguchi R., 2004. Ontology-based systematization of functional knowledge,
Journal of Engineering Design, Vol.15 (4), pp.327-351
Klein Meyer J.S., Cabannes G., Lafon P., Troussier N., Roucoules L. and Gidel T., 2007.
Product modeling for design alternatives selection using optimization and robustness analysis.
16th International Conference on Engineering Design, Paris, France.
Liu Y. and Bason A.H., 2007. An ontology based approach to a flexible aid for mechanical
conceptual design. Proccedings of the 16th International Conference on Engineering Design
ICED 07, Paris, France.
Mizoguchi R., 2003. Tutorial on ontological engineering - part 1: introduction to ontological
engineering, OhmSha&Springer, ed., New Generation Computing, 4, pp. 365-384.
Pahl G. and Beitz W., 1984. Engineering Design, Design Council, 1984.
Pahl G. and Beitz W., 1986. Systematic approach to the design of technical systems and
products, VDI Society for Product Development, Design and Marketing, Berlin.
Protégé3.4, 2008. Stanford University http://protégé.standford.edu.
Sasajima M., Kitamura Y., Ikeda M. and Mizoguchi R., 1995 FBRL: A Function and Behavior
Representation Language, Proceedings of IJCAI, pp.1830-1836.
Storga M., Andreasen M.M. and Marjanovic D., 2005. Towards a formal design model based on
a genetic design model system. Proceedings of the 15th International Conference on
Engineering Design ICED 05, Melbourne, Australia.
Szykman S., Racz J. and Sriram R., 1999 The Representation of function in Computer-Based
Design, Desing Engineering Technical Conferences, ASME, Las Vegas, Nevada.
Suh N.P., 1990. The Principles of Design, Oxford University Press, New York.
Takeda H., 1994. Towards multi-aspect design support systems, Technical Report NAIST-ISTR94006, Nara Institute of Science and Technology, Nara, Japan, 1994.
Tor S.B., Britton G.A., Zhang W.Y. and Deng Y.M., 2002 Guiding functional design of
mechanical products through rule-based causal behavioural reasoning, International Journal of
Production Research, Vol.40 (3), pp.667-682.
Tor S.B., 2000, Design automation of two-stage collapsible core using design prototype,
International Journal of Computer Integrated Manufacturing, Vol.13 (1), pp.31-39.
Umeda Y., Takeda H., Tomiyama T. and Yoshikawa H., 1990. Function, Behaviour, and
Structure, Gero J., ed., Applications of Artificial Intelligence in Engineering V, 1 Springer, Berlin,
pp. 177-194.
Umeda Y., Kondoh S., Shimodura Y. and Tomiyama T., 2005. Development of design
methodology for upgradable products based on Function-Behavior-State Modeling, Artificial
Intelligence for Engineering Design, Analysis and Manufacturing (AIEDAM), Vol.19, pp.161182.
Van Rennsen A., Vermaas P.E. and Zwart S.D., 2007. A taxonomy of functions in gellish
english. Proceedings of the 16th International Conference on Engineering Design ICED '07,
Paris, France.
Vermaas P.E. and Dorst K., 2007. On the conceptual framework of John Gero's FBS-Model
and the prescriptive aims of design methodology, Design Studies, Vol.28 (2), pp.133-157.
Veyrat N., Blanco E. and Trompette P., 2007. When shape does not induce function: why
designers must not lose the big picture, 16th International Conference on Engineering Design,
Paris, France.
290
Zhang W.Y., Tor S.B. and Britton G.A., 2003. Funcdesigner a Functional Design Software
System. The International Journal of Advanced Manufacturing Technology I, Vol.22, pp.295305.
Zhang W.Y., Tor S.B. and Britton G.A., 2005. A graph and matrix representation scheme for
functional design of mechanical products, The International Journal of Advanced Manufacturing
Technology, Vol.25 (3-4), 2005, pp.221-232.
Zhang W.Y., Tor S.B. and Britton G.A., 2002. A Heuristic State-Space Approach to the
functional design of mechanical systems, The International Journal of Advanced Manufacturing
Technology, Vol.19 (4), pp.235-244.
Acknowledgements
The authors are greatly appreciative of the funding provided by the Spanish Ministry of
Education and Science (project DPI2006-15570-C02-01, with FEDER funds). They would also
like to thank the rest of the Engineering Design Group (GID) at the University Jaume I in
Castellón (Spain) for their support.
David Cebrián Tarrasón
Av. Sos Baynat, s/n. E-12071 Castellón.
Tel. +34964729252 Fax +34964728106
Rosario Vidal Nadal
291
DEVELOPMENT OF PROJECT BASED LEARNING TECHNIQUES BY
THE IMPLEMENTATION OF INFORMATION TECHNOLOGIES
Mesa, J. M.
Villanueva, J.
Rodríguez, V.
Rodríguez, F.
University of Oviedo
Abstract
The Project Engineering Area of the University of Oviedo uses, in order to obtain an important
part of its learning objectives, the development of practice projects by groups of students. This
technique is called Project Based Learning (PBL). From the learning point of view, the objective
of this technique is to promote the teamwork and to stimulate the realization and evaluation of
technical judgment on projects related with every technical discipline. In this paper, a project of
educational innovation sponsored by the University of Oviedo and developed during the period
2007-2008. Such Project aims to provide the students with the necessary tools to develop
those activities through the use of information technologies and it is structured in two main
blocks: firstly a set of tools for the complete management of Project teams has been
customized and secondly a virtual library for the use of the students with previous Project
examples was designed and installed.
Keywords: Project Based Learning (PBL), Information Technology (IT), Innovation in education
1. Introduction
This paper describes de development of a project approved on the 7th Call for Innovation
Projects (2007-2008) summoned by the Vice Chancellorship for Quality, Planning and
Innovation of Oviedo University.
The Oviedo University Project Engineering Area uses group-based learning techniques for the
practices of the Project Management subject following the Project Based Learning methodology
(PBL) (De Graaff, 2003), to achieve a very significant part of the learning objectives of the
practicum for the subject on Project Management.
The Practicum Projects aim to reflect as close as possible the professional work of the
students, with special encouragement to team–based work, analysis and discussion. Therefore,
both internal coordination and work planning are essential for success. Individually speaking,
the development of the project means taking responsibilities which affect directly on the work of
every group, on searching information and finding of solutions (Palmer, 2003; Pulko, 2003).
292
Students spend much of their working time on searching, managing and sharing a great
amount of information and documentation within each group (Mesa, 2007). This work also
requires the use of a series of tools for its internal management, so as to document decisionmaking, assign duties and responsibilities, and to control the development of the project.
Before this project was developed, communication among group-members was mainly via email, which resulted in difficulties with websites and sending files. To avoid this, students would
create internet groups in websites such as Google or Yahoo. However, these sites do not meet
the specific requirements to develop a practicum project, since they are space-limited. This
issue among others makes monitoring difficult for the tutor.
On the other hand, the use of former practicum projects in paper format as reference for current
students faces some problems:
•
It is not available any time required for students. It is therefore subjected to a restricted
schedule.
•
It gets damaged gradually, resulting in untidiness or even loss of documents.
•
It demands a great storage capacity.
•
It requires periodical updating and management to keep the standards.
•
It makes it difficult to be checked for corrections or comments by the academic staff.
2. Objectives
Facing the above mentioned difficulties, the purpose of this Learning Innovation Project was to
provide the students with the necessary tools they may need to perform these activities by
using Information Technologies (IT). As well as the working environment in real engineering
companies has implemented the use of new tools and working methods, the teaching staff must
also be adapted to the possibilities that the Information Technologies can offer.
It dealt with developing an environment adapted to the use of methodology of the Project Based
Learning and to the specific competence and professional skills demanded for the Project
Engineering Area subject included in the project. Therefore, the objectives are as follows:
•
To provide each group with a web site for managing, storing and sharing files.
Learning how to manage this tool is a goal for this subject, since it will be essential for
success in their future professional careers.
•
To improve coordination and communication within groups out of monitoring
meetings, providing communication tools such as forums and chats. This allows
exchanging any kind of information or opinion among students and teachers, thus
improving collaboration in the learning experience.
•
To contribute with a virtual tutorial tool for teachers; this allows the groups to
make the most of doubts or clarifications made by teachers.
•
To serve as a tool to develop evaluation criteria and personal tutor sessions. It
provides information on the performed activities and implication of each student in the
group.
•
A more efficient management of groups in terms of access, grants, modifications,
notices, etc.
293
•
A tool restricted to the teacher staff, which allows them the assessment of
practice groups. Both the tutor and the examining board will enter their assessments as
well as their personal comments on presentation and documentation.
As regards the consulting material, it is suggested:
•
To develop a Virtual Library to store all Practicum Projects performed in the
previous years. Such information, once evaluated and checked, will be available for
future students as on-line consulting material to access during practices. This
information should be available under student registration and log-in. It is also essential
that the uploaded documents would be protected against editing.
•
The compilation of rules, regulations, journals, specific bibliography, price tables
and similar documents, as well as links to different useful web pages. Added and
updated contents available every year.
3. Methodology and development
The first basic requirement of the project is the data-processing equipment and corresponding
software. The Learning Innovation Project is structured into two major parts. Firstly, a set of
tools for the Project Groups Management and secondly, a series of utilities based upon the
development of a Virtual Library for this subject.
Regarding the software selection, there is plenty of teaching experiences using Educational
Platforms such as WebCT (Manning, 2004) o Moodle (Moreno, 2007) to improve the teaching
quality in Higher Education. However, in this particular case, a specific tool for Project
Management was chosen (Microsoft Sharepoint), so as to favour the approaching to a real
professional working environment. Such a tool is normally used by companies to manage their
engineering projects; therefore students must learn how to use it properly in order to pass the
subject. That is the reason why it is important not to customize the system just as an
educational tool, but also as a professional one. This software allows companies to build a web
portal which enables different users (students in this case) to interact with each other and
create work groups. At the same time, it brings the opportunity to create work groups in their
own internet portal, allowing them to manage and control all information available on the site.
Besides, SharePoint tool allows users to work with Microsoft Office Applications and web
explorers to create, manage and share information among the members of an organization.
Therefore, this tool fulfils the necessary requirements besides being the best integrated with
Microsoft Office Programs, which is the most popular software used to create project
documents.
Afterwards, Microsoft Sharepoint Tools were customized and adapted to the project
requirements. After a trial period of its functionalities, the system was eventually adapted to the
development of Projects Subject on Industrial Engineering.
A Users Guide for students was also developed, handed along with the Practice Guide at the
beginning of the term, so as to provide the students with all the necessary information and
guidelines to help them make a proper use of this application. This guide is also available in
digital form to help immediate access online.
Detailed below is the teaching material developed in this project:
The User Guide was first developed. It covered the following points:
294
Learning objectives agreed for this practice.
Meetings and deliverables schedule.
Practice methodology and dynamics
Evaluation criteria
Available resources
A website model was developed to be reproduced by each practice group. Each group was
provided with the following elements online after validation:
•
Documentation area
–
Shared documents: stores useful external documents
–
Final documentation: stores the final documents for the project.
–
Internal management: records the minutes and other internal documents
–
Temporary documents: stores documents made by the students of the
practice group.
•
Listing area
–
Schedule: to arrange meetings and to manage dates for documents and
presentations.
–
Project planning: project implementation planning.
–
Contact information: stores contact details (phone number, email) of each
group member.
–
•
Web links: gathers web addresses useful for the project development.
Discussion area:
–
Group discussion: for internal group discussing and decision-making
–
Tutor Query: doubts and questions for group tutors.
Illustration 1 shows the developed work environment:
295
Figure 1: Web Site for the Industrial Engineering Practicum.
Taking into account the availability of the Visual Library from every Microsoft Sharepoint
environment, it is expected to grow and develop for the academic years to follow until it is
completed and filled with the following parts currently established:
•
Common errors: where the most frequent mistakes are included, taken from practice
project documentation.
•
Legislation and regulation: it compiles norms and regulation sorting them by sector and
coverage.
•
Example documents: it gathers different real project documents to be used by students
as practice reference.
•
Links: links to some useful web pages for the development of the project.
•
FAQ: frequently asked questions
As indicated, new sections and contents will be added periodically depending on necessities.
Illustration 2 shows.
296
Figure 2: Virtual Library.
The use of this application as working environment in the first term of the course has been
positive, especially the use of some of its tools. Both Documentation Area and Discussion
Groups have been generally in use; the latter was used for internal coordination as well as for
making any queries to the tutors.
Among the least used tools, were the internal planning and schedule tools. It is here where
future attention should be paid to improve its use.
4. Conclusions
The academic team involved in this teaching innovation project considers the application of
these developed tools to be a remarkable improvement in the quality of teaching Projects
Subject. It is also properly integrated in the use of the methodology of Based on Project
Learning. Finally, some relevant conclusions are detailed below:
• The implementation of Information Technologies encourages team work and helps internal
coordination among students.
• It provides a virtual tutorial tool where questions get registered to be consulted afterwards
by fellow group students.
• Apart from providing monitoring through weekly meetings, the web platform provides the
teacher with relevant information about the students’ virtual work.
• Though most of the students were generally positive about the use of the web platform for
the practices, some of them were reluctant about the uploading of internal work group
documents.
Finally, it is essential to remark the achievement of all the main objectives of the subject
practices, emphasizing the performance of some remarkable final projects from some of the
groups.
297
References
De Graaff E. and Kolmos A., 2003. Characteristics of problem-based learning, International
Journal of Engineering Education, 19, pp. 657 - 662
Manning, C. K., 2004. Web CT discussions in problem based learning (PBL): analysis of use by
representative groups in an introductory food science course, Journal of the American Dietetic
Association, 104, Supplement 2, 40-341.
Mesa J. M., Ortega F., Villanueva J., Álvarez V., 2007. Aplicación de la metodología PBL a la
enseñanza de la dirección de proyectos de ingeniería, 1ª Jornada Internacional UPM sobre
Innovación Educativa y Convergencia Europea
Moreno, L., Gonzalez, C., Castilla, I., Gonzalez, E., and Sigut, J., 2007. Applying a
constructivist and collaborative methodological approach in engineering education, Computers
& Education, 49, 3, 891-915.
Palmer S. R., 2003. Framework for undergraduate engineering management studies, Journal of
Professional Issues in Engineering Education and Practice, 129, pp. 92 – 99.
Pulko S. H. and Parikh S., 2003. Teaching 'soft' skills to engineers, International Journal of
Electrical Engineering Education, 40, pp. 243 - 254
José Manuel Mesa Fernández
University of Oviedo, Project Engineering Area.
C/ Independencia, 13, 33004 Oviedo, Asturias, Spain
Phone: +34 985 10 42 72
Fax: +34 985 10 42 56
URL: www.api.uniovi.es
298
TECHNIQUES OF REMOTE DETECTION IN PRECISION
VITICULTURE
Beltrán, P.
Montesinos, S.
Geosys S.L.
Abstract
Since 1999 several commercial satellites capable of taking very high resolution images, have
had the capacity to obtain information about any object on the earth's surface larger than 60
cm. Such satellites orbit the Earth and supply customized information (the user chooses how,
when and where the satellite acquires the image). The satellite images record the interaction
between vineyards by means of the electromagnetic radiation of the visible and near-infrared
regions of the electromagnetic spectrum. These interactions are directly related to the
greenness of the plants, the photosynthetic activity rate and the biomass of the plant.
Information from satellite images can be immediately used to improve vineyard management
because it allows us:
i) To assess the current variability of exploitation during critical periods of vegetative growth,
and
ii) To establish areas with homogeneous characteristics in their exploitation
Information obtained before harvesting allows improved planning of the work and a more
selective way of carrying out the process.
An integrated analysis, using data from the plant-climate-soil sensors and information from
satellite images, provides continuous spatial-temporal information for agricultural management
aiming to achieve greater homogeneity of the area under cultivation.
Keywords: Precision growing viticulture, teledetection, remote sensors, plant-climate-soil
sensors
1. Introduction
The vine growing and wine making industry has experienced substantial growth since the
1990s. The wine producing countries of the New World (Australia, the United States, South
Africa, Chile and New Zealand) have undergone great change, increasing their yield and
aligning production towards international markets. As a consequence, the sector becomes
every day more competitive (figure 1).
By practising precision viticulture, countries such as Australia, Chile and the United States have
299
achieved significant improvement in the quality of their wines and also sustained very
competitive levels of productivity. Such an approach regards the variation within a vineyard as
being the result of factors intrinsic to the place of production, such as climate, topography and
the nature of the soil, plus extrinsic factors like irrigation, fertilization and other agricultural
practices.
Traditional vine growing methods are uniform over a single area and do not take into account
the variability in factors that impinge upon production. The process is not maximally efficient
since different areas are treated in the same manner, leading to economic and environmental
losses.
The principal economic impact of precision viticulture is an increase in the quantity of premium
quality wine and a reduction in production costs.
USA
Italy France
Spain
Australia
Chile
South Africa
New Zealand
Germany
Portugal
Hungary
Bulgary
Competitivity
Aggressivity
Figure 1: Illustration of world wine production (Montesinos y Quintanilla, 2006).
Australia and California (USA) have been pioneers in the application of precision viticulture,
each adopting a different approach.
In California, the emphasis has been upon remote sensors, or what amounts to the same thing:
obtaining multispectral images of the vines from satellites and planes (Johnson et al. 1996 and
Nemani et al. 2001).
In Australia, where the wine industry is highly mechanised, data is collected via GPS (Global
Positioning Satellite) receivers, installed on the combine-harvesters themselves (Bramley,
2000).
In Spain, we have been advocating since 2004 a methodology based upon the integrated use
of remote sensors and plant-climate-soil sensors (Montesinos et al., 2006 and Álvarez et al.
2006). While the latter give us continuous data about a single plant, remote sensors provide
multispectral data about all the plants in an operation. Their continuous flow of spatial-temporal
information is recorded digitally, thus allowing it to be analysed via GIS (Geographical
Information System).
300
2. What is Teledetection?
Teledetection is the capacity to acquíre information by analysing the electromagnetic radiation
reflected or emitted from the earth's surface by an object. The human eye is a sensor that picks
up electromagnetic radiation in a person's field of vision. However, the eye cannot detect
infrared reflection - the very area of the spectrum where vegetation has maximum reflection,
determined by a crop's dominant features and its vegetative growth rate (Figures 2 and 3).
Multispectral sensors mounted on satellites and planes can detect such infrared reflection and
allow us to “see” features of the vegetation that are invisible to the human eye.
Figure 2: Physical principles of teledetection. The sun is a source of energy that emits radiation all along the
spectrum. This radiation passes through the atmosphere via so-called “atmospheric windows” and interacts with
objects on the earth's surface. Suitable sensors can pick up this radiation and convert it into a digital image.
Figure 3: Reflectance spectra of the principal elements that cover the earth's surface. The spectral response of
vegetation is more significant in the infrared region than in the visible range.
These sensors codify radiation in the form of multispectral images that can be geo-referenced
(in suitable form for cartographic presentation) and also analysed by computer programs that
301
allow us to extract the information they hold.
The images record the interaction of the vines with electromagnetic radiation in both the visible
and near-infrared areas of the spectrum. Such interactions correspond to the greenness of the
plants, their rate of photosynthetic activity and their overall biomass.
There is a direct relationship between the Vegetation Indexes revealed by multispectral images
and the Leaf Area Index (LAI) estimated in the field.
These Vegetation Indexes are used to categorise the vines into homogeneous areas according
to the lushness of their growth. It has been demonstrated that such categorisation is of great
value in maximising the production of quality wines.
Recent studies in France, undertaken by the ICV (Institut Coopératif du Vin Vallée du Rhône)
and Cemagref have demonstrated that there is a direct correlation between the readings of the
Vegetation Indexes extrapolated from an image and the quality of the grape juice.
3. Methodology
Beginning in 1999, several satellites capable of taking very high resolution images have been
launched, each one capable of obtaining information from any object on the earth's surface
larger than 60cm. These satellites are constantly orbiting the earth and are able to record
images of anywhere on its surface.
Compared to other sources of data such as orthophotos (as produced by SIGPAC and PNOA),
satellite images afford a more personal form of information gathering where the user selects
how, when and from where the satellite should record the image.
Images are taken during the ripening stage, when the foliage is fully developed, and at a critical
moment for obtaining high quality fruit.
As we have seen, the images record the interaction of the vine with electromagnetic radiation in
both the visible and near-infrared areas of the electromagnetic spectrum. Using this data and
applying arithmetical calculations that combine the spectral response of the plant in these areas
of the spectrum, one can calculate the Vegetation Indexes.
There is a direct correlation between the Vegetation Indexes obtained from a combination of
the red and near-infrared bands, and the Leaf Area index (LAI), estimated in the field (Nemani
et al. 2001).
Vegetation Indexes are used to categorise the vines into homogeneous areas according to the
lushness of their growth. It has been demonstrated that such categorisation is of great value in
maximising the production of quality wines (Johnson et al. 1996).
Such information obtained in advance of the harvest allows the administrator to plan grape
collection with greater precision and to undertake harvesting in a selective manner.
Informed by the overall vision of the area provided by a satellite and aware of the varying
degrees of grape ripeness, the farmer can implement a plan to reduce such variation. By
employing suitable agricultural practices, it is possible to reduce variation within the farm so that
the characteristics of the grapes in each plot are as similar as possible. This same process can
be applied to plots planted with the same variety of vine.
302
Figure 4: The following figures show data generated from multispectral images: i) image of a vineyard in infrared, ii)
chart of vine varieties by sector, iii) zoning of the cultivated area according to the greenness of the plants, their rate
of photosynthetic activity and vegetational mass and iv) detail of zoning.
An image enables us to obtain multispectral information about all the plants under cultivation in other words, continuous spatial information, even though this is limited to the moment
when the image is taken.
To show the vines' vegetative growth rate, it is useful to combine this continuous spatial
information with the continuous temporal information provided by the plant-climate-soil
sensors.
The plant-climate-soil sensors continuously measure the humidity of the soil at various depths,
the variation in thickness of the vines' trunks and also significant climatic variables.
Our practical experience of viticulture means that we can characterise a variety by means of a
carefully selected point of sampling (Álvarez et al. 2006), while multispectral images allow us to
pinpoint representative points of sampling.
An integrated analysis of data drawn from plant-climate-soil sensors and information from
multispectral imaging allows its extrapolation to the whole of the farm and provides a useful tool
for carrying out agricultural management practices which allow greater homogeneity of the
cultivated area (figure 5).
303
2005
2006
Figure 5: Monitoring the homogenising plan for the Merlot variety (2005-2006). In red, the position of the plantclimate-soil sensor. In the upper part, zoning based upon multispectral images taken in 2005 and 2006. In the lower
part, graphs indicating precipitation (in green), irrigation (in blue), reserves of water in the soil (in black) and plant
growth (in pink).
With this simple example, figure 5 demonstrates the potential of such an integrated diagnostic
methodology.
The two images on the upper part of the figure show a pair of maps of the zoning applied to a
plot over two consecutive years (2005 and 2006). Each colour represents the area of the plot
where the vegetation has similar characteristics. The variation in colour indicates differences in
vigour of growth and leaf cover of each zone, ranked from minor to major according to the
gradation: pale yellow, yellow, ochre, green and dark green.
Both images were taken in summer, at a time when variation in the state of the vegetation is at
its maximum, thus allowing a direct comparison. In both instances, the colour distribution
304
permits the highlighting of zones with common characteristics within the plot and also an
appreciation of the degree of difference between each.
The red rectangle indicates the position of the plant-climate-soil sensor positioned on the plot;
this collected data continuously during the 2005 and 2006 seasons.
In the lower part of the figure, the graph represents the continuous recording of four variables
during the 2005 and 2006 seasons. i) rainfall intensity (in green), ii) the degree of irrigation (in
blue), iii) the percentage of useable water in the soil (in black) and iv) the thickness of the
plant's stem (in pink).
The graph plotting the degree of humidity in the soil shows the reserves of ground water
between the boundaries (indicated by horizontal black lines) of Field Capacity and Withering
Point. The evolution of the plant is represented (in magenta), calculated from the correlation
between its growth rate and cumulative growth during the season.
In this example, it is noticeable that both the satellite image and the graph representing the
degree of stem growth indicate that in 2006 the vegetative growth was less than in 2005, a year
when, according to the winery's technician, the quality of wine made from grapes grown in this
area was at its best.
It can be seen that the accumulated growth of the plant in 2005 was greater than in 2006. This
is because during the most critical period for vegetative growth (June 2005) there were
significant rains, which increased water reserves to more than 60% and continued until the end
of June. However, in June 2006 water reserves were very low, resulting in slower leaf growth.
In the light of this information, we can conclude that if irrigation had been carried out previously,
this would have corrected the situation and improved the quality of the product.
As can be observed from figure 3, all the data obtained by this integrated diagnostic method is
automatically incorporated into a GIS (Geographical Information System), creating a data base
of “experiences”. In this way we can attempt to replicate conditions obtaining in those years
when best results are achieved in terms of quality and/or quantity.
This Geographical Information System is a computing tool that allows us to store in a rapid,
simple and user-friendly way all the data collected by the sensors (images, maps and graphics)
in what is termed a geodatabase. It also performs all the functions required to recover, visualise
and analyse the data, without our needing to be specialised technicians.
4. Conclusions
Awareness of the variation within vines can lead to improvement in a business's profitability by
allowing us to adopt preventative agricultural measures within homogeneous crop sectors.
To know the variation within their vines, 40 wine producers in Spain are these days using
diagnostic methods based upon the combined use of remote sensors and plant-climate-soil
sensors.
Such integrated diagnostic practices give a continuous spatial-temporal information flow for
every variety present in our vineyard. This knowledge will allow us to identify and define the
most productive zones for each variety and also to institute suitable agricultural practices for
rendering the remainder of the plots more homogeneous.
All data obtained by such a diagnostic method are automatically integrated into a GIS and
305
transformed into a database of “experiences” that can be replicated in subsequent years.
References
Álvarez, R.; Huete, J.; López, M.; Bea, M.; Fernández, L. and Montesinos, S., 2006. Integración
de nuevos métodos de diagnóstico en la gestión de información agrícola para mejorar la
rentabilidad y la sostenibilidad, ("Integration of new diagnostic practices in the management of
agricultural information in order to enhance profit and sustainability”). Fruticultura profesional,
no. 161: 81-87.4
Bramley, R., 2000. Measuring within vineyard variability in yield and quality attributes. Vineyard
monitoring and management beyond 2000. Wagga Wagga: 8-14
Montesinos, S. and Quintanilla, A., 2006. BACCHUS: Methodological Approach for Vineyard
Inventory and Management. European Commission. DG Research. RTD Action: Energy,
Environment and Sustainable Development. Murcia.190 pp.ISBN: 84-611-0107-3.
Montesinos, S.; Fernández, L.; Bea, M.; Martínez, A.; Erena, M.; Francisco, J.; García, P.;
González, F. and López, J.A., 2006. BACCHUS: Una herramienta para mejorar el inventario y
la gestión de la viña ("BACCHUS: A tool for improving inventory keeping and vineyard
management"). V Foro Mundial de la Viña (5th World Vine Forum). Logroño.
Pedro Beltrán Medina
Director Cátedra de la Viña y del Vino
Escuela Técnica Superior del Medio Rural y Enología.
Avda. Blasco Ibáñez, 21. 46010 Valencia.
Teléfono 963 87 71 44. Fax: 963 87 71 49
Página web: www.catedravidyvino.es
306
Area: Rural Development and Cooperation Projects
TOPIC AREA: RURAL DEVELOPMENT
AND COOPERATION PROJECTS
307
THE EUROPEAN REGIONAL POLICY, KEYS FOR THE SUCCESS:
ANALYSIS OF THE SPANISH EXPERIENCE
Regalado, J.
Colegio de Postgraduados
Díaz-Puente, J. M.
Figueroa, B.
Colegio de Postgraduados
Abstract
The regional policy of the European Union (EU) in order to reduce inequalities deal with the a
set of basic elements: the need of regional policy, the presence of an institutional model and
the participation of regions, a set of operating principles, a development objectives with
territorial orientation and financial instruments for regional development. The present paper
examines the result of this policy and its various elements in Spain. The methodology consisted
of the review of variables and findings documented in various studies, monitoring of a set of
indicators, which affect the structural funds. It concluded that the financial instruments have
been emphasized in particular on the creation of infrastructure and human capital formation;
have an impact on national convergence with the EU, but on the regional convergence has
been lower. Success is more on improvement institutional, linked to the Regional Development,
from the point view of
programme management and coordination between national and
regional plans, improved institutional evaluation, which has strengthened by public investment.
Keywords: European Union, Structural Funds, Regional Development, Convergence
1. Introduction
The evolution of the regional question in the historic construction of the European Union (EU) is
divided in four general action lines (Yagüe,2007): 1)the lacking of a explicit regional orientation
from 1950 to 1974, 2) the creation of FEDER and the foundations of a community regional
policy (1975-1985), 3) the consolidation of a community regional policy (1986-1999), which
coincides with the first reform of the Structural Funds, and 4) the definition of a regional policy
searching for a stronger economic and social cohesion for those less favored regions (20002006) and (2007-20013). The E.U. in a progressive way was creating the Structural Funds at
the same time that the economic and social cohesion, and regional integration objective was
established. These funds have as central purpose elaborating regional, agrarian, and labor
structural reforms.
These structural funds have been the European Fund of Regional Development (FEDER),
created in 1975, the European Fund of Orientation and Agrarian Guaranty –
309
Orientation Section (FEOGA-Orientation), constituted in 1962 and 1964; the European Social
Fund (FSE); and after the last reform in 1993 –a fourth fund was created, the Fishing Orienting
Financial Instrument (IFOP). The cohesion fund appeared from the signing of the Maastricht
Treaty in 1994. The centrality of the Structural Funds in the model of application, management,
and cohesion of the UE regional policy allows addressing questions related to Spain. Related
subjects to intervention areas, contribution to the regional and national convergence, and the
institutional learning derived from the principles of application and management of the
Structural Funds. Documenting these changes, is the central objective of this paper. The
research is based on diverse studies about the effects of the regional policy in Spain.
Specifically in some autonomous communities the impact of infrastructure and the building of
human capital is analyzed. These variables are treated from the perspective of a dataset of
indicators contained in the National Strategic Program of Reference 2007-2013.
2. Application and management model of the Structural Funds
2.1 Application and management principles
The application and management of the principles among the member States is based on the
results from the first reform of the Structural Funds, which produced four principles:
concentration, cooperation, programming, and additionality. In later reforms due to the EU
needs other principles were added. This days twelve principles are documented including the
first four already mentioned: A) Solidarity, B) Subsidiarity, C) Association (also called
cooperation and partnership), D) Co-financing, E) Complementarily and coherence, F)
Additionality, G) Concentration, H) Programming, I) Monitoring and evaluation, J) Efficacy, and
K) Equality and sustainable development.
3. Impact of the EU regional policy in Spain
3.1 Financial resources allocation
As a first outcome of the systematized process financial support was granted to Spain as a
consequence of a shared responsibility among the playing parts: European Commission,
member State, regional and local governments.
Financial
1986-1988
1989-1993
1994-1999
2000-2006
2007-2013
Structural Funds
------
-------
41.080
49.569
28.207
Cohesion Funds
------
------
9.574
12.322
3.250
4.822
18.707
50.654
61.890
31.457
Instruments
Total
Table 1.Resources granted to Spain from 1986 to 2013 (millions of Euros, 2004).
310
3.2 Prioritary lines of intervention
The incoming financial resources from the Structural Funds are addressed to infrastructure
improvement (transportation and communications). An important aspect is the investment in
human capital building. In the 2007-2013 period once again human capital building and
physical infrastructure investments are outstanding. Environment preservation and equality in
opportunities among others are also included. (Sánchez, 2004. (Inforegio, 2007). Addressing
financial resources from the Structural Funds in these areas generates results in the regional
and national convergence and in the institutional learning as well. These aspects are described
below as impact areas of the regional policy.
3.3 Results in infrastructure and human capital
The community regional policy through investments in infrastructure and human capital building
enables the regions to attract more investments (Drap and Herce, 1999). This allows the EU to
be more competitive in both, production and distribution of goods and services. This is the case
for the transportation infrastructure and especially to the Great Capacity Network, which
includes toll-highways, freeways, railways, and double-way roads. This network shows an
outstanding growth since 1983 from 314 Km to 25.415 Km in 2005.
30500
Longitude ( Km)
25500
20500
15500
10500
5500
500
1983
1993
2000
2005
Years
Figure1. Evolution of the Great Capacity Network in Spain 1983-2005.
In a study about the strategic role played by roads explains such growth from the Great
Capacity Network analysis (Delgado, 2003). This study emphasizes that the elements related to
this growth are: 1) the establishment of General Plans of Roads from 1984 to 1991 and the
Infrastructure Director Plan 1993-2007, 2) the Autonomous Communities participation, and 3)
the incorporation of Spain to the European Community. This confluence of efforts makes Spain
to have in 2007 163.000 Km of toll and free roads, railways, double-way and conventional
roads. From this amount 24.415 Km are State roads, 70.755 correspond to the Autonomous
Communities, and 69.476 to the provincial deputies.
It is clear the role played by the provision of transportation infrastructure for regional
development, which includes energetic and sanitation infrastructure. Since 2000-2006 the
monitoring of some indicators was established. About the final use of energy consumption in
Spain there are some variations. The use of petroleum and coal tends to decrease among the
population, 4.4% in 1998, 4.0% in 2002, and a provision of 3.6% in 2006 confirms this trend. A
similar scenario is observed in the final consumption of petroleum: 66.5% in 1997, 62% in 2002,
and 16.6% projected for 2006. These kinds of advancements are reinforced by the program of
Savings and Energy Efficiency in Spain 2004-2012. In this context it is very important the
311
electricity produced from removable sources, in which Spain shows significant advancements
being the European leader in wind-power production. During 1997, 2002, and 2006 Spain has
increased its production in Gwh/year; 26.246, 37.121 and 43.320. About sanitation
infrastructure the national government makes efforts in order to provide these physical assets
to the regions. The results show that this kind of infrastructure has been increased during the
period from 1998 to 2006. For example, the increase in number of hospitals is one indicator of
this. 1998 there were 394 hospitals, 406 in 2006, and for 2006 the number was projected in
413. This growth in the number of hospitals also increases the number of hospital per 100.000
inhabitants, and the number of physicians per 1.000 inhabitants. However, the indicator number
of beds in hospitals per 1.000 inhabitants shows a decrease for the projection in 2010, which
later is increased by 2012.
These results represent the total engineering projects funded with resources from the Structural
Funds addressed to infrastructure development in transportation, energy, and social services.
In rural development, mainly in agro-food and livestock industries, engineering projects also
have been developed. In the period 2007-2013 the European Fund of Regional Development
(FEDER) is proposing to apply financial resources for projects related to: 1) Transportation and
energy infrastructure (highways, railways, railroads, and airports) and 2) Development of
removable energy distribution and energy saving networks.
Education is an investment in human capital that renders benefits later. In this matter Spain has
made important efforts in order to improve the schooling level of the population (De la Fuente A
y Doménech R, 2006). In 1960 about 10% of the population was illiterate, and most of the
population only had primary school. By 2001 the illiteracy was dropped below to 3% mainly
among the elderly. This increase in human capital provision has been more intense during the
1980s. During this period, the proportion of people with primary education has decreased in
contrast to those with secondary and university studies, which have been increased.
3.4 Convergence at national level
The analysis of economic and social profitability of the Structural Funds shows convergence in
profits per capita and social cohesion (Cordero, 2006). This study finds evidence in the
contribution of financial resources from the Structural Funds with the implementation of the
HERMIN model. Its results show that from the 9,2 percent points reduced in the difference
between the Spanish rent and the European average rent between 1988 and 1999, 3.5 points
are due to the effect from the structural help. This is more than one third of the reduction in the
difference was due to the application of structural resources provided from the community
budget. In the same line to demonstrate the impact of the European regional development
model through the Structural Funds management and implementation in Spain, it is
demonstrated that the Structural Funds have significantly contributed to the growth of the
poorer counties. This study used an econometric model composed by three components (a
value-added function, an employment equation, and one private investment function) and
estimates that the structural policies or community cohesion have been worked very well in the
Spanish case (De la Fuente, 2003).
Another study makes a balance in the integration of Spain with the EU and is analyzed from
three variables (Jordan, 2003), the economic growth, the growth in the employment and in the
labor productivity. The results show that the real GNP is variable with periods showing
increases and others not. About the employment rate, the data show that there is an increase
compared to the 15 European states. While Spain in the period of 1985-1990 the growing rate
312
was 3.3% in the 15 states of the EU was 1.4%. In the period of 1995-2000 Spain grew 2.7% in
contrast to the EU, which grew 1.2%.
From the analysis of the regional development policy impact a positive effect in the rents
convergence is observed. This situation is confirmed in Spain with some variations among the
Spanish Autonomous Communities (Lazaro, 2004). The Spain’s GDP per capita in purchasing
power parity has been closed to the community average in the EU. This growth rate is of one
point percent per year from 1986 to 2001. The Spanish GDP per capita in 1986 was equivalent
to 70% of the community average, and in 2001 to 84%. These data indicates that there is a
convergence with the EU. However, this has not decreased internal regional unbalances in
Spain, which on the contrary have increased. This situation is called by the author as the
paradox of the divergent (to inside) convergence (to outside).
3.5 Convergence at regional level
A study was conducted to know the Structural Funds in the incidence of Spanish regional
cohesion. The justification was that previous studies in the Spanish regions considered
variables related to the regional rents to know the convergence levels but not socioeconomic
characteristics, which in this case were the core of the analysis. In order to know the incidence
factor, this study utilized the strategic factors grouping methodology that means a grouping by
cluster in order to group the Autonomous Communities by conglomerates. As a final reflection,
the study regards that inside the Autonomous Communities from 1990 to 2000 significant
territorial inequalities still persist in the milieu of the socioeconomic variables. The big number
of clusters and the fact that some territories consolidate their divergence related to the less
favored Autonomous Communities confirms this trend. This means that the territorial
distribution of the Structural Funds has not increased significantly the Autonomous
Communities cohesion level (Torres and Salas, 1998). With the same purpose of knowing how
much the Spanish regions converge, but using other variables and without mentioning the
regional policy effects (Cuadrado, 1998), other variables are considered such as general
unemployment rate and observed trend, evolution of labor productivity, and value added per
capita. The more general conclusion from this study is that the convergence in the Spanish
regions is not being achieved. This fact is demonstrated when the economic disparities
measured in Gross Value Added –productive effort made by each sector among the Spanish
regions during the period 1983-1995-, remain staged.
An extensive descriptive study of the Spanish regional policy and its impact retakes the
financial issue (Perez and Rowland, 2004). The efficiency of regional concessions is analyzed
and evaluated, not only those from the Spanish government but also those granted by the
European Community. Firstly, the study documents that the Europe’s regional assistance
received by Spain in terms of resources, was in most of the cases much bigger than that
received from the Spanish government, which is an example of the importance of the role
played by the European Community (and later the EU) in the Spanish regions development. In
the context of addressing financial resources the initial hypothesis is that the wellbeing of the
poorest regions is improved compared to the rich regions only when the regional policy is
effective. If, on the other side, the relative wellbeing of the poorest regions has not improved,
the transferences program is not effective and the regional policy has failed. Using a measure
of wealth it was tried to know how the regions responded to the donations from the Spanish
government through the Inter-territory Compensation Fund (FCI) and those from the European
Community. The results suggests that these programs have not significantly affected the
poorest regions, therefore the regional policy has not been effective.
313
However, other authors argue in their studies the incidence of the Structural Funds at regional
level. The analysis about the correcting policy of regional disequilibrium in Spain concludes that
the structural policies on economic and social cohesion during 1989-1999, have signified a
major convergence among the Spanish Autonomous Communities, from the investment in the
provision of basic infrastructure, labor qualification, and improvement in the productive
environment (Farinos, 1999). In the same topic, other study is developed using two
complementary methodologies (Ramon and Garcia, 2001). One is about conventional
calculations of convergence, and the second deals with switching models. The results from the
study highlight that the Funds influence the infrastructure and human capital levels that can
obtain the receptive Autonomous Communities from the funds. The provision of these factors
plays a positive role in the regional convergence of Spain and in the reduction of regional
inequalities. As a consequence of the convergence, the Spanish citizens’ wellbeing increases.
These results present a close relationship with the premise of the EU in terms of the strategy to
advance towards the regional convergence through infrastructure provision and human capital
building. In general, these studies deal with the impact of the regional policy in Spain from the
Structural Funds perspective, arguing with pros and cons about the role of the Funds in the
convergence. In what they agree is that such incidence has been more at country level than at
the regional domain. In a dataset from 1986 to 2006 the process of convergence in the Spanish
regions in contrast to the EU is shown.
160
140
U E = 15
120
100
GDP 1986
80
GDP 1996
60
GDP 2006
40
20
Ca
st
Ex
tr e
m
a
An d u r
d a
il la alu
cí
la
a
M
an
ch
Ca
G a
a
st
il la li ci a
y
Le
Ca ó
nt n
ab
ri
M a
ur
c
C a ia
na
r
As ia s
t
u
Va
r
l e ias
nc
ia,
Ar C .
ag
C a ón
ta
lu
R i ña
oj
a,
N a La
va
rr a
M
Pa adr
id
ís
Va
B a sc o
le
ar
es
0
Autonomous communities
Figure 2. Regional convergence in contrast to the EU=15 1996-2006.
This data show an increase in the GNP per capita in the Autonomous Communities since the
incorporation of Spain to the EU. This shows that if Spain grows, the regions grow in the same
way. This growing relationship is explained through evidence confirming the hypothesis about
the progressive closeness on such relationship found by other studies in 1995. This
demonstrates that the Spanish growth is linked to the community growth (Cuadrado, 1995).
This trend is confirmed when territorial inequalities are analyzed into the Spanish State. A study
makes a balance of the GNP per capita from 1955 to 1995. The main conclusion is that
territorial inequality, measured in terms of rent per inhabitant, decreased from1955 to 1980, and
later a stagnation has taken place or a slight divergence has been observed. In this fashion, the
final tradeoff is a process of convergence among the different regions (Delgado, 1999).
314
However, the presence of disequilibrium among the regions is still observed (Figure 2).
Minimum and maximum percentages of the GNP per capita of the Autonomous Communities in
contrast to the EU=15. In 1986 the minimum value of GNP per capita shown by the
Autonomous Communities was 45%, which corresponded to Extremadura, and the maximum
was 91% corresponding to Baleares. By 1996, Extremadura rises to 54%, and still was the
minimum, while the maximum corresponded to the Community of Madrid. In 2006 Extremadura
is again the minimum with 73% and the Pais Vasco showed the maximum. This relationship
between the minimum and maximum values reflects a difference of 46% during 1986-1996, and
52% in 2006. These data not only confirm the presence of regional disequilibrium and the lack
of convergence, they are also growing. This is the case of the Pais Vasco, Madrid, Navarra,
and Cataluña, which outweigh both, the Spanish and the EU average. Others such as
Extremadura, Andalucía, Galicia, and Castilla la Mancha increased their growth rate during the
last 20 years; however, they still remain below the community average. Due this growth rate,
these later communities are in the target for the convergence in the budget period 2007-2013.
3.6 Institutional learning
Beyond the analysis of the macroeconomic factors of the regional community policy, this policy
has had an important impact on the public administration (national, regional, and local). This
was a result of the provision of a methodology which focuses on the improvement of resource
management through the establishment and evaluation of programs. The main improvement in
the application process of the Community Funds has been made on institutions linked to
regional development. This improvement has been observed in better program management,
better coordination between national and regional plans, and better program evaluation, which
has reinforced the public investment. This analysis concludes that the financial provisions
coming from the financial instruments of the regional policy and cohesion serve as catalysis. In
addition, these resources contribute to mobilize national and regional policies and resources
and to address them more properly to the EU objectives. This situation has changed the
institutional structures in all levels to adequate them to the resources management. The
incorporation of functioning principles and lines has been helpful in addressing general policies
of the member states and those from those recently incorporated. The experience generate by
the regional policy in Spain is that the resource mobilization is not enough, as is not effective by
itself, if it is not accompanied by a set of basic principles that rule their application and
management. The institutional learning of this experience is linked to resource management
and institutional evaluation. These results are not isolated; they have their origin in the
organizational processes in Spain before its incorporation to the European Communities. This
is, its conformation in Autonomous Communities and the experience derived from the
autonomic financing. In the organizational processes, but especially in the financing system,
figured some of the application and management principles, which later appeared in the
European model of regional policy: sufficiency of the granted resources, autonomy in the
expenditures management, coordination, and solidarity. With this background and the principle
of programming applied to the elaboration of national plans, Spain joined the EU. The
institutional arrangement changed as a consequence of this merge, and it has been adjusting to
the financial resources application and management model established by the EU.
The incorporation of the management principles in the financial model of the regional policy has
promoted the modernization of the public management in the country. It has transformed both,
the management model and the understanding of public management. Therefore, one of the
most outstanding effects of the European influence is the deep cultural transformation, mainly
315
in the introduction of the idea of the citizen as a client. In this process, the evaluation plays a
fundamental role. This activity is included in the management and application principles of the
Structural Funds since 1993. In Spain there was no culture in evaluation of results.
Administrative and legal controls were the dominant methods. However, as a member State
that manage resources from the Structural Funds, Spain had to adopt community programming
norms and procedures, which require an ex-ante, intermediate, and ex-post evaluations of the
operative programs and performing frameworks. In this sense, the European Commission
contributed creating the evaluation Technical Teams, organized conferences, and published
documents to guide the evaluation practice. From the Commission the different levels of
evaluation were organized at local, regional, national, and community levels. A programming
logic was designed in order to evaluate programs. The control, follow up and evaluation were
articulated, and the different stages of the evaluation (ex-ante, intermediate, and ex-post); key
aspects of evaluation were defined and recommendations were given on the most appropriate
methods and techniques for evaluating programs financed by the Structural Funds. This
expansion of the evaluation culture in Spain has as a result the creation of the Spanish Society
of Public Policies Evaluation in 2001, and the creation of the State Agency of Public Policies
and Service Quality in 2007.
4. Conclusions
The objectives and intervention ground of the funds, territory criteria of application, main
principles of application, financial resources, responsibility of the European Commission, the
States, and other regional institutions, as well as the beneficiaries from the help, procedures for
programming, procedures for administering, following up, evaluation and financial management
constitute the model of the European regional policy.
Studies indicate that the resources coming from the Structural Funds contribute to the
convergence in Spain with the UE community average. While in the regions, even when an
economic growth is perceived, the regional inequalities still persist. The convergence showed
by the Spanish State regarding the EU, is linked to the incidence of the model to address
financial resources coming from the Structural Funds in areas of infrastructure for
transportation, telecommunications, energy, sanitation, human capital building, among other
areas of expenses.
The conclusions given by the studies indicate that the success factors of the model are more
notable in the development of institutions linked to the regional development, improvements
that go from program management, coordination between national and regional plans, a better
institutional evaluation that has reinforced the public investment.
References
Cuadrado R. J. R., 1999. Divergencia regional versus convergencia de las disparidades
regionales en España. EURE (Santiago), set, Vol. 24. No 72, pp. 5-31
Cuadrado R J. R., 1998. Convergencia regional en España: Hechos, tendencias y
perspectivas. Vol. 8: Colección economía española.
Cordero M. G., 2005 La rentabilidad económica y social de los Fondos Estructurales:
experiencias y perspectivas. Presupuesto y gasto publico 39,pp. 151-172.
316
De la Fuente A., 2003 El impacto de los Fondos Estructurales: Convergencia y Cohesión
interna. Revista de Economía Publica, Vol. 165, pp. 122-129
De la Fuente A y Doménech R., 2006. Capital humano, crecimiento y desigualdad en las
regiones españolas. Instituto de Análisis Económico (CSIC), pp. 47
Delgado R. MA de J. y Álvarez, A. I., 2003. Efectos de la red viaria de gran capacidad sobre el
desarrollo territorial de España. Economía Industrial No 353/ V, pp. 25-32
Delgado C. M y Sánchez F .J., 1999. Las desigualdades territoriales en el Estado español
1955-1995. EURE (Santiago), set, vol. 25 No 75, pp.41-62
Drap M. y Herce A., 1999 Infraestructura y crecimiento: un panorama, Revista de Economía
Aplicada 6:, pp. 129-170
Inforegio., 2007. Marco Estratégico Nacional de Referencia 2007-2013 de España.pp. 227
Jordán G. M., 2002 La ampliación de la Unión Europea hacia el Este: repercusiones para
España. Boletín Económico de ICE No 2733, pp. 9-17
Farinos D. J., 1999. La política correctora de desequilibrios regionales en España: efectos de la
integración comunitaria. En cuadernos de geografía de la Universidad de Valencia, Num.
65/66, pp. 193-211
Jordán G. M., 2002. La ampliación de la Unión Europea hacia el Este: repercusiones para
España. Boletín Económico de ICE No 2733, pp. 9-17
Lázaro A. L., 2004 Integración económica y cohesión: una visión española de la experiencia
europea. Seminario sobre integración económica y social: lecciones aprendidas y
perspectivas., pp. 85-98
Perez G. y Rowland V. P., 2004 Políticas económicas regionales: cuatro estudios de caso.
Documentos de trabajo sobre economía regional No 47, pp. 58,
Sánchez G. L., 2004 Integración económica y regionalismo: la Unión Europea. Presentada en
la sexta reunión de economía mundial en Badajoz 2004. Facultad de Estudios Empresariales y
Turismo. Universidad de Extremadura. pp. 13
Serrano C. L., 2006. Los reglamentos de los Fondos Estructurales 2007-2013 y sus
implicaciones para España. Boletín Económico de ICE No 2897. From November 27th to
December 3rd, pp. 3-16
Torres S. T y Sala R. M., 2003. Incidencia de los Fondos Estructurales en la Cohesión
Regional Española, pp. 25 available in www.ub.es/ecopusBCN/ponencias
Perez G. y Rowland V. P., 2004 Políticas económicas regionales: cuatro estudios de caso.
Documentos de trabajo sobre economía regional No 47, pp. 58,
Yague B. J. L., 2007 Modelo de Regionalización para el Desarrollo Local: Aplicación a la
Republica Oriental del Uruguay. Doctoral Dissertaion. Departamento de Planificación de
Proyectos de Desarrollo Rural. Universidad Politécnica de Madrid
317
Correspondence (for further information, please contact)
José Regalado López
Colegio de Posgraduados
Km. 36.5 Carretera Mexico-Texcoco Montecillo Edo. de México. C.P. 56230
Phone: (595) 95 2 02 00 Texcoco / (55) 58 04 59 00 D.F
318
TAKING INTO ACCOUNT FOOD AND NUTRITION SECURITY IN THE
MEASUREMENT AND ASSESSMENT OF HUMAN DEVELOPMENT
Afonso Gallegos, A.
Trueba Jainaga, I.
Abstract
Human Development Index (HDI) has become a very useful tool for measuring development
and making inter-country comparisons. The observation of its trajectory and evolution by
country arouses the curiosity of students of development and invites them to question the
underlying causes of the behaviour of the index. In order to provide answers to such questions
the Human Development Report complements the HDI by offering varied, valuable and relevant
statistical information related with every dimensions of human development. From its very
beginning, experts have debated about the strengths and weaknesses of the report and have
acknowledged a need to include new indicators in the measurement of human development. It
is in this context that Mahbub ul Haq, the creator of the Human Development Reports said:
“Here we have a broad framework; if you want something to be included in this list, which may
deserve a table in the Human Development Report tell us what, and explain why it must figure
in this accounting. We will listen”. As a response to this demand, this paper proposes the
inclusion of food security in the measurement of human development, as well-being can only be
achieved when there is not food insecurity.
Keywords: Food Security, Human Development, Indicators
1. Human Development: Concept, Measurement and Analysis
It is opportune to start by establishing the general foundations of the matter of discussion. This
paper deals about the measurement of human development. Hence, defining such a concept
and studying and analysing how it is measured is the starting point to which this first section is
devoted.
The concept of human development was formulated by the United Nations Development
Program in 1990 and can be summarized as follows: Human development is the process of
enlarging people’s choices. These choices are infinitive and can change over the time but some
of them are changeless and common to every human being: adequate nutrition, access to safe
water, better health services and access to knowledge. The primordial goal is to satisfy the
most basic of human needs. A country’s level of development may be considered as the degree
to which basic human needs are met.
1.1. GDP as an indicator of human development
Gross Domestic Product (GDP) and GDP per capita have been used for a long time to measure
and compare development, on the grounds that development can be identified with economic
growth since welfare arises from economic growth.
319
Economic growth is important for the achievement of development as it creates the “means” to
reduce deprivation. However, if growth is not properly managed, its advantages can be
squandered. There is no an automatic and direct relationship between economic growth and
human development (Sen, 1999).
It is more meaningful to measure development as a reduction in the extent of deprivation
amongst the most needy rather than as a process of enrichment that tends to benefit most
those who are already rich. Moreover, poverty is not equivalent to lack of income but rather to a
shortage of entitlements. Thus, income growth is not an end but a means by which to achieve
development. But it is not right to equate development with economic growth because income
is significant as a means of obtaining entitlements which are intrinsically important but it is not
the only instrument through which opportunities and entitlements can be generated. Besides,
the relationship between income and entitlements varies between different communities and
even between families and individuals (Sen, 1999).
1.2. Human Development Index and Human Development Reports
In the nineties in international debates on development policies it was acknowledged that
development is more than merely material growth and that it should have broader objectives
related to human wellbeing (Fukuda-Parr, 2002). Development is, therefore, no longer simply a
matter of economics and the concept of Human Development emerges.
For a long time the perennial question of those concerned with development has been: How
much is a country producing? From the nineties the leading question has been: How are
people going on? The reason for this change is an increasing acknowledgement that the real
objective of development is to enlarge people’s choices (Ul Haq, 1995a).
It is within this context that the concept of the Human Development Index (HDI) emerged. HDI
is an index used to measure and compare development between countries since 1990 and it is
calculated as the mean of three components: GDP per capita, life expectancy at birth (LE) and
educational level (EL) – measured as adult literacy rate and combined gross enrolment ratio for
primary, secondary and tertiary school.
The three variables included in the HDI are inter-related. When GDP per capita increases, LE
and EL also tend to increase. However, HDI establishes a ranking of countries that differs
significantly from the ranking obtained when considering only GDP per capita. For example, the
position occupied by Equatorial Guinea is one hundred places below its position when only
GDP is considered. Similarly, Botswana and South Africa occupy a lower place (about 70
places below) their positions when only GDP is taken into account. On the contrary, other
countries as, for example, Sweden, Cuba or Uruguay have higher quality health and social
services than could be expected considering their GDP per capita.
There are countries which have achieved an increase in LE and EL as a result of economic
growth, for example, Algeria, Iran or Mexico. Others have achieved economic growth out
through improvement of social services as for example post-reform China, Sri Lanka and
Indonesia. Amartya Sen (1999) distinguished between two types of successful development
which he termed respectively Growth-Mediated and Support-Led processes.
HDI has become a very useful tool for measuring development and making inter-country
comparisons. The observation of its trajectory and evolution by country arouses the curiosity of
students of development and invites them to question the underlying causes of the behaviour of
the index. In order to provide answers to such questions the Human Development Reports
(HDR) published by the UNDP complements the HDI by offering two other types of information:
statistical information and analytical information.
Varied, valuable and relevant statistical information related with every dimensions of human
development is provided by the HDR. This statistical information is grouped in tables including
320
indicators with the aim of putting forward a global assessment of the achievements of countries
with regard to different fields of development. Main tables are organized by topics. Indicators
which are considered to warrant inclusion in the HDR change from year to year. For example,
recent reports include statistics about cell telephone users or women who play a role in politics
that were not included in the early reports.
Statistical data provide a framework for the analysis of global human development. Every year,
the HDR analyses an important issue with the aim of stimulating debate which might eventually
lead to the implementation of new policies with repercussions on human development.
1.3. HDR encourages innovative use of statistics
The most significant contribution of the HDR has been the Human Development Index (HDI)
that has always aroused and continues to arouse great expectations. The success of the index
rests on the fact that it provides the basis for a ranking of countries from the most to the least
developed as well as a classification of countries between high, medium and low developed
(Streeten, 2000). For this reason, from its very beginning, experts have debated about its
strengths and weaknesses and have acknowledged a need to include new indicators in the
formulation of the HDI.
It is in this context that Mahbub ul Haq, the creator of HDR, made the point that: “Here we have
a broad framework; if you want something to be included in this list, which may deserve a table
in the Human Development Report (and may even be considered for inclusion in the Human
Development Index) tell us what, and explain why it must figure in this accounting. We will
listen”4
The HDI does not include any indicator of the extent to which the most basic need of all, that of
food security is met. This has been acknowledged as an important gap: “How can we
incorporate food security in the HDI? We do know how to make shoes of leather and energy out
of coal or water power. But we know very little about how precisely to transform social services,
adequate food and certain institutional arrangements into long, healthy, productive, creative,
enjoyable lives” (Streeten, 2000).
At present is widely acknowledged that the expansion of education and health constitute
important targets of development that have to be considered in assessing poverty together with
income (Kanbur,2001). There is now a tendency to identify the concept of human development
with its measure (Fukuda-Parr, 2002). However, it would be a great mistake to concentrate too
much on the HDI, or any other such aggregative index. The real merit of the Human
Development Report lies in the attention it brings to bear on so many aspects of development.
The openness of the creators of the Report to suggestions to extend the ways in human
development is evaluated is also important (Sen, 2000).
2. Why to include Food and Nutrition Security Dimension in the measurement
and assessment of human development?
Arguments are presented bellow to defend the idea that, being both an end and a means of
development, food and nutrition security should be taken into account in the measurement of
human development, as well-being can only be achieved when there is not food insecurity
(Afonso and Trueba, 2005).
4
Quotation from the article by Amartya Sen, (2000), entitled “A Decade of Human Development”,
published in the first edition of the Journal of Human Development.
321
Good nutrition it is an end of development as it is a basic human need: a chronically hungry
person will never reach a satisfactory state of well-being. A hungry person is deprived of a
basic need and of the most fundamental of human rights (Streeten, 1982). It is also a means of
development as it conditions every aspect of development: malnutrition is reflected in high rates
of disease and mortality (Grantham-McGregor et.al., 1999), (Caulfield, et.al. 2004a) and
(Caulfield, et.al. 2004b), limited neurological development (Scrimshaw, 1998) and low
productivity among current and future generations (Arcand, 2001). It is a major constraint to a
country’s ability to develop economically, socially, and politically (Who, 1997). Observation and
analysis of statistical data provides different examples of how an improvement in development
can be achieved by improving nutrition. Support-Led Development can be promoted by
investing in food security.
The measure of adequate nutrition happens to be a good indicator of lack of poverty: health,
shelter, educational level and health level are reflected in the nutritional status (Lipton, 1983)
and so that nutritional situation of a country is internationally acknowledged as an indicator of
development (OMNI, 1998). In order to achieve a sustainable improvement of the situation of a
country food needs should be known and considered in any strategy of development (FAO,
2004).
Hunger and malnutrition are internationally acknowledged as a main problem to be solved by
humanity. When the Millennium Development Goals where established (Millennium Project,
2000) the first goal was to eradicate extreme poverty and hunger with the target of reducing by
half the proportion of people who suffer from hunger by 2015. It is fair to acknowledge that
International Community pays more attention to malnutrition problem and identifies it as a main
concern that should be implicit in any human development program. There is an increasing
awareness of the importance of achieving world food and nutrition security for development.
Yet, hunger and malnutrition remains being a problem largely silenced. A datum that
corroborates it is the fact that malnutrition problem is overlooked both in the Human
Development and in the World Health Organization Reports.
Malnutrition has never been considered sufficiently relevant for human development as to be
the analysis topic of any of the 18 reports published till date (UNDP, 1990-2008). Different
topics that rise world concern have been addressed, as for example, democracy, cultural
diversity, international cooperation, world water crisis or climatic change. But world food
security has never been the topic of discussion even after 2000 when the Millennium
Development Goals fix the world agenda for human development.
Hunger and malnutrition are also overlooked when evaluating world health. World Health
Organization (WHO) publishes every year data concerned with health and causes of mortality.
Although WHO acknowledges that malnutrition is a main source of many illnesses that afflict
humanity it is not included as an explicit cause of mortality in the statistics.
The inclusion of food and nutrition security in the measurement and assessment of human
development could be a tool of fighting against hunger and malnutrition. And seeks to meet
both: the general demand of new dimensions to measure the complex multidimensional
concept of human development and the specific demand to take into consideration the
nutritional situation of countries. It seems that “the inability to recognise malnutrition as a
predominantly social problem is widespread in the medical profession. This inability causes the
problem being addressed in an inappropriate and, sometimes, callous manners” (Ramani et.
al., 2006).
3. Why is it necessary the elaboration of a new indicator of food and nutrition
security?
The inclusion of food security dimension in the measurement and evaluation of human
development requires the elaboration of a new indicator. To measure food and nutrition
322
insecurity, different indicators have been proposed but they are inadequate to asses the
nutritional situation of a country at per capita level for different reasons.
Some of them only contemplate one dimension of nutrition. This is the case of FAO’s indicator
which is an estimation of the proportion of people having access to fewer kilocalories than the
minimum daily requirement for a healthy life (Mernies, 2003) and only measures energy
deficiency (food quantity) neglecting micronutrients deficiency (food quality). Different indicators
measure deficiency of particular micronutrients but do not include other dimensions of nutrition.
There are other indicators that are indirect indicators and thus, the relation between them and
the real nutritional situation is indirect and difficult to determine. This is the case of the
proportion of children under five who are underweight, that is an indicator of the nutritional
status of children an is also an indicator of the nutritional status population as a whole (Onis,
and Blössner, 2003). This indicator can reflect other dimensions of nutrition besides quantity of
energy (Smith and Haddad, 2001) but only gives a partial approach of the situation.
Underweight may be an outcome of other causes different from malnutrition (Osmani, 1992)
and some undernourished children are not underweight (Van den Broeck et. al. 1994). Besides,
the nutritional status of children is not always extrapolating to the whole population (FAO,
2001).
Being aware of these deficiencies, the appropriateness of creating new systems to evaluate the
nutritional situation considering all relevant factors has been emphasized: malnutrition is much
more widespread than energy deficiency, and likely larger than deficiencies identified through
anthropometric measures (Mason, et. al. 2001).
4. A new Indicator of Food and Nutrition Security (ISAN by its Spanish Acronym)
A new Indicator of Food and Nutrition Security (ISAN) is proposed. The design and formulation
of ISAN has been developed in a PhD thesis and was presented and discussed in a Seminar
held in FAO in July, 5, 2007. The ISAN has been designed as a mean of three sub-indexes that
respectively measure three dimensions of food and nutrition security – quantity, quality and
biological utilization) as shown in the figure bellow (Figure 1).
1. DES per capita
Quantity
Energy
Dimension
2. % of carbohydrates,
I1 Energy
Dimension
Nutritional Status
proteins and fat
3. nº of food groups
Quality
Diversity
Dimension
providing at least 90 % of
energy
I2 Diversity
dimension
ISAN
4. % contribution of the
main food group
5. % of population with
access to safe water
Biological
utilization
Sanitary
Dimension
6. % of population with
I3 Sanitary
Dimension
access to sanitation
DIMENSIONS
VARIABLES
SUBINDEX
Figure 1. Food and Nutrition Security Index (ISAN). Towards a multidimensional Index.
323
Access to enough food energy in adequate proportions of macronutrients is measured through
I1 sub-index that is developed from the combination of two variables: Dietary Energy Supply
(DES) per capita and the proportion of DES of carbohydrates, proteins and fat.
Access to needed food is determined through I2 sub-index that measures diversity dimension
combining two variables: number of food groups providing at least 90 per cent of energy and
share of the contribution of the main group.
Variables used to evaluate the sanitary dimension obtaining I3 sub index are proportion of
population with access to safe water and proportion of population with access to sanitation.
The relevance of ISAN lies in the fact that it brings together considerations of dietary energy
supply and the quality of nutrition. It also breaks new ground by bringing in a water/sanitation
dimension.
The validity of ISAN rests on the selection of the variables involved and the criteria used to
combine them into a single index, which requires scientific knowledge of the factors
conditioning nutritional status. ISAN fulfils the conditions of good indicators, in particular: it
proposes targets to reach and measures the degree of attainment of such targets; it is sensitive
to changes and changes happened in any variable are measurable in the short term and are
reflected in the index; it has temporal and spatial specificity as there are data of the variables
that conform the index for a large number of countries; it is efficient in terms of cost benefit as
can be readily applied at low cost because its variables require the use of data that is already
available on FAO and UNDP websites. In addition, the estimate of malnutrition in the world
using ISAN is very consistent with previous estimates.
5. ISAN as an indicator of Human Development
The ISAN seeks to respond to the general quest for new indicators which throw greater light on
the complex and multidimensional concept of development, and on the particular issue of its
nutritional dimensions. It also responds to a FAO’s quest for better indicators of food and
nutrition security in order to enlarge the range of measurement including the trends of
micronutrients, besides the usual points linked to food energy (Mason, 2002).
This paper does not propose that ISAN should replace the publication of the other food security
indicators, but argues that it should be treated as a complement to the information already
being published.
It also argues that ISAN should be presented in a table in the Human Development Report in
order to complement the statistical information with which to assess human development
achievements at country level. The inclusion of ISAN as a forth dimension of HDI could also be
considered.
5.1. Presenting ISAN and its Components in a Single Table
The usefulness and efficacy of an indicator when utilizing it for assessing and evaluating
progress or and in the process of policy decision-making process require that those who are
interested know how to interpret the data and understand them correctly. If the process of
obtaining the data is not clearly understood the implications of the indicator could be
misinterpreted (UNDP, 2006).
The way in which the indicator is presented is important to a better understanding of its
message. Since ISAN is an index which combines three dimensions, obtained as a mean of
three sub-indexes, it is advisable that the table in which it is normally presented should also
include the three sub-indexes. When ISAN is accompanied by the three sub-indexes from
324
which it is derived, an observer interested in the nutritional situation of a country can check,
through examining the three sub-indexes, which is the main source of weakness and the
reason why the ISAN has not reached a higher value. In order to facilitate the interpretation of
the data, a short description of what each sub-index measures can be included at the head of
each column, as illustrated below (Figure 2).
Energy
Dimension
Country
..
..
..
..
Diversity
Dimension
(Sufficient food
and balance of
macronutirients)
(Needed Foods
and
micronutrients)
I1
I2
..
..
..
..
..
..
..
..
Sanitary
Dimension
(Environmental
Safety)
ISAN
1/4 (I1+I2+2I3 )
I3
..
..
..
..
..
..
..
..
Figure 2: A Statistical Table for the ISAN.
5.2. Incorporating the ISAN within the HDI: HDI-2
When the Human Development Index was designed as a mean of three dimensions (life
expectancy, educational level, and income per capita) the convenience of including other
variables that reflect other development dimensions was considered. Among those variables,
the inclusion of DES (kilocalories per capita) was considered but it was not included in the end
as it was concluded that it was not a good indicator of malnutrition because malnutrition is
conditioned by many other factors (Ul Haq, 1995b).
The ISAN reflects these different factors and has been designed as an index such that its
incorporation in the HDI as a new dimension should be feasible. Although there is a high
correlation between the three-dimensional HDI and ISAN, the ranking obtained by applying this
new tetra-dimensional HDI, which has been called HDI-2, differs substantially from the ranking
obtained with the three-dimensional HDI. The new ranking differs from the former ranking in
such a way that positions of countries fall by as much as 22 places or rise by up to 20 places.
6. Conclusion
Undernutrition denotes a flaw in the development process where it has failed to reach certain
groups of the population. In order to achieve a sustainable improvement in a country’s situation,
food needs must be identified and considered as part of any strategy of development (FAO,
2004). Therefore, no appropriate method to measure development can afford overlooking the
nutritional dimension: food security should be given explicit attention in the assessment and
measurement of human development. In this paper, the incorporation of a Food and Nutrition
Security Index that enables to establish a ranking of countries in per capita terms, in official
databases related to human development has been proposed. This might lead countries to
increase their level of attention (and resources) to the reduction of hunger and malnutrition.
325
References
Afonso, A. and Trueba, I., 2005. Incidencia de la seguridad alimentaria en el desarrollo. In
Trueba, I. (ED), El fin del hambre en el 2025: Un desafío para nuestra generación, pp. 165-182.
Mundiprensa, Madrid. Available in http://www.fao.org/tc/tca/esp/pdf/findelhambre/index_es.htm
Arcand, J.L., 2001. Undernourishment and Economic Growth: The Efficiency Cost of Hunger.
FAO Economic and Social Development Paper.
Caulfield, L. E., Richard S. A and Black. R. E. 2004a. Undernutrition as an Underlying Cause of
Malaria Morbidity and Mortality in Children. American Journal of Tropical Medicine Hygiene 71:
Suppl. 2 S55 – 63
Caulfield, L.; Onis, M.; Blössner M. and Black, R. 2004b. Undernutrition as an underlying cause
of child deaths associated with diarrhea, pneumonia, malaria, and measles. American Journal
of Clinical Nutrition, Vol. 80, No. 1, 193-198.
FAO, 2001. The World Food Summit Goal And The Millennium Development Goals Committee
on World Food Security. Twenty-seventh Session. Rome, 28 May - 1 June 2001.
FAO, 2004. Incorporating Nutrition Considerations into Development Policies and programmes.
Brief for Policy-makers and programme planners in developing countries”.
Fukuda-Parr, Sakiko. 2002. Rescuing the Human Development Concept from the HDI:
Reflections on a New Agend in Sakiko Fukuda-Parr and A. K. Shiva Kumar (eds.). Readings in
Human Development: Concepts, Measures and Policies for a Development Paradigm. New
Delhi:Oxford University Press.
Grantham-McGregor, S.M.; Fernald, L.C.; Sethuraman, K. 1999. Effects of health and nutrition
on cognitive and behavioural development in children in the first three years of life. Part 2:
Infections and micronutrient deficiencies: iodine, iron and zinc” Food and Nutrition Bulletin. 20:
76-99.
Kanbur, R. 2001. Economic Policy, Distribution and Poverty: The nature of Disagreements.
World Development. Volume 29, Issue 6. Pages 1038-1094.
Lipton, M. 1983. Poverty, Undernutrition and Hunger. World Bank Staff Working Paper nº 597.
Mason, J.B., 2002. Measurement of hunger and malnutrition. Keynote paper in the International
Scientific Symposium on Measurement and Assessment of Food Deprivation and
Undernutrition. Rome.
Mason, J.B., Hunt, J., Parker, D. & Jonsson, U., 2001. Improving child nutrition in Asia. Food
Nutri. Bull., 22 (Suppl. 3).
Mernies, J., 2003. Measurement of food deprivation. FAO Statistics Division.
Millennium Project, 2000. Millennium Development Goals. Targets and indicators.
OMNI, 1998. The Final Report. The OMNI Experience: Using global lessons to move local
programs. Washington D.C.
Onis, M. and Blössner, M., 2003. The World Health Organization Global Database on child
growth and malnutrition: methodology and applications. International Journal of Epidemiology,
pp. 518-526
Osmani, S. R. 1992. On Some Controversies in the Measurement of Undernutrition In: S. R.
Osmani, ed. Nutrition and Poverty. Oxford: Clarendon Press, pp. 121-164.
326
Ramani, A. V.; D’Souza, Ravi 2006. Misunderstanding malnutrition. Indian Journal of Medical
Ethics. Vol III No 3
Scrimshaw, N., 1998. Malnutrition, brain development, learning and behaviour Nutrition
Research. (18) 351-379.
Sen, A., 1999. Development as Freedom. Oxford University Press.
Sen, A., 2000. A decade of Human Development. Journal of Human Development. (1) 17-23
Smith, L. C., and Haddad L. 2001. How important is improving food availability for reducing
child malnutrition in developing countries?. Agricultural Economics 26 (3) , 191–204
Streeten, P., 2000. Looking Ahead: Areas of future research in human development. Journal of
Human Development. (1 ) 25-48
Streeten, P., 1982. First things first: meeting basic human needs in the developing countries
Oxford Universities Press.
Ul Haq, M. 1995a. The paradigm of Human Development. In: Reflections of Human
Development. Chapter 2 and 3. Oxford University Press.
Ul Haq, M. 1995b. The Birth of Human Development. In: Reflections of Human Development.
Chapter 2 and 3. Oxford University Press.
UNDP (United Nations Development Programme). 2006. Human Development Report. New
York: Oxford University Press.
UNDP, 1990-2008. Human Development Reports. New York: Oxford University Press.
Van den Broeck, J., Meulemans, W. & Eeckels, R. 1994. Nutritional assessment: the problem
of clinical-anthropometrical mismatch. Eur. J. Clin. Nutr. 48: 60-65.
WHO (World, Health Organization), 1997. Health and Environment in Sustainable
Development. Five years after the eight submit.
Ana Afonso Gallegos
Universidad Politécnica de Madrid.
GESPLAN (Grupo I+D Planificación y Gestión del Desarrollo Rural-Local).
Dpto. Proyectos Y Planificación Rural
ETSI Agrónomos. Avda. Complutense, s/n. 28040 Madrid.
Phone: +34 91 336 39 86
327
SELECTING AND ASSESSING FAIR TRADE SUPPLIERS USING
MULTICRITERIA TECHNIQUES
Cobo Benita, J.R.;
Ortíz Marcos, I ;
Mataix Aldeanueva, C.
Abstract
Fair Trade has consolidated itself during recent decades as a useful instrument in co-operation
for development, especially for small producers in the South. However, there is some concern
that a situation of dependency between the producer organisations and the organisations that
market their products in the developed countries may materialise. To avoid this situation of
dependency, the producer organisations must progress and develop in order to become strong,
sustainable entities even in traditional competitive markets, out of the protective reach of Fair
Trade networks. This paper presents the development of a holistic methodology, based on the
application of multicritera techniques, and is focused on supporting the selection and
assessment process of Fair Trade Suppliers so as to understand what organisations are ready
to enter the Fair Trade Networks and which ones are prepared to do business in competitive
traditional markets.
Keywords: co-operation for development, fair trade, multicriteria analysis
1. Introduction
A weak, maybe small producer organisation lacking in many things decides to become part of
the Fair Trade circuits to find a more favourable market that will allow it easier access to the
developed markets. For a time, the producer organisation changes, improves, becomes
strengthened and a few years later is in a position to manage on its own outside the protection
afforded by the Fair Trade market. It is now time to launch out into the traditional market and
relinquish its position to another organisation in a less developed stage.
In respect of supplier management, standard ISO 9001 states: “The organisation must assess
and select suppliers in accordance with their capability to supply products that meet the
organisation’s requirements. Selection, assessment and re-assessment criteria must be
established”. It is precisely these criteria that give rise to a certain complexity in the process,
since in most cases their nature is eminently subjective (Ballou, 1999), hence there are marked
differences when assessing a supplier as this depends on who is making the assessment.
As supplier management is of the utmost importance, it is imperative to eliminate subjectivity
(Herrera y Osorio, 2006). This requires more specific tools to bring tranquillity to the decisionmaking process, regarding both the outcome and the intervening process.
In the current organisational environment, the multicriteria paradigm has appeared as an
effective aid to organisational decision-making and management (Romero, 1993), offering a set
of techniques and methods capable of taking account of the decision-making centre’s
328
preferences and which assist decision-making in any area of scientific research or human life.
Supplier selection is not outside the scope of application of these techniques and methods.
This paper presents the development of a methodology based on the application of methods by
experts and on multicriteria techniques focused towards the process of selecting Fair Trade
suppliers. The paper is organised as follows: the first part explains what Fair Trade is and the
need to apply multicriteria decision-making techniques to supplier assessment; further on, the
model put forward for solving the decision-making problem is set out, and finally the
conclusions about the model and its implementation.
2. Objectives and scope
The object of this paper is to find a validated hierarchical structure for the application of the
Analytic Hierarchy multicriteria assessment method (Saaty, 1994) to the assessment and
selection of Fair Trade organisations working with small producer groups in developing
countries. The tool produced will allow knowing:
•
which organisations are prepared to join the Fair Trade network,
•
which are prepared to be sustainable in the traditional market.
This paper sets out a set of criteria suited to the design of a model to assess and grade Fair
Trade suppliers that will allow identifying progress and the detection of best practices that could
then be propagated throughout the studied network or other Fair Trade networks.
Extending the model to the realm of Fair Trade in both the area of supply management and the
construction and application of multicriteria decision-making methodologies suited to the
assessment of Fair Trade suppliers is deemed to be of great interest, not only for its practical
use – already confirmed by other sectors – but also for its innovative nature that makes feasible
its dissemination among the scientific community.
3. Fair Trade. Criteria, current situation and a description of the problem
Fair Trade is built on a basis of equality and transparency in labour relations that will enable the
living conditions of producers in countries in the South to be improved and guarantee the
consumers in the North that the products they purchase have been produced in decent
conditions. The activity of producers is always sustainable in their economic, environmental and
social context.
In order to reach these goals, the products are purchased as directly as possible from the
smallholders and craftsmen; these are offered better remuneration for their toil and the prices of
their products are fixed in agreement with the producers. The producers in the South commit
themselves to working and making decisions democratically at the core of their organisations,
thereby creating more participatory structures. Fair Trade is radically opposed to the
exploitation of child labour and encourages women to take part in decision-making on an equal
basis, the starting point always being that women and men should receive the same wages.
For their part, the Fair Trade organisations in the North commit themselves to giving products
from the South direct access to the markets in the North and paying a fair price that will let the
producers cover their basic needs and production costs while leaving a margin for investment.
The organisations in the North also commit themselves to paying part of the price of the
production in advance to avoid indebtedness. The trade relations agreed are always medium to
long term, which gives the organisations in the South sufficient time to develop and become
sustainable.
329
The importers offer technical back-up for producing and distributing the product, such as
information and assessment concerning European trends and fashions, health and safety
regulations concerning the product, or the end packaging required to make it easier for the
producer to gain access to the international markets. If needed, help is also provided to develop
new products with finance being made available through credits and technical and
administrative training. The ultimate goal is to give producers in the South an opportunity to
become self-sufficient.
At the same time, Fair Trade organisations organise awareness-raising and political impact
campaigns aimed at changing the present unfair international trade frameworks. Work is also
undertaken to educate western society about responsible consumption and make it aware of
the culture, identity and living conditions of the producers.
4. Model for assessing organisations and co-operatives and integrating them into
the Fair Trade network
4.1.
General framework and development of the procedure
There are 44 craftwork organisations in the Fair Trade network studied. Up to the present, the
supplier relationship with these suppliers has been dependent on the organisation’s level of
development; however, performance in quality issues, delivery dates, and the product’s
success in the market has not always been satisfactory, with a considerable effect on sales
figures.
Supplier selection.
Defining and characterising
quality.
Reduce, maintain or
increase orders.
New opportunities.
Terminate contract.
Supplier evolution.
Improvement or a step
back.
Best supplier.
Registered supplier.
Quality characteristics.
Management assessment.
EFQM.
New product development.
Follow up to action plans
deriving from the
assessments.
Figure 1. Continuous improvement cycle for the assurance and development of Fair Trade suppliers.
Stages and expected results in each of the stages.
The solution put forward classifies organisations according to their development as an
organisation, their product market share and the quality of these products, with the purpose of
learning what measures to take and what budgetary amount to allocate to each organisation in
line with the criteria that are most relevant for the companies purchasing products from these
small organisations and co-operatives.
The procedure employed to approach the problem is set out in the figure below followed by an
explanation of the results.
330
Defining the criteria for assessment
1º. Select experts.
2º. Select criteria for assessment.
3º. Design the grading scale.
Determination of the relevant
importance between criteria
1º. Objective weighting.
2º. Subjective weighting.
3º. Final weighting.
Supplier assessment
1º. Design
matrix.
the
supplier-criteria
2º. Levelling down.
Grading the suppliers
3º. Standardisation.
Overall performance assessment of
critical suppliers
Figure 2. Procedure for supplier assessment and selection.
4.2. Defining the criteria for assessment
The organisations purchasing Fair Trade products draw up an annual purchasing budget where
it is decided how much is to be purchased from each South organisation. This decision is
mainly based on commercial criteria (the producer’s capacity for growth, development of new
products, and the market potential of its products) and on the development criteria for each of
the organisations worked with. It is attempted to strike a balance between both points.
When an organisation is assessed according to development criteria it is attempted to quantify
the group’s key development factors. To do so, the Human Development Index (HDI) is borne
in mind together with the Human Poverty Index to which the group or organisation belongs, the
producer group’s gender composition and the organisation’s level of sustainability. In this way,
purchases are prioritised from groups in countries with a low level of growth that comprise
women and small groups, with little experience, with not very competitive product ranges and
with hardly any sales diversification.
Another of the great challenges to be overcome consists in improving producers’ compliance
with product delivery dates. For this reason, it has been proposed to include quality criteria in
the model. These criteria not only ensure that products arrive on time but that they also meet
the specifications in the contract.
The model, therefore, is designed with these criteria in mind:
331
Making FT’s organisations sustainable in
the traditional market
Main objective
To select and assess fair Trade
suppliers
Criteria
Development
Sub-criteria
Market
Quality
Ability to
adapt
HDI
Gender
composition
Specifications
Delivery
date
New product
development
Sustainability
Product
potential
Alternatives
Productor 1
Productor 2
Productor 3
…
Productor n
Figure.3 Hierarchical structure of the problem.
4.2.1. Criteria for Development
•
Sub-criterion: ‘HDI’. The Human Development Index is a per country measurement
drawn up by the United Nation’s Programme for Development (UNPD). It is based on a
statistical social indicator comprising three parameters: a long healthy life (measured
according to life expectancy at birth), education (measured according to the adult
literary index and the combined gross rate of enrolment in primary, secondary and
further education), and decent standard of living (measured by the per capita GDP in
USD).
•
Sub-criterion: ‘Gender composition’. The number of women with a remunerated job
unrelated to agriculture has gradually continued to grow. The greatest increases have
been recorded in the regions where women were less present in the labour market (in
central Asia, western Asia and Oceania). The decision-making centre sets five levels for
this criterion depending on the proportion of women making up the producer group, as
depicted in the following table.
332
Level
Very high
High
Medium
Low
Very low
% women in
the group
95-100%
75-94%
55-74%
35-54%
0-34%
aij*
1
0.85
0.65
0.45
0
Table 1. Experts’ ratings for the sub-criterion: ‘gender composition’.
•
Sub-criterion: ‘Producer group level of sustainability. In order to measure each producer
group’s degree of organisational progress a tool has been designed based on the
EFQM Excellence management model, which comprises five criteria (leadership and
strategy, financial resources and materials, people management, alliances and
processes) and four indicators (beneficiary results, people results, customer results, key
results) which allow identifying areas for improvement with an eye to market
sustainability. The decision-making centre can score each producer group in line with
the annual progress reflected in the score obtained in the EFQM model designed.
Level
Δ EFQM
score
aij*
Very high
High
Medium
Low
Very low
>100
70 to 100
30 to 70
10 to 30
<0
1
0.85
0.65
0.45
0
Table 2. Experts’ ratings for the sub-criterion: ‘level of sustainability’.
4.2.2. Market criterion
•
Sub-criterion: ‘Ability to adapt’ to changes suggested by the organisation. The producer
group’s degree of adaptability to the changes suggested by the organisation for
continuous improvement.
Level
Very high
High
Medium
Low
Very low
Definition
Shows no resistance to change and quickly implements
the required changes and proposes new improvements to
remain sustainable in the traditional market
Shows no resistance to change and implements the
proposed changes
Accepts the changes to be carried out but with some
reluctance
Shows resistance to change and finds it difficult to adapt to
the changes
Shows strong resistance to change and does not agree to
make changes
Score
aij*
5
1
4
0.85
3
0.55
2
0.35
1
0
Table 3. Experts’ ratings for the sub-criterion: ‘ability to adapt’.
333
•
Sub-criterion: ‘Developing new products’ and innovation. The creation of new products
and the ability to adapt to the demand required by the organisation.
Level
Very high
High
Medium
Low
Very low
Definition
The producer group regularly proposes developing new
products and successfully adapts, insofar as possible to the
demand for new products
The producer group regularly proposes developing new
products but does not always successfully adapt to the
demand for new products
Occasionally proposes developing new products but does
not always successfully adapt to the demand for new
products
Rarely proposes developing new products and does not
adapt to the demand suggested by the organisation
Never proposes developing new products and does not
adapt to the demand suggested by the organisation
Score
aij*
5
1
4
0.85
3
0.65
2
0.25
1
0
Table 4. Experts’ ratings for the sub-criterion: ‘developing new products’.
•
Sub-criterion: Product potential’. Level of success of the products in the preceding
financial year classified according to product type, sales channel, purchases, and sales
and profitability of the product in the two previous financial years.
Level
Very high
High
Medium
Low
Very low
Products
sold
95-100%
75-94%
55-74%
35-54%
0-34%
aij*
1
0.85
0.65
0.45
0
Table 5. Experts’ ratings for the sub-criterion: ‘product potential’.
4.2.3. Quality Criterion.
•
Sub-criterion ‘Meeting specifications’. Level of quality of products supplied.
Level
Very high
High
Medium
Low
Very low
Compliant
products
95-100%
90-94%
85-89%
75-84%
0-74%
aij*
1
0.85
0.65
0.45
0
Table 6. Experts’ ratings for the sub-criterion: ‘meeting specifications’.
334
•
Sub-criterion ‘Delivery date’. With the dispatch of an order, the purchaser proposes a
shipping date. If the producer agrees, the purchaser will deem the delivery date to be
agreed. If the producer does not accept the proposed date, they negotiate the delivery
date or either partial delivery dates are agreed or a final date.
Total number of
days’ delay
0 to 5
6 to 10
11 to 15
16-20
>20
Level
Very high
High
Medium
Low
Very low
aij*
1
0.85
0.65
0.45
0
Table 7. Experts’ ratings for the sub-criterion: ‘delivery date’.
4.3. Multicriteria assessment applied to the assessment and selection of suppliers.
Results of the model.
The different methods of multicriteria Assessment have a series of characteristics,
requirements and properties that define each one individually. Likewise, the type of
assessment, type of data to be considered, the nature of the objectives as well as the decisionmaking centre’s point of view, have a bearing on the choice of a particular method (Qureshi et
al., 1999). Other issues, such as computer system capability should also be assessed when
deciding whether or not to use a specific method, since although the initial input information
may occasionally be similar whatever method is used, certain components of its internal
assessment structure may condition the way they are used (Hwang and Yoon, 1998).
Intelligent phase
Defining the problem
One objective.
Various
objectives.
Conflictive.
Complementary.
Understanding and acceptance
of the context of the decision
and the study.
Agreement regarding the
problem.
Generating alternatives. Identifying and
shaping criteria
Scoring criteria
Standardisation
Weighting criteria
Design phase
Discussing and accepting the models:
preparing decision matrices
Selecting assessment methods
Selection phase
Selecting alternatives
Sensitivity analysis
Recommendations and explanation
of results
Figure 4. Phases of a multicriteria process (Barba-Romero y Pomerol, 1997; Malczewski, 1999).
335
An analysis of the problem set has revealed the following aspects:

The number of producer groups, at present, is modest and small in number (44 in this
case) and it is aimed to keep this as such by providing an outlet to the traditional market
for these and an entry for more needy organisations into the Fair Trade network.

The decision will not only involve groups entering or leaving the network but also in what
proportion the budget allocated for Fair Trade purchases will be distributed among the
different organisations.

Both quantitative and qualitative information is worked with.

The decision-making centre will be made up of Fair Trade experts.
Bearing these points in mind, the Saaty Analytic Hierarchy method will be used to know what
weighting is obtained by each organisation assessed in line with the criteria defined above.
Having identified the assessment method to be used, together with the different criteria to be
borne in mind in the model, and having decided the weighting for each of these criteria, it will
then be possible to know which organisations are best prepared to be sustainable in the
traditional market and which parameters can be improved in other organisations to enable them
to continue advancing in this direction. The following figure shows the weightings obtained for
each organisation after applying the analytic hierarchy methodology:
Figure 5. Assessment of alternatives regarding the general objective.
336
5. Conclusions
Taking into account the results of the model, organisations can be classified into four groups
according to their growth rate and their product market share. In so doing, the
recommendations and steps to be followed can be made for each of these groups.
M
A
R
K
E
T
0.05
DEVELOPMENT
Market share
Figure 6. Classification of organisations in respect of Development and Market criteria.
3
1
4
2
Growth rate. Development
Figure 7. Classification matrix of Fair Trade organisations.
337
According to the results of the model the supplier organisations of Fair Trade products can be
classified into 4 types:
Type 1 organisations
⇒ Good level of organisational development.
⇒ High market share of the products marketed.
⇒ Organisations prepared for access to the traditional market.
⇒ Consultancy on international markets.

Health and safety standards for products.

Required certificates and quality seals.
⇒ Good level of organisational development.
⇒ Low market share of the products marketed.
⇒ Organisations that need to improve their product to pass to level 1.

Improvement of the process of the developed product.

New product lines.

Successful consultancy on other products: information and assessment on
European trends and fashions.
⇒ Low level of organisational development.
⇒ High market share of the products marketed.
⇒ Organisations that need to improve their management system.
⇒ Technical and administrative training.
⇒ Recommendations on management systems.
⇒ Indicators on progress in organisational management.
⇒ Low level of organisational development.
⇒ Low market share of the products marketed.
⇒ Organisations at risk of leaving the Fair Trade network.

Need to improve their management system to pass to level 2, or

Need to improve their product to pass to level 3.
338
References
Anaya Tejero, J. J., 2000 Logística Integral. La gestión operativa de la empresa. Editorial ESIC,
Madrid.
Ballou, R., 1999 Business Logistics Management. Planning, Organizing and Controlling the
Supply Chain. Fourth edition. Prentice Hall. New Jersey.
Barba Romero, S. Y Pomerol, J., 1997: Decisiones multicriterio. Fundamentos teóricos y
utilización práctica. Colección de Economía. Servicio de Publicaciones. Universidad de Alcalá
de Henares. Madrid.
Beuthe M., Eeckhoudt, L., Scanella, G., 2000. A practical multicriteria methodology for
assessing risky public investments. Socio Economic Planning Sciences, 34.
Colson y De Bruin., 1989. Models and Methodsin in Multiple Objectives Decision Making.
Pergamon. London. 1989.
Fernández Fábrega, J.M., Santadreu Capdevilla M. J., 2003: Claves estratégicas en compras y
aprovisionamientos. AERCE, Barcelona, 1ª edición.
Herrera Umaña, M., Osorio Gómez, JC., 2006 Modelo para la gestión de proveedores
utilizando AHP difuso. Universidad ICESI.
Hwang, C., Yoon, K., 1998. Multiple attribute decision making methods and applications. Berlin.
Springer Verlag.
Informes de Newsletters., 1997. La revolución de la gestión de compras. Cadena de
proveedores, Gestión logística y Nuevas tecnologías. Editorial Recolectos S.A.
Jankowski, P., Nyerges, T.L., Smith, A. Moore, T.J., Horvath, E., 1997. Spatial group choice: a
SDSS tool for collaborative spatial decision making. International Journal of Geographical
information Science, 11 (6).
Malczewski, J., 1999. GIS and multicriteria decision analysis. New York. John Wiley and Sons
Inc.
Mareschal, B., 1986: Stochastic multicriteria decision making and uncertainty. European
Journal ot Operational Research, vol. 26, pp. 58 – 64.
Prida Romero, B., Gutiérrez Casas, G., 1996. Logística de Aprovisionamientos. El cambio en
las relaciones proveedor-cliente, un nuevo desafío para la empresa del siglo XXI. Mc Graw Hill,
Madrid.
Qureshi, M.E., Harrison, S.R., Wegener, M.K., 1999 Validation of multicriteria analysis models.
Agricultural Systems, 62.
Romero, C., 1993. Teoría de la decisión multicriterio: conceptos, técnicas y aplicaciones.
Alianza. Madrid.
Saaty, T. L., 1980: The Analytic Hierarchy Process: Planning, Priority setting, resource
allocation. Mc Graw Hill, New York.
Saaty, T. L., 1994. Fundamentals of Decision Making and Priority Theory with the Hierarchy
Process, Volume VI the Analytic Hierarchy Process Series, RWS Publication.
Voogd, H., 1983 Multicriteria Evaluation For Urban and Regional Planning. Pion. London.
339
José Ramón Cobo Benita
C/José Gutiérrez Abascal, 2. 28006 Madrid.
ETSIngenieros Industriales.
Universidad Politécnica de Madrid.
Phone: 913363146
340
Area: Production Process Engineering
TOPIC AREA: PRODUCTION PROCESS
ENGINEERING
341
DEVELOPMENT OF A WATER TREATMENT PLANT
FOR A ROLLING-MILL TRAIN
Martínez, S.
Goicoechea, I.
Universidad de Vigo
Abstract
A plant of water treatment consists of an installation in which the water is treated and submitted
to diverse processes depending on the industrial use to which it is destined. In case of a plant
of treatment of waters proceeding from the refrigeration of a constant train of lamination, the
elimination and separation of oils, filtration, refrigeration or the thickened one and dehydrated of
mires they are some of the processes to which these waters surrender. In the majority of these
plants, a circuit is dedicated to the direct refrigeration (water circuit of contact) and other one to
the indirect refrigeration (water circuit of not contact). As the water circuit of contact it takes
pollutant materials that reduce the useful life of the equipments of the rolling mill the principal
aims of the plant they will be to extract these pollutant materials and to reduce the temperature
of the fluid that is re-circulated to the train. In order to design the process we use a series of
pre-established technical requirements, such as, for example, the increase in temperature
needed to cool the fluid, the speed, flow, and pressure at specific points and the different
concentrations of materials permitted in the water.
Beside fulfilling the above mentioned requirements, one tries to optimize the process applying
the following criteria:
- To re-circulate the whole of the wealth that flows for the lines.
- To reduce the consumption of the water.
- To improve to the maximum the quality of the water that is re-circulated to the train.
- To reduce the costs of maintenance, project of installation and costs of equipments.
Using this information we will determine a water treatment process and then design and
dimension the civil works. The fluid conduits will be calculated, and the equipment for pumping,
filtering, cooling, extracting solids and oils will be selected, along with the instrumentation
needed to control the plant.
Keywords: Waste-water purification plant, rolling-mill train, water treatment.
1. Introduction
In the rolling-mill train the “billet” used as raw material is heated to 1080 ºC using natural gas in
a glory hole with a radiant dome and three temperature control zones: preheating, heating and
equalisation. The billet moves forward by means of walking beams and lateral discharge and
343
exits the furnace by means of an internal roller conveyor. When the product exits the furnace it
is sent via a roller conveyor to the rolling-mill train formed by 18 continuous lamination jigs
which reduce the diameter of the material for market consumption (after it has passed through
a blunting and shearing area).
Figure 1: Billet exiting the furnace.
The material then passes on to two different final phases.
•
First phase for the production of sections, rounded bars and corrugated bars.
•
Second phase for the production of wire rolls.
The water has the following function in the lamination process:
All rolling-mill trains must have a water treatment plant for cooling purposes and to eliminate the
particles which the water picks up in the lamination process.
In the entire process water is present with the main function of cooling, essentially due to the
need to protect the machinery which enters into contact with the material during the production
process from excessive temperatures, performing the quenching of the product in the water
boxes and the cooling of lubrication oils and hydraulic equipment.(Montalvo,2007)
There are two different effluents to be treated: contact water and non contact water.
Contact water cools the different points and materials subject to high temperatures by contact.
It is a non-watertight circuit which cools the rolling-mill train (guides, rollers, etc.) as well as the
furnace. It is also used in the flushing process, which consists in the application of a direct jet of
water to eliminate remains of steel deposited on the furnace floor. This effluent is subject to a
series of conditions that increase its temperature, and it is also mixed with solid particles
(mainly scale - iron oxides), plus grease and oil which is removed by the effluent. (Mataix,
2006)
In the case of non contact water the circuits are watertight and cooling affects the billet
heating furnace components and the oil in hydraulic systems and oil used for lubrication in the
production process.
These cooling points have their own filtering equipment, and so the non contact water is only
heated during the productive process.
344
The two types of effluents must be treated in totally independent circuits, without any possible
communication between them.
2. Environmental study
All of the Regulations to be considered for the Supply of Water, Discharge and Purification, the
Environment, Environmental Impact and Waste, are analysed, and the following points
considered:
2.1.
Waste Products. Destination and Management.
The following products may be analysed as elements which have an environmental impact:
● Drainage Water
The appropriate authorisation for discharges used for the discharge of controlled water is
required for the system. In addition, a Water Service Technical Report may be provided
certifying this for any such plant. Said report shows that the concentrations of chlorides and the
conductivity present in the drained water do not exceed the concentration limits established by
the council’s General Urban Plan.
● Inorganic sludge after the filter press process.
This sludge is the result of the process performed in the filter press, and once obtained is taken
to the corresponding recycling plant.
● Chemical additives.
The chemical additives are stored in the tank in such a manner that they do not come
into contact with the atmosphere until they are mixed with the water. These additives,
responsible for carrying out a chemical process, have the following functions:
-
Biocide action
-
Dispersing action
-
Rust inhibitors
-
Incrustation inhibitors
● Oil and grease from the separation process in the Oil Skimmer.
This oil collects in a gutter in the decanters and is carried to an oil tank (after having
been separated from the water). From this tank it is carried to the corresponding recycling
plant.
2.2.
Water inflow requirements: Consumption and source
Water inflow comes from a river located some 67 metres from the scale ditch and 108 metres
from the non-contact reservoirs. These are the inflow points. The effluent receives an antialgae treatment via the incorporation of a biocide product which is automatically added using a
pump and tank.
345
It is used when the reservoirs are first filled during plant start up, and later, on occasions when
the concentrations have to be varied.
3. Specifications to be satisfied by the facilities
3.1. Requirements in the effluent circuits
In general terms, two water circuits are required, one for the contact water, and another for the
non-contact water. The contact water circuit requires 2500 m3/hour, according to the supplied
design information, and has been designed to support up to 3000 m3/hour and a pressure of
6.5 bar, plus the pressure required to overcome losses. This water flow is divided into two
circuits: the first circuit (1400 m3/hour) will be used from the kiln to box 18; the remaining flow
will supply the quenching process and will be used from box 18 to the output of the lamination
process. The water is then treated in the Scale Pit (or in the Receiver Pool in the case of the
water from the quenching process and the water used to cool the area from box 18 to the
outflow). It then undergoes a second decantation in the dual decantation reservoirs and the
process continues via the oil skimmer, filtration and cooling, with cooling towers calculated to
withstand a temperature variation of 15 degrees centigrade (45ºC inflow and 30ºC outflow).
The non-contact water circuit requires a flow of 1200 m3/hour and has been designed to
function with a flow up to 1440 m3/hour at a pressure of 5 bar, plus the pressure required to
overcome losses. It will supply the kiln/lamination circuits and the lubrication circuits. The water
is then cooled using towers which are designed to withstand a temperature drop of 15 degrees
centigrade (45ºC inflow and 30ºC outflow).
For the evaluation of the drops in pressure in the water treatment system's pipes, the maximum
speed for the circulation of the water has been taken to be 1.5 m/sec. As a consequence the
characteristics of the pumps and the dimensions established for the pipes have been chosen in
accordance.
3.2. Effluent flow
The flows of effluents supplied by the water treatment plant will depend on the operating status
of the production process. The following cases have been identified:
•
Full production: the nominal flows of the effluents correspond to this situation.
•
No production and the kiln hot: in this situation production is halted, but the kiln is hot and
ready to begin production in a short period of time. This situation corresponds to moments
when different types of repairs and adjustments to the production machinery are being carried
out.
•
No production and the kiln cold: all of the machinery has been halted. There is no need to
provide cooling water.
In all the water flow figures for the design of the water treatment plant, an overload margin of
20% has been used. Tables 1 and 2 show the flows for contact water and non-contact water
respectively, in each area, once overdimensioning has been applied.
346
CONTACT WATER FLOWS
NOMINAL PRODUCTION FLOW
OVERDIMENSIONING
(m3/h)
DESIGN FLOW
(m3/h)
Kiln-box 18 inflow
1100
20%
1320
Box 18-kiln outflow
900
20%
1080
Thermal treatment
500
20%
600
Total to be processed
2500
20%
3000
OVERDIMENSIONING
DESIGN FLOW
NOMINALS FLOW WITHOUT PRODUCTION
(m3/h)
(m3/h)
Hot kiln
1000
20%
1200
Cold kiln
0
20%
0
Table 1: Contact water flows
NON-CONTACT WATER FLOWS
NOMINAL PRODUCTION FLOW
OVERDIMENSIONING
(m3/h)
Total to be processed
DESIGN FLOW
(m3/h)
1200
NOMINALS FLOW WITHOUT PRODUCTION
20%
1440
OVERDIMENSIONING
DESIGN FLOW
(m3/h)
(m3/h)
Hot kiln
1200
20%
1440
Cold kiln
0
20%
0
Table 2: No-contact water flows.
3.3 Physical-chemical characteristics of the effluents
The following tables show the physical-chemical characteristics of the effluents in terms of
solute concentrations, Ph, temperature, pressure, etc. during the normal operation of the
system.
347
CONTACT WATER
Ph
7-8,5
Alkalinity
50-300 mg/l
CaCO3
30-380 mg/l
Mg2
0-120 mg/l
Cl max
250 mg/l
Fe max
0,5 mg/l
SO4 max
200 mg/l
PO4 max
0 mg/l
SiO2 max
75 mg/l
TDS (total dissolved solids)
1500 mg/l
Total suspended solids
50-80 mg/l
Maximum size of suspended solids
200 µm
Maximum oil content
10 mg/l
Minimum pressure
4 bar
Maximum temperature
30 ⁰C
Table 3: Physical-chemical characteristics of the contact water.
NON‐CONTACT WATER Ph 7‐8,1 Alkalinity 3,2 mg/l CaCO3 19 ppm +2
Ca 240 ppm Mg+2 84 ppm Cl‐ 102,9 ppm Fe+3 4,4 ppm PO4 1,85 ppm SO4 725 mg/l HCO3 195,2 ppm Total suspended solids 10‐20mg/l Maximum size of suspended solids 0,1 mm Minimum pressure 4 bar Maximum temperature 30 ⁰C Table 4: Physical-chemical characteristics of the non-contact water.
348
Bearing in mind that, amongst other reasons, given the evaporation of the water in the
effluents, the concentrations of solutes will tend to increase, the water treatment plant, should,
as far as possible, counteract this effect. During continuous operation the plant will not be able
to treat the concentration of each solute independently. As a result, the measure used to control
the level of solutes will be by means of conductivity meters, measuring the conductivity of the
contact and non-contact water. Based on these readings the circuits will be flushed with clean
water. The plant will have suitable resources to perform this operation.
The effluents will be subject to periodical chemical analyses, performed by qualified staff. They
will determine, based on the information received, any preliminary treatments needed and the
amount of water inflow required so that the flushing operations have the desired effect. They
will also determine the frequency with which said analyses need to be performed and the
corresponding conductivity of the aforementioned solute concentrations. The water treatment
plant will not contain devices for the chemical conditioning of the inflow water, and will only
have effluent drainage and inflow lines. As a consequence of this, the water treatment plant
must ensure that the following parameters are observed for the effluents:
CONTACT WATER Conductivity To be determined Maximum size of suspended solids 200 µm Maximum oil content 10 mg/l Minimum pressure 6,5 bar Maximum temperature 30 ⁰C Table 5: Parameters for the contact water.
NON‐CONTACT WATER Conductivity To be determined Minimum pressure 5 bar Maximum temperature 30 ⁰C Table 6: Parameters for the non-contact water.
It should be borne in mind that, for the contact water circuit, the evaporation of water as it
passes through the production process has been established at 2.5% of the total flow. All of
the devices used to cool the non-contact water have individual filtering equipment. Thus, control
of the size of solids in suspension in this effluent will not depend on the water treatment plant.
3.4 Particles removed by the effluents
During cooling, when the effluents pass through the production process, their temperature
increases and they also remove a series of different particles. The contact water removes
349
solid particles of different sizes, mainly scale. If we consider that the maximum capacity of the
rolling-mill train is 90 Tonnes/hour, and we estimate a formation of scale equal to 0.2% of the
production total, the maximum amount of scale that the contact water will remove is 18 Kg/hour,
with a density of 7.65 g/cm3.
It will also remove a maximum quantity of 40 l/hour of oil and grease from possible overflows
and leaks in the greasing and lubrication mechanisms.
Given that the contact water circuit is closed and water-tight, it will not remove significant
amounts of particles. Moreover, all of the equipment which functions in the productive process
possesses its own individual filtering devices. As such, for the treatment process of this effluent
the possibility of the existence of particles which interfere with the normal operation of the
circuit will not be considered.
3.5 Thermal information
The maximum temperature of the water inflow after having passed through the production
process will be 45ºC; the water treatment plant will be capable of reducing its temperature tp a
value less than 30ºC. In relation to tower cooling a wet bulb temperature of 21ºC is applicable.
3.6 Refrigeration towers
The possible presence of the legionella bacterium in the cooling towers needs to be borne in
mind, as this bacterium develops rapidly in water in the temperature range of 20 to 45ºC, and it
is carried on the air. The following design factor swill be considered:
•
The cooling towers will have highly-efficient drop separators. (The amount of water
removed will be less than 0.1% of the flow passing through the tower.)
•
To avoid the formation of rust in the towers plastics, galvanized steel and stainless steel
will be used.
•
Sections of dead-end pipes will be avoided, and drainage valves will be fitted.
•
The equipment will be accessible for their inspection and cleaning.
•
The water collector will have a minimum gradient of 2% to drain off the water and will
also have water outlets.
Other preventative actions will be carried out via factors external to the design, as a specialised
company will be hired which will be responsible for the inspection and cleaning of the
equipment.
Disinfection will be carried out using authorised disinfectants. 5 ppm of chlorine will be injected
along with the biodispersant in the tray. The pumps will then be left running for five hours with
the fans switched off.
Next, all the water will be removed from the circuit and it will be thoroughly cleaned, with further
water added until the drained off water appears to be clean. Lastly, it will be filled with clean
water and 10 ppm of chlorine will be added, with the pumps operating and the fans switched off
for five hours. The chlorine levels will be checked every hour.
The water’s conditions will be controlled in a continuous, automatic manner, via the flushing out
of dirty water and replacement with clean water, the addition of biodispersants and biocides,
350
lime prevention agents, and agents designed to prevent the rusting of the metal parts in the
circuit.
The water will be drained from the tray when the apparatus is no longer in use.
3.7 Surface treatment of tubing which is prone to rusting
All of the pipes which are prone to rusting must be treated on the outside with an appropriate
coating for outdoor weather conditions, with the average lifetime for the system being 15 years.
3.8 System reliability
All of the effluent pumping systems will have redundant systems (one reserve system). This will
start automatically if any of the pumps suddenly fails. Floating level indicators will be fitted and
will operate in conjunction with the ultrasonic level indicators to thus increase the safety of the
level measurements if the ultrasonic level indicators fail.
3.9 Power supply
The electrical power supply for the water treatment plant will be a triphase voltage of 380 V / 50
Hz, a monophase voltage of 220 V / 50 Hz , and an earth connection.
It will also have a 7 bar compressed-air pipe, used to supply and operate the pneumatic
equipment, and another 2 bar line to control the lines using suitable devices (such as pressure
gauges).
351
4. Process description
Figure 2 shows a simplified version of the process followed in the water treatment plant.
As we mentioned in section 1, there are two water circuits: the contact water circuit and the
non-contact water circuit. (Diaz, 2007). The path of each circuit is described below.
Figure 2: Basic schema of the process in the plant.
This is a closed circuit where water is not contaminated by scale particles, grease or oil. The
circuit runs between the cooling towers and the train, where the water cools the lubrication
systems and some of the furnace components. There are two phases in this process: effluent
cooling and feedback to the production process.
As can be seen in figure 2, part of the flow used for cooling the furnace, during lamination, and
for cooling purposes during quenching is directed to the reception pool or the mill scale pit; the
other part of the flow is used to cool the rest of the furnace and lamination.
In the billet heating furnace water is used as a fire damper in open areas of the combustion
chamber. In the lamination section this effluent is sent via manifolds to each lamination cylinder,
and a direct jet of water is applied to the surface of each cylinder in order to protect them from
overheating due to contact with the material in the production process. The jets also remove
scale particles from the cylinders’ surfaces which could cause defects in the product. As a
result of passing through the interior of the lamination boxes the effluent contains oil and grease
particles (from leaks and spillage in the machine’s lubrication systems).
The flow used in the quenching process has a temperature of less than 30ºC and is directed
through tubes from the water treatment plant to the suction header of the quenching pumps. It
is then carried to the inside of the guiding tubes through which the material circulates in the
production phase (shaped so that the material acquires the required physical properties).
As a result of the direct contact between the water and the incandescent steel (of the product)
the water carries scale particles to an open channel which is used to direct the effluent. The
channel has the same morphology as that of the furnace/lamination area.
352
Figure 3. Layout of the contact circuit in the train.
After having performed these functions, the water is carried by force of gravity through two
similar channels. The channels have a gradient which ensures that the effluent circulates at a
sufficient speed so as to avoid sedimentation of the solids in suspension. As can be seen in
figure 3, one of the channels leads to the scale pit and the other to the reception pool.
The flow being treated consists of water with a maximum temperature of 45ºC, with scale
particles of different sizes, and a maximum of 40 l/hour of oils/grease. The loss of effluent via
evaporation during the production process will be 2.5% of the total inflow.
The phases of this circuit are as follows:
•
Separation of solids in suspension
Separation of solids in suspension is carried out by decantation in several phases taking into
account that the solid particles to be separated in the first phase will be scale particles, with a
density of 7.65 g/cm3.
This phase is performed in both the scale pit and the reception pool, where the water is carried
by force of gravity.
The second phase of decantation takes place in the decanters, where the water is sent through
a pressure pipe line from the pumping areas of the scale pit and the reception pool.
In addition, the decanters will receive water from the cleaning of the filters, from the thickener,
and the floating material reservoir, after carrying out their corresponding processes as
explained in detail below. The dimensions of the decantation reservoirs will be determined
taking into account that the minimum size of the particles to settle in the scale ditch is 200 µm
(worst possible case), with a variable decantation time in each reservoir. In the scale ditch the
particles are heavier and less time and decantation space will be required. The decanted solids
353
will be extracted using a suspended amphibious grab and will be deposited in areas destined
for that purpose. The decanters will be calculated for a decantation time of 75 minutes.
•
Separation of oils
Oil separation occurs in a pond designed for that purpose. Separation takes place due to the
difference in density between the oil and water. The oils tend to form a film on the surface of the
water. This is then extracted using a belt oil remover. The extracted oil is then sent to a
hermetic storage tank.
•
Filtering
As a precautionary measure a series of automatic-cleaning filters with a micronage of 200 µm.,
are used, in case any particles are still present after the decantation phase.
•
Cooling of the effluent
The flow to be processed is sent by means of a series of pumps through the abovementioned
bank of filters to the top of several prefabricated cooling towers. The cooled water is stored in
pools beneath the towers which support the towers themselves.
•
Feedback to the production process
Part of the total flow is pumped at a specified pressure from the contact water ponds to supply
the furnace-lamination areas; the rest of the flow is carried by its own weight from the pool to
the quenching area.
5. Conclusion
If one compares the design of this water treatment plant to that of the majority of the plants
currently used in the iron and steel industry, the following points standout:
•
The water treatment process is based on decantation and not filtration (that is why time
required for decantation is so high). It results in economic savings as less filtration equipment is
required.
•
The decanters are rectangular and not circular, which results in
more uniform
sedimentation. As for the scraping beams, the force which the beam applies when moving
along the pond is also more uniform, whilst in the case of beams used in circular decanters the
end of the beam furthest from the centre is subject to much higher momentum, thus requiring
the use of a more powerful electrical engine and a more robust beam.
•
Filtering before the cooling towers is performed using a ring filter and not sand filters. The
most important advantages of using this system are as follows:
In the sand filters, during the washing phase generated by back-flow water supported by
an air flow in the same direction, of particles of scale are found in the filters then, the back-wash
will first expel sand particles of a lower density. These scale remains will be retained in the
filter. When this scale rusts it may cause discolorations and impurities in the outflow water,
amongst other effects. This is not the case in the ring filters. Here the particles are trapped
between the lower and upper discs.
A considerably lower amount of cleaning water is required during the cleaning phase of
the ring filter. A specially designed system allows the rings to be separated which helps when
removing the filtered particles from the equipment. For sake of comparison, in the set of
354
designed ring filters, the cleaning phase may last a total of 6 minutes every two hours, creating
a washing flow of 5.3 m3/every two hours. With an equivalent sand filter, the duration of the
cleaning phase may be 10 minutes per filter. If a system with 6 filters was designed, then this
would mean 60 minutes every two hours, that is, 250 m3/every two hours. This would create the
need for more tubing, pumps, pool civil works, consumed energy, introduction of sand into the
system, recycled water flow, etc.
-
The plant only requires a 9 by 2 metre area, and thus takes up less space.
-
This avoids the need for reloading or cleaning of sand because of oil or grease.
Filtering efficiency is constant and never varies: it does not vary as the size of retained
particles depends on the distance between the rings; and filtering is constant because
“preferred channels” do not form (which occurs in sand filtering).
This means that the proposed solution is more compact, more efficient, and requires
less energy and civil works.
For these reasons and also considering the experimental results obtained in plants with similar
designs, we may conclude that this design has several advantages with respect to others.
References
CEPIS-Environmental Health protection Programme-HPE, 1981. Evaluación de plantas de
tratamiento de aguas. Tomo I, Manual DTIAPA C-5.
Diaz,O and House, M (Russula), 2007 Improved steelworks process water treatment.
Millennium Steel International Edition, Editor: Dr. David James Price, pp 167-169
Giama (Universidad de La Coruña) and Russula., 2006. Estudio de Generación y de la
tratabilidad de las aguas residuales producidas en un tren de laminación de acero en caliente.
Universidad de la Coruña.
Mataix, C., 2006. Mecánica de fluidos y máquinas hidráulicas. Ediciones del Castillo., pp 660
Montalvo L.T, 2007. Ingeniería hidráulica. Ediciones VJ, pp. 420
E. T. S. Ingenieros Industriales y de Minas. Universidad de Vigo. Departamento de Diseño en la
Ingeniería. Área de Proyectos en la Ingeniería.
Campus Universitario Lagoas - Marcosende 36200 Vigo
Phone: +34 986 813647
Fax: + 34 986 812201
355
WORKPLACES REDESIGN IN A FOOD INDUSTRY TO MINIMIZE
MUSCULOSKELETAL INJURIES IN WORKERS
Planas, A.E.
Ducun, M.
Tomás, J.A.
Azcona, I.
Mutua Universal
Abstract
Musculoskeletal disorders are a cause of concern not only because of the health effects on
individual workers but also because of the economic impact on businesses and social costs.
More than 70% of occupational diseases and 33% of accidents declared at work with days off
work are musculoskeletal. Working layouts and processes, awkward postures, high demand
repetitive movements, sustained efforts, insufficient muscular rests, are responsible for these
losses. Optimal workplace design, selection of appropriate equipment and the ability to predict
musculoskeletal load and physical demands of each process are key tools to prevent the
emergence and development of injuries. A case study conducted in a food industry which
recorded continuously days off work because of epicondylitis and wrist tendinitis is presented.
The factory had old machines, without economic possibilities of being renewed. Muscular and
biomechanical demands on workers, ergonomic risk factors and characteristics of various
production processes were analyzed. Workplaces were redesigned, accessories to the old
machines were joined and production processes were changed. As a result, a performance
improvement, an increase in productivity and the disappearance of days off work because of
epicondylitis and tendinitis were obtained.
Keywords: redesign, workplace, ergonomics, musculoskeletal, off work
1. Introduction
This case study is presented as a result of an ergonomic analysis carried out in a food industry
dedicated to packing and distribution of snack products. In the last year 49 work accidents with
days off and 3 accidents with relapse were registered for a staff of 390 workers. The incidence
rate was 125,6 accidents every 1000 workers, this number is higher than the average one in its
activity sector that was 107,4. 34 of such accidents were musculoskeletal and mainly located at
hands, arms and back. Several cases of epicondylitis and tendinitis in wrist were diagnosed. In
addition to accidents with leave, workers expressed to suffer from discomfort in hands, wrists,
forearms, lumbar spine, cervical spine and legs.
Activities related with the most of the injuries where located at packing products line. The main
part of these activities involved picking up packets from a conveyor belt, putting them into
356
boxes or bags by hand according to its size, closing boxes or bags by hand and putting them
into a pile to transport them. A packet weighted between 100 g (the lightest) and 1 Kg (the
heaviest), conveyor belt speed was regulated according to this parameter.
Afterwards a detailed analysis of the ergonomic risks in the packing line was developed. It was
found that the critical activity was the packing of nuts in 1 Kg packets. Workers exerted
repetitive movements of fingers, hands and forearms in addition of high and continuous forces,
standing for a long time in the same place.
2. Objective
Once workplace environment and processes causing injuries and discomfort in workers were
identified, a collection of changes in procedures and layout of the workplaces were suggested,
specially the redesign of packing nuts in packets with the greatest weight.
The objective of this work is to valuate quantitatively the improvement achieved in muscular
activity and energy worker solicitudes by the comparison of the old packing workstation design
with the new packing workstation redesign. This previous assessment before implementation
will be useful to guarantee the investment.
3. Methodology
To detect risk factors related to the old design and to check the improvement related to the new
design has been applied the following methodology:
•
Register of conditions of the packing line during work in the food industry.
•
Simulation of working conditions at laboratory, comparing both ways of carrying out the
process: old design and redesign.
•
Film of the work processes.
•
Register of muscular activity of muscular groups responsible for movements and forces.
4. Results
4.1. Old Design
Packing nuts operation involves opening a plastic bag, taking 1 Kg nuts packets from the
conveyor belt, one by one, putting them into the plastic bag, getting the full bag into a knot, and
putting it into the container for finished products (see Figure 1). This cycle lasts 18 seconds on
average. Each worker handles on average 1000 Kg per hour (16,6 Kg/min). There is not a
stipulated break for a rest, workers only take a break when a change of product is made.
357
Figure 1: Packing of five 1 Kg-packets into a bag.
Risk assessment for this workplace yields the following results:
• High probability of developing specific injuries in distal part of upper limbs (epicondylitis, de
Quervain´s disease, carpal tunnel syndrome, trigger finger) and other non-specific injuries.
According to Moore and Garg (Moore, 1995), it has been obtained a strain index of 72.
• High manual forces related to repetitive handling of heavy products, higher than 1,6 times
values recommended by Snook and Vaillancourt (Snook, 1995) for similar tasks. The workers´
hands are tensed during 62% work cycle.
• Awkward postures for hands and wrists. Flexo-extension movements for right hand ranged
between 40º extension and 50º flexion, and for left hand between 70º extension and 30º flexion.
Lateral wrist movements ranged between 15º radial deviation and 40º ulnar deviation for left
wrist, and between 35º radial deviation an d 30º ulnar deviation for right wrist. All these
movements exceed completely limits recommended by UNE-EN 1005-5:2007.
• Flexion of neck and back, in a static and continuous way, with values higher than limits
recommended by UNE-EN 1005-4:2005.
4.2. New Design
To solve awkward postures, high forces and repetitive movements, workplace redesign was
recommended. To avoid stressful postures in hands and wrists, high strains in muscles and
tendons during picking, a new output ramp for the machine was redesign. This new output
ramp guided packets to final destination (a bag or a box). Then, workers operation only
involved pushing packets, without picking and without handling weights.
Figure 2 shows the redesign recommended for packing into sacks or bags. Figure 3 shows the
redesign for packing into boxes. With this new procedure, the final container for nut packets
was located at the extreme of the sloping conveyor belt. Nut packets could fall by gravity.
Worker only should supervise the operation and help for the correct guide of the packets to the
final container. It was recommended an automatic sealing device to avoid manual knotting for
bags, because it was proved that this action implied a high physical demand to the worker.
Putting the boxes or sacks on a mobile cart could make packets evacuation easier and could
avoid a new manual handling.
358
Nuts packet
Flange for turning packets
Conveyor belt
Mobile cart
Sacks
Figure 2: Redesign for packing into bags.
Nuts packet
Conveyor belt
Box
Mobile cart
Figura 3: Redesign for packing into boxes.
For packing into boxes, it was proposed to put mobile and adaptable guides suitable to each
packet size.
As workers were standing for long periods of time, it was suggested to change the standing
workstation for a sit-stand workstation. It was introduced a pedestal stool with adjustable height,
to let workers change posture and let legs muscles rest.
359
Figure 4: Pedestal stool to introduce a sit-stand workstation.
4.3. Comparison Old Design – New Design
Packing operation with the old design and with the new design has been analysed in order to
assess the physical load for the worker. The activity of muscular groups responsible for
movements and postures has been examined for a working cycle. Muscular groups analysed in
right and left upper extremities are flexors and extensors of the wrist, biceps brachii, triceps
brachii, trapezius and anterior part of deltoid.
Muscular activity has been registered by surface electromyography. To be able to compare
between processes, results have been calculated as maximum voluntary contraction (MVC) of
the worker.
Right Deltoid- anterior
part
C1 Right Flexors
C5 Left Flexors
C9
C2 Right Extensors
C6 Left Extensors
C10 Right Trapezius
C3 Right Biceps Brachii
C7 Left Biceps Brachii
Left Deltoid- anterior
C11 part
C4 Right Triceps Brachii
C8 Left Triceps Brachii
C12 Left Trapezius
Table 1: Muscular groups analysed.
Muscular Activity (%MVC)
C1
C2
C3
C4
C5
Old Design 19,7 20,4 20,4 21,7 18,5
New
Design
16,5
12
6
11
11,5
C6
C7
C8
C9
C10
C11
C12
12
10,4
5,7
9,2
11,5 10,5
8,1
10
7
5
3
6,5
4,5
4
Table 2: Muscular activity for each muscular group for old design and new design of packing workstation.
360
In Figure 4 is shown the comparative graph of muscular activity required in old design and in
new design for the packing workstation. For all muscular groups, effort suffered by the worker is
smaller in new design than in old design. Numerical values are presented in Table 2.
Muscular Activity
25
%MVC
20
15
Diseño
antiguo
Old Design
Rediseño
New Design
10
5
0
1
2
3
4
5
6
7
8
9
10 11 12
Muscular Groups
Figure 4: Muscular activity required in both designs.
New design of the packing workstation means a decrease in muscular activity of the right upper
extremity of 16% in flexors, 41% in extensors, 70% in biceps brachii, 49% in triceps brachii,
67% in anterior part of deltoid and 43% in trapezius. Also it means a decrease in muscular
activity of the left upper extremity of 37% in flexors, 16% in extensors, 32% in biceps brachii,
12% in triceps brachii, 62% in anterior part of deltoid and 44% in trapezius.
Muscular energy demands have been examined in packing to complete the analysis. As it is
shown in Table 3 and in Figure 5, the redesign of the workstation implied a decrease of the
energy demands in the right upper extremity of 61% in flexors, 72% in extensors, 85% in biceps
brachii, 76% in triceps brachii, 85% in anterior part of deltoid and 73% in trapezius. Also it
means a decrease of the energy demands for the left upper extremity of 71% in flexors, 60% in
extensors, 69% in biceps brachii, 57% in triceps brachii, 81% in anterior part of deltoid and 74%
in trapezius.
361
Muscular energy demands
3500
3000
uV.s
2500
2000
Old Design
Diseño
antiguo
New Design
Rediseño
1500
1000
500
0
1
2
3
4
5
6
7
8
9 10 11 12
Muscular Group
Figure 5: Muscular energy demands required in old design and in new design of the packing workstation.
Muscular Energy Demands (μV.s)
C1
C2
C3
C4
C5
C6
C7
C8
C9
C10
C11
C12
Old
Design
2084 3032 2367 1652 1578 1480 1449 717 1681 830 1563 721
New
Design
811
822
333
396
457
581
449
305
244
220
285
183
Table 3: Muscular energy demands required by each muscular groups in old design and in new design of the
packing workstation.
5. Conclusions
The redesign of the workstation of packing has allowed to reduce between 57% and 85% the
muscular energy demands to the worker. Movements and unnecessary manual handling have
been avoided, especially manual handling of weights. Postural comfort of workers have
improved. This has led not only to higher welfare for workers but the possibility of increase the
production rate without risks for health of workers. A small change in the output ramp of the
machine, the inclusion of a mobile cart, a device for automatic sealing bags and the purchase
of a pedestal stool with adjustable height have allowed this improvement.
A suitable ergonomic intervention implies a very positive cost-benefits relationship for the
company. To guarantee the success of the ergonomic intervention is important to know in detail
the technical faults of the workplaces and processes design about postures, movements and
forces demanded, what risks for heath they introduce, how they become evident, why they are
362
caused, and also to know which the most suitable options are, their viability and the way of
implanting them.
In many occasions, as the case study shown, a detailed analysis of working postures and
movements of the worker, the alternatives that a machine could take, the introduction of
unsophisticated tools and a small change in the layout of the workplace, all of them without a
high economical investment, allow to decrease the musculoskeletal discomfort in workers, to
improve the safety at work and to reduce the productive and social costs.
References
Moore J.S., Garg A., 1995. The strain index: a proposed method to analyse jobs for risks of
distal upper extremity disorders. American Industrial Hygiene Association Journal, 56(5), pp.
443-458.
Snook S.H., Vaillancourt D.R., Ciriello V.M., 1995. Webster B.S., Psychophysical studies of
repetitive wrist flexion and extension, Ergonomics, Vol. 38, pp.1488-1507.
UNE-EN 1005-4., 2005. Seguridad de las máquinas. Comportamiento físico del ser humano.
Evaluación de las posturas y movimientos de trabajo en relación con las máquinas.
UNE-EN 1005-5., 2007. Seguridad de las máquinas. Comportamiento físico del ser humano.
Evaluación del riesgo por manipulación repetitiva de alta frecuencia.
MUTUA UNIVERSAL
Laboratory of Ergonomics
Polígono Industrial Arazuri-Orcoyen, Calle C, nº 25, 31170 Arazuri (Navarra), Spain
Phone: +34 948 01 04 70
Fax: + 34 948 01 04 90
URL: http://www.mutuauniversal.net/m10/prevencionindex.html
363
Area: Project Management
TOPIC AREA: PROJECT
MANAGEMENT
365
ANALYSIS OF DECISION-MAKING MODELS FOR PROJECT
MANAGEMENT
Poveda-Bautista, R.
García-Melón, M.
González-Cruz, M.C.
Abstract
In the present paper an empirical study about the decision-making processes in Project
Management areas in valencian SMEs is presented. The work is based on previous studies
where it was concluded that:
-
the decisions made in Project Management environments are of the type multicriteria,
multiexpert and discrete,
Valencian SMEs do not use any Decision Support System.
The following work is focused on establishing relationships between the companies decision
making models and their Maturity Model in Project Management. The methodology used is
based on qualitative research techniques, e.g. focus interviews. These relationships obtained
will be used as guidelines to propose a Decision Support Methodology based on MCDA
techniques to help these companies.
Key words: Project Management, MCDM, PMMM, decision making
1. Introduction
A previous survey conducted by the authors of the present work (Poveda et al., 2005; García
et al., 2006; García et al., 2005) revealed that the analysis of Project Management decision
processes has been mainly based on existing approaches, models and tools for General
Management, a field which presents significantly different characteristics from the one under
study. Therefore there is a methodological niche and the urgent need for the development of
a decision analysis tool that takes into consideration the particular nature of project
management. This support tool will allow a more systematic and rigorous decision-making
process throughout the project life cycle as well as high excellence in the project
management process for successful project implementation.
On the other hand, the need of working with faster and more flexible organizational
structures, which forces companies to operate through projects that successfully materialize
their objectives, as well as a more competitive business market make it necessary that the
367
successful results of one project can be extended to future projects through the use of
standardized procedures. In this context, new management techniques and tools adapted to
project development that allow for systematization and standardization of project
management procedures have to be developed to reach acceptable levels of maturity in
project management.
Maturity in Project Management consists of developing repeatable processes and systems
leading to project success (Kerzner, 2000).
The present study is part of a wider research work on Project Management decision-making
processes with the main objective of developing a decision support tool to help Valencian
companies in their project management decision process. The tool will allow decision makers
to gain a better understanding of the decision problem and to model the main parameters
involved in the decision process (alternatives, criteria, criterion weights, utility aggregation
procedure …)
The decision support tool must solve decision problems found in Project Management (PM)
processes and adapt to the level of maturity of the companies involved in the process. In this
sense, the tool should provide a modelling system for solving complex PM problems.
Therefore, it seems reasonable to gain a deeper knowledge on the way decisions are made
as well as on the organizational features of the PM processes used in the companies under
analysis.
In a previous survey, a questionnaire was used to analyze the typology of the commonest
PM decisions and the way companies formulate the decision-making problem. The present
work describes the findings of an empirical study conducted on Valencian companies with
the aim of classifying these companies according to their PM culture using a model that
assesses their level of maturity.
2. Background
A previous survey conducted by the authors showed that the PM decisions made in
Valencian companies are of a multiexpert multicriteria and discrete type. Based on the
findings of that preliminary survey, in the present work the relationship between the use of
conventional decision analysis models and the company’s organizational culture in PM is
analyzed.
The aim is to gain a deeper understanding both of the formal aspect of the decisions and of
the organizational aspect of PM as a general framework for the application of the decisionmaking processes. For this end, in this work we analyze the level of maturity in PM within the
organizational culture of the companies involved in the projects.
It is difficult for most companies to accurately determine the level of application of PM
practices. One of the main reasons for classifying Valencian companies according to their
level of maturity is to establish an unbiased indicator of the degree of definition of the
processes and of the correct application of widely accepted PM practices and tools.
According to Ibbs and Kwak (Ibbs and Kwak, 2002) most companies consider the use of
practices and support tools useful in project management processes as they permit them to
adapt to changing business environments, yet they need a reference model for the efficient
implementation of such tools. Project Management Maturity Models (PMMMs) emerge to
provide companies with the necessary mechanisms to allow them to identify the key
368
opportunity and improvement areas in project management activities. Additionally these
models serve to develop comparison indicators for the application of PM practices and
techniques across organizations operating in the same business environment or sector.
Recent studies have identified the advantages of applying methodological practices and welldefined procedures in the area of PM. Most of such studies are based on research works
conducted in the professional and academic environments. Some studies basically focus on
qualitative assessment of PM practices and tools rather than on establishing quantitative
indicators (Al-Sedairy, 1994; Boznak, 1988; Bu-Bushait, 1989; Cleland, 1993; Deutsch, 1991;
Gross and Price, 1990; Kwak et al., 1995; Ziomek and Meneghin, 1984).
Other works analyze possible improvements in project success through the use of projectoriented organizational structures based only on the analysis of PM practices (Donnelly and
Kezsbom, 1993; Gobeli and Larson, 1986; Larson and Gobeli, 1989; Lundin and Soderholm,
1994; McCollum and Sherman, 1991; Might and Fischer, 1985). More recently, Dooley
(Dooley, 1998) has studied the impact of maturity and best practices on new product
development projects.
In this work we present a new proposal for the quantitative measurement of the level of
application of procedural methods and well-defined processes in the companies under study.
The measurement procedure is based on Project Management Maturity Models (PMMMs).
This will allow the quantitative comparison of the PM organizational culture in Valencian
SMEs with the aim of establishing a relationship between this company’s culture and the PM
decision process.
The results of the analysis will then be used to define adequacy criteria which in future works
will be used to develop a PM decision support system adapted to Valencian companies. For
this end, our study uses the data from questionnaires sent to the Valencian companies under
study. The results of the questionnaires were used to classify the companies according to the
decision making model used: well-structured decision models vs ill-structured decision
models.
3. Objectives
The objectives of the present work are:

To classify Valencian companies according to their Project management culture using
a model that measures their PM maturity level.

To study the relationship between the companies’ maturity in project management
and the decision process. With this objective in mind, the impact of organizational
maturity in project management on decision making processes is analyzed.
4. Research methodology
4.1 Stages of the survey
The different stages of the research are

To review the state of the art in the areas of Project Management and Decision Making.

To develop a methodology of general analysis of the companies’ maturity level and of
the use of decision making models.
369

To design a questionnaire to identify the decision model used in the company.

To analyze the data on the basis of how the company structures the decision making
process.

To strategically select the study cases (companies) of the survey.

To design an interview outline in order to measure the level of maturity in Project
Management based on PMMMs.

To assess the degree of application of methodological practices and well-defined
processes in PM (classifying the companies in levels according to PMMMs).

To analyze the companies’ strengths and weaknesses in decision making and PM.

To establish the relationship between the decision models used and the companies’
level of maturity in PM.
4.2. Survey design
The methodological approach is aimed at establishing the relationship between the
companies’ decision models and their level of maturity in PM. The methodology and survey
were based on the Maturity Models proposed by the Software Engineering Institute (SEI) in
1990, Kerzner (Kerzner, 2000) and other renowned authors, as well as on PMMMs used by
Project Management practitioners. The PMMMs propose 5 levels of maturity (initial,
repeatable, defined, managed and optimizing). The most significant features of the
methodology developed are:
Strategic
selection of
the study
cases. Exes
Company
size
Decision
models
Variables/Factors
-Project Management
-Decision Making
process
Model/Procedure
- PMMM
- Decision Model
vs. PMMM
Research methodology
Tools/Analysis
-Decision model
questionnaire / Statistical
analysis
-Interview guidelines
PMMM / Qualitative
analysis
Objective
Relationship
between
decision
models &
PMMM
Table 1: Methodology for establishing the relationship between decision models and maturity levels (designed by
the authors).
The analysis conducted at this stage of the research is the basis for the subsequent
development of a decision support tool. An in-depth interview with the decision makers was
considered the best way to reach the first objective, i.e. to better understand the relationship
between project management decisions and project management maturity.
The purpose was to discuss with the decision makers on key Project management issues:
Decision Making Models, and PM methodological practices and standardized procedures;
however a discussion on issues as intangible as these is a difficult task. For this reason we
decided to focus the analysis on specific situations so that the respondent explained his/her
370
experiences in more detail. These personal experiences were then used as the basis of the
discussion and further interpretation of relevant aspects for the present study.
The survey was devised in such a way that the respondent could analyze the project
management processes and activities conducted on a specific decision-making environment.
For example, decision makers were asked to describe the procedures for resource allocation
followed in a particular project, focusing on how resource allocation decision problems were
formulated.
The interviewed decision makers were asked to think about and describe situations
according to certain criteria:
The situations should belong to a well-defined Project: the Project should present
well-defined strategic goals, extend over a given time length and be strategic for the
company.
The situations should be complex in the sense that they involve highly-qualified
personnel from other departments and/or organizations and areas of knowledge.
The respondent should have been actively involved in the resolution process.
The situations should be extrapolable to the overall performance of the company in
similar decision problems.
With their descriptions, the respondents were asked additional questions about their
organization with the purpose of finding out the company’s level of maturity.
Therefore, most of the empirical work focused on the identification of paradigmatic indicators
about the companies’ level of maturity in project management.
At this stage of the research, therefore, the basic tool was the in-depth interview, in the
sense of a non-programmed standardized interview as defined by Vallés (Vallés, 1996), i.e.
with certain general guidelines that are adapted to the particular conditions of each specific
discussion. Most of the interviews lasted less than one hour and were recorded for further
analysis.
4.3. Design of the in-depth interview guidelines
The guidelines of the first part of the survey were based on a questionnaire sent to the
companies at an earlier stage of the research work. The results of the questionnaire allowed
the classification of the companies into companies that follow a well-structured model and
those which follow an ill-structured model in their decision making process. This classification
was used in the strategic selection of the study cases (companies) for the present survey.
The second part of the survey focused on questions about how the companies develop their
project management activities through standardized procedures. That is, how they structure
and systematize project management processes by measuring their level of maturity.
The interview covered the following aspects:
Detailed description of the selected decision case.
Description of the stakeholders indicating their responsibilities and roles in the
process.
Questions about the decision making process.
Questions about the company’s degree of orientation towards the project.
Questions about the status of the project management process. Mechanisms, project
plan, monitoring and control tools.
Questions about the stakeholders’ degree of involvement in the project.
371
Although during the interview, the respondent felt free to express his/her opinions and
viewpoints on the different issues, certain elements of the PM theoretical framework were
included. Key conceptual categories were included in the interview with the aim of identifying
the level of maturity (for example, when the respondents were asked to define “Project
Office”).
4.4. Strategic selection of the study cases
The selection of the survey sample followed the sampling criteria reported in Vallés (Vallés,
1996) about strategic selection of study cases based on the theoretical sampling approach
that consists of generating a typological matrix by crossing two criteria in order to identify
cases or situations within each resulting category.
The matrix that collects all the study cases (companies) distributed according to the decision
model - company size exes is as follows: the numbers correspond to the companies as listed
on the ARDAN Business Database of the IMPIVA (Valencian Institute of Small to Medium
Sized Industries).
Size
Well-Structured
Ill-Structured
Medium
3, 10, 15*, 20, 22*, 33, 44, 51,
59*, 63, 64*, 69*, 74, 79, 80, 82,
83, 87, 90*, 93, 99*, 100*, 105,
107
50*, 70, 73*, 95, 106, 113,
116, 118, 121
Small
2, 4, 5*, 6, 7, 8*, 9, 11*, 14, 16, 17,
18, 19, 23, 25, 26*, 29, 31*, 32, 34,
35*, 36, 37*, 39*, 40, 41, 42, 43,
46, 47*, 48, 49, 52, 53*, 54*, 55,
56, 57, 60*, 61, 62, 67*, 71, 75, 77,
81*, 85, 88, 91*, 97, 98, 103, 108,
109, 110*, 111, 112, 122*, 124,
127
1*, 12*, 13, 21, 24, 27*, 28,
30, 38*, 45, 58*, 65*, 66,
68, 72*, 76, 78, 84, 86, 89,
92, 94, 96, 101*, 102, 104,
114, 115, 117, 119, 120, 123,
125, 126, 128, 129
Table 2: Surveyed companies distributed in the typology matrix. Strategic selection of study cases:*surveyed
companies.
The survey was conducted on companies belonging to the four different categories as
defined in the matrix. The companies were distributed in the matrix cells according to their
answers to the questionnaire that had been sent at an earlier stage of the research and
whose data were used to classify the companies on the basis of the decision model used.
The survey of the four company groups was as follows:
Midsize companies with well-structured decision models: a total of 8 interviews were
conducted on project managers or heads of department who had actively participated
in project management.
Midsize companies with ill-structured decision models: the respondents belonged to
two companies of this group and had profiles similar to the respondents of the above
group.
Small companies with well-structured decision models: subjects with similar profiles
from 17 companies belonging to this group were interviewed.
372
Small companies with ill-structured decision models: the interviewed subjects
belonged to 8 different companies.
Although the strategic selection was determined by the cells in the matrix, the criterion of
non-uniformity within each group was taken into consideration for the selection of the
companies to interview in order to cover a wider range of companies. Thus, about 20% of the
companies that had answered the questionnaire were interviewed. In order to cover a wide
spectrum of possibilities, the selection was done on an organizational basis such as type of
business organization (familiar, managerial different from ownership, filial-Head), industrial
activities and geographical location.
On certain occasions it was not possible to personally interview the subjects in the selected
companies. For such cases, a qualitative questionnaire that included the main interview
issues was elaborated and sent by e-mail.

4.5. Survey data analysis and interpretation
After the interviews, the recorded material was processed and the interviews were typed and
further summarized in the final report of results.
The survey guidelines described in Section 4.3 above helped the analyst to better organize
(coding, classification, integration) the survey data.
The analytical processing of the data followed these four steps:
The transcripts of the interviews were analyzed, and the segments related to key
issues were underlined. Notes were taken on the sides to indicate the interview
section the fragment belonged to.
After coding each interview transcript, all fragments belonging to one section were
grouped together. This meant separating the fragments from their original interviews
and grouping them in the corresponding section.
Finally, the material collected in each section was re-classified and interpreted,
adding sub-sections when necessary.
After integration, section by section, the next step consisted of the coherent
organization of all sections following a narrative coherent sequence.
The analysis of the information collected in the interviews suggested that Valencian
companies have different levels of PM maturity mainly due to the great differences in their
organizational structure. About 45% of the surveyed companies are at low maturity levels
(levels 1 and 2) whereas about 45% are at high or very high maturity levels (levels 4 and 5),
although very few companies reach the Optimizing level. The rest of the companies (10%)
are at an intermediate maturity level.
However, the objective of the present work was not the accurate characterization of the
companies’ maturity level; for this goal the definition of indicators difficult to measure would
have been necessary. Rather the main objective of this study was to provide a qualitative
classification of the organizations in order to find differences in their project management
process. Two company groups were identified: companies with high or very high levels of PM
maturity (including the companies with an intermediate level of maturity as they were
considered to possess a well-defined structure in project management) and companies with
low or very low levels of maturity.
In the group of companies with well-structured decision models we found that:
These companies are project-oriented.
The Project and the Project chief manager are formally announced and designated.
373
They generally use Project management qualitative tools.
Project management processes are standardized.
The companies tend to continuous improvement in their Project management
processes.
The main conclusion drawn from the interviews and their transcripts was that there is a close
relationship between the way the organizations manage their PM decisions and their level of
PM maturity. The companies that use well-structured decision models are at high levels of
PM maturity, whereas the companies that use ill-structured decision models are at low levels
of PM maturity.
No clear relationship was found between maturity levels and company size as high and low
levels of PM maturity were found in medium as well as in small companies.

5. Conclusions
From the findings presented and analyzed in the previous section the following conclusions
can be drawn:
There is a close relationship between the company’s level of maturity in project
management and their decision process. The higher the level of maturity the more
they use well-structured decision models.
The statement that the higher the level of maturity the better the company performs in
modelling its project management decisions is based on the fact that higher maturity levels
bring about benefits to decision models. As an illustration of this we can mention:
The higher the level of maturity, the greater the availability of well-defined and welldocumented projects, which improves the PM decision process.
A better communication among the stakeholders involved in the project, resulting in a
higher satisfaction with and acceptance of the decisions made. This is partly due to
the use of a common language at high levels of maturity.
Alignment and coordination of goals across projects and with the company’s strategic
goals in aspects such as human resources, purchases, taskforce …
Definition of a framework for the decision making process and its consequences in
the project plan and project change management (this generally takes place at high
maturity levels).
Availability of databases about the company PM capabilities. This will facilitate
decision making management.
According to Project Management Process Maturity (PM)2 (Ibss and Kwak, 2002), at maturity
levels 4 and 5 there is an integrated planning and control of the company’s project portfolio,
and complex decisions, like project portfolio selection, are better solved.
Authors like O´Toole and Mikolaitis (O´Toole and Mikolaitis, 2002) identify maturity level 3
with the use of decision models implemented in the company and well-documented project
processes, and maturity levels 4 and 5 with assisted decision making and continuous
improvement in the decision making process.
The results of the analysis do not reveal any characteristic PM maturity level in
Valencian companies. Certain authors like Andersen and Jessen (Andersen and
Jessen, 2003) postulate that the level of maturity of a company is subjective rather
than objective. On the other hand, the percentage of Valencian companies at an
intermediate maturity level is significant. This consideration is valid only for
374
companies with high maturity practices. Ibbs and Kwak (Ibbs and Kwak, 2002) in the
application of (PM)2 consider that certain companies can be assigned a maturity level
N+1, yet they do not possess all the features of level N. This is due to the fact that
companies that lie at level N+1 are capable of selecting efficient practices and tools
for their projects. For this reason, level-3 companies are included in the group of
companies with high maturity levels.
Although we could not establish a typical maturity level in Valencian companies we have
drawn some conclusions that help define adequacy criteria for the future development of a
decision support tool for these companies:
In general, level-1 companies do not have a well-structured project plan; in addition,
they have no project management information system available. Level-2 companies
use support tools and practices for project planning as well as a well-structured
organizational project system. Companies at maturity level 3 define and manage
formal project management methodologies and models, and possess information
systems that collect, review and deliver well-documented processes for project
management. Companies at maturity level 4 have integrated or comprehensive welldefined project control processes that are coordinated by the different departments of
the company. The project management information system comprises all the
company’s areas and projects. In companies at maturity level 5 the project
management processes are planned and optimized with a focus on continuous
improvement.
High maturity levels are mainly observed in companies that manage projects in which
different organizations or departments are involved. These companies need a
common PM methodological framework that ensures project success; they also need
well-defined management processes to solve conflicts in organizations in which
projects are an essential part of the business (Mulder, 1997).
Certain companies classified as with high maturity levels, however, do not present a
well-defined project management training program, i.e. the participants in the project
may not have upgraded knowledge.
Most companies in the groups level 3 and higher, although they have project
management teams and departments, the team members also play other roles in the
company.
The conclusions presented here suggest that a decision support system adapted to the
organizational project management culture of Valencian companies should take into account
the following considerations:
Companies at maturity levels 3 and higher possess project management information
systems that allow them to count with valuable data and documents during the
decision making process. However, for companies with maturity levels 1 and 2, most
of which do not have such information systems available, the decision support tool
should help them in structuring and modelling project information.
For maturity levels 4 and 5, characteristic of companies that develop acrossorganizations projects, the decision support system should allow remote problem
solving and help solve conflicting interests among the stakeholders.
The decision support system should adapt to interdisciplinary problem solving and
help professionals with little knowledge on project management.
375
6. Future research lines
For future research work we propose to develop a decision support system for project
management decision making, based on Multicriteria Decision Analysis techniques, adapted
to the maturity level of the Valencian companies. The system may help in the modelling of
complex project management problems. In this sense, the support system will help
companies at low levels of maturity to structure decision problems.
References
Al-Sedairy, S.T., 1994. Project Management Practices in Public Sector Construction: Saudi
Arabia, Project Management Journal, pp. 37-44
Andersen, E.S. Jessen, S. A., 2003. Project maturity in organisations, International Journal of
Project Management 21 pp. 457-461
Boznak, R.G., 1988. Project Management- Today’s Solution For Complex Project
Engineering, IEEE.
Bu-Bushait, K.A., 1989. The Application of Project Management Techniques To Construction
and R&D Projects, Project Management Journal, pp. 17-22.
Cleland, D., 1993. The Future of Project Management, PM Network, pp. 6-8.
Deutsch, M.S., 1991. An Exploratory Analysis Relating the Software Project Management
Process to Project Success, IEEE Transactions on Engineering Management, Vol. 38, No. 4.
Donnelly, R.G. and Kezsbom, D.S., 1993. Overcoming the Responsibility-Authority Gap: An
Investigation of Effective Project Team Leadership for a New Decade, AACE Transactions.
Dooley, K., Subra, A., and Anderson, J., 1998. The Impact of Maturity and Best Practices in
New Product Development, Quality Management Conference, Arizona State University,
USA.
García Melón, M., Aragonés P., Poveda R., Zabala, J., 2005. Decision making in NPD
projects. An empirical study, International conference on engineering design, ICED 05,
Melbourne, Australia.
García Melón, M., Poveda, R., Aragonés, P., Pastor, J., 2006. Analysis of the decision
making processes in innovative companies. Empirical study for the Valencian Region
(Spain), 18th international conference on Multiple Criteria Decision Making, MCDM 2006,
Chania, Grecia.
Gobeli, D.H. and Larson E.W., 1986. Matrix Management: More than a Fad, Engineering
Management International, Vol. 4.
Gross, R.L. and Price D., 1990. Common Project Management Problems and How They Can
be Avoided Through The Use of Self Managing Teams, IEEE International Engineering
Management Conference.
Ibbs C.W., Kwak Y.H., 2002, Assessing project management maturity, Project Management
Journal, 31(1):32–43.
Kerzner H., 2000. Applied Project Management best practices on implementation. Ed. John
Wiley & Sons. USA.
376
Kwak, Clark, Grilo, Betts, and Ibbs., 1995. “Contemporary Strategic Planning Tools and
Applications for Construction Managers”, First International Conference on Construction
Project Management, Singapore, pp. 25-40.
Larson, E.W. and Gobeli, D.H., 1989, Significance of Project Management Structure on
Development Success, IEEE Transactions on Engineering Management, Vol. 36, No. 2.
Lundin, R.A., and Soderholm, A., 1994, Conceptualization and Learning- The Black
Environment of PM, PMI 25th Annual Seminar/Symposium”, Vancouver, Canada.
McCollum, J.K. and Sherman J.D., 1991. The Effects of Matrix Organization Size and
Number of Project Assignments on Performance, IEEE Transactions on Engineering
Management, Vol. 38, No. 1.
Might, R.J. and Fischer, W.A., 1985. The Role of Structural Factors in Determining Project
Management Success, IEEE Transactions on Engineering Management, Vol. EM-32, No. 2.
Mulder, L., 1997. The importance of a common project management method in the corporate
environment, Blackwell publishing: R&D Management, 27(3):189-196
O'Toole, M., 2002. Corporate Event Project Management, John Wiley & Sons. USA
Poveda Bautista R., García Melón, M., Aragonés B., 2005. Análisis de los procesos de
decisión en Dirección y Gestión de Proyectos. Estudio empírico en PYMES innovadoras de
la comunidad Valenciana. IX congreso internacional de ingeniería de proyectos, Málaga
Vallés, M.S., 1996, Técnicas cualitativas de investigación social. Reflexión metodológica y
práctica profesional. Síntesis. Madrid.
Ziomek, N. L., Meneghin, G. R., 1984. Training- A Key Element In Implementing Project
Management, Project Management Journal, pp. 76-83.
Rocío Poveda-Bautista
Departamento de Proyectos de Ingeniería
Camino de Vera, s/n
46022 Valencia
Phone: 96 387 98 60
E–mail: [email protected]
377
R&D&I PROJECT MANAGEMENT IN SPANISH UNIVERSITIES
Varas, M.
Universidad de Antofagasta
Caamaño, J.
Universidad del País Vasco
Abstract
The new role that universities possess in innovation, was promoted from an evolution both of
the universities themselves and of the social frame in which they are inserted Thus,
university research becomes one of the props of the development of society But, this new
role, carries new difficulties and organizational challenges to an institution already complex
(Bricall, 2000), (Sanyal & Bikas, 1995)]. In this process of adjustment universities have tested
several models of management, to give fulfillment to the expectations of society.
The intention of this paper is to understand how the R&D&i projects managed in spanish
universities. The analysis is carried out from a field study that shows that researchers, feel
alone and lacking of the necessary resources to manage the R&D&i projects
Keywords: Management R+D+i, Project management
1. Introduction
1.1. History of the changes in university
Before the democratic transition, Spanish universities had very few opportunities to
assimilate the Humboltiano model, based on the narrow link between teaching and research
that had spread over Europe and The United States throughout the 19th century. According
to (Sanz, 2003), the "transformations in the European and American universities scarcely
reached the Spanish universities. Only after Franco's regime, they approached the
necessary institutional reforms that helped to define, in the context of an extraordinary
autonomy and capacity of self-government of university, the possibilities of normal
development of the research and of reinforcement of the bows with the productive
environment".
In the seventies other opinions arose about the problems that the R&D systems that had
been created had and a worry developed about the difficulties in the utilization of the
knowledge produced by universities. Thus, the transfer of technology and the connection
with industry emerged as essential topics for the government, together with the later
378
incorporation of technological innovation. Simultaneously, from the theories and policies of
regional development (Malecki, 1991), it pointed out that universities, and their scientific
technical capacities were a potential mechanism, a lever for regional development.
Therefore, to both traditional missions of university, consolidated in the binomial teaching
and research throughout the 20th century, a new mission was added in the last years: the
transfer and exploitation of research results. In line with these observations new
argumentations have been constructed about the radical changes that this new orientation
produces in the nature of the way of knowledge production (Gibbons et al, 1994); it says that
a new mode (Mode 2) characterized by the specificity, the trans- disciplinarity, has arisen. In
addition the relations among university, government and company have changed with the
emergence of a "triple helix " [(Etzkowitz & Leydersdorff,1997), a "new" type of institutional
arrangement for the production and utilization of knowledge.
Thus, the evolution of university begins with efforts carried out by the state to promote
research in general and, specifically, to improve the research level of Spanish university. In
any case, the major weight of university in the Spanish research activity is also due to
significant growth of the expense in high education that takes place in the eighties and first
part of the nineties (OECD, 2002).Another factor that has influence on this evolution of
Spanish university towards research has been the availability of public competitive funding of
research activities, so that the normal development of R&D projects was allowed.It seems
relevant, to indicate that the project policy understood as a general promotion policy of
research activities by means of public summons, had its introduction in the eighties, and it
meant a radical innovation in the Spanish science and technology system. Also thanks to this
policy, and to the creation of the National Agency of Evaluation and Futurology (ANEF), one
of the institutional structures in which the above mentioned policy was shaped, a healthy
culture of competitive summons has been developed in Spain.
On the other hand, the normative reforms carried out in the university in 1983, with the
approval of the Law of University Reform (LRU), aimed at a model's change in which
research and service to society, by means of the transfer of technology, carrie to represent
important values added to the traditional one of training by means of teaching (Jimenez
&Moya, 2003).
The LRU raised a program of university opening to society and companies, besides a
conceptual integration of research in its activities.
The article 11, which authorized the contracting of scientific works out of the university,
created specific incentives to university teachers for the development of research under
contract with companies and public administrations This article, now replaced with 83 of LOU
(6/2001), established a special regimen for university government employees, in which these
could increase their income by means of the establishment of " research contracts " or the
rendering of services to Public Administrations or private entities. Therefore, individual
incentives, supported by improvement of teachers' fee, were created for the development of
R&D under contract.
Later on, this system was complemented by the approval of "researchers' sexenios" that
were established remunerations for researchers dedicated to the Basic R&D, which have
served to stimulate research commitment and scientific production (Jiménez Contreras et al.,
2003), simultaneously they have turned into elements that demonstrate the researchers is
379
reputation, in spite of not being so lucrative in remunerative terms as the derived from the
contracting with companies.
In the same way, LRU introduced, and LOU confirmed, the possibility that public universities
could formalize agreements and contracts with the private sector to finance the
accomplishment of research and development activities. This one was the first step that put
the base for a culture of collaboration that is developing progressively, in the year 2002 a
19,63 % of the budget expenditure in research was financed by the private sector.
Nevertheless, the research in Spanish university has been financed mainly by Public
Administrations.
This progressive adaptation of university was made through different organizational schemes
and even with new legal structures to manage these services. In this sense, Spanish
university has created foundations in charge of continuous formation, has promoted
independent legal institutions to favour the connection between university and company, has
set up companies to manage some of its services, etc. To sum up, it has tested which is the
best way of organization to give each type of service. Some recent analyses of the ways to
carry out research and continuous formation in universities offer many examples of the
enormous variety of organizational solutions that have been given in the performance of the
different university services.
1.2. The services of research management in universities
The services of research management in universities are created from the Sciencie Law, in
1986, which entrusts functions of contracting (Article 17), creating or taking part in mercantile
societies (article 19). For this reason, structures dedicated to research management, with
different functions and with different organic and functional dependence are created.
In addition, the administrative services from universities have among their functions the
support, advice, management and rendering of necessary services for the fulfillment of the
university purposes. The article 39 of LOU, ratifies the exposed in the LRU, and establishes
"research as a university function".
Therefore, the services of research management are in charge of facilitating the researcher
with necessary specific support, with rapidity and efficiency, for the correct execution of
research. According to the article 40 of LOU, the researcher has the right to have the
information, advice and necessary administrative assistance to carry out research, since
research is a right and duty of university professors.
2. Research Methodology
2.1. Selection of a sample
To select the public universities that took a part in the study they considered the income that
they obtained from their research activities, including public and private financing derived
from research projects. The criterion of selection was all those universities that possess a
incomp bigger or equal to 10.000.000 Euros, qualifying 20 universities, representing 42 % of
the total population of public universities.
380
2.2. Evaluation of Vice President for research
To know the opinion of Vice Presidents about research management, Vice Presidents /
chiefs of research were consulted about the processes that they carried out to manage
projects.
The sample target of study is formed by:
1-Vice Presidents or chiefs of research from 20 selected public universities.
Population universe
20 public universities
Perimeter of research
Spain
Modes of data collection
Personel mail intercept survey and personel in
home survey.
Date of accomplishment
Begin: 15 de March 2007
End: 15 de September 2007
Sample size
20 Vice President research
Rate of response
55 % (11 Vice President Research )
Table 1: Specification sheet of the survey Vice President Research public universities
2.3. Questionnaire Vice President Research
The development of the questionnaire, was based fundamentally in checking the procedures
of delivery of information, follow-up, control and evaluation of research projects, present on a
R&D&i project cycle of life a, that Vice President Research carry out.
Initial Phase
In this phase, it is necessary to manage and promote the participation of university
researchers in projects summons and the collaboration with other organizations in the field
research.
The carried out activities are:
1. Dissemination of public summons, information about requirements, conditions and period.
2. Advice, taking into account the needs of every group of research.
3. Support on interested companies search.
4. Collaboration in contracts negotiation.
Execution Phase
This phase begins with the project authorization to researcher, or with the signing of the
research contract (in case of contracts art. 83 L.O.U.).
In this phase,the different regulations that are going to mark in any way the economical
administrative management of a research can't be forgotten.
Tasks of the research management service:
381
1. Information to the researcher of the conditions of the project authorization (initial date,
admitted expense, possible modifications, date of ending, justification and reports). That is to
sag, the requirements established by sponsor.
2.-. System of budgetary countable follow-up of expense and income derived from projects.
3. Exact Information to the researcher on the economic situation of research project.
4. Information on system of assistant personal and scholars selection.
5. Control of times of dedication and remuneration of PDI (researcher teaching staff),
involved in the aforementioned project.
6. Juridical advice.
7.-Reports elaborated by the internal Control Unit / intervention in projects. Does transfer of
funds exist among projects?
Closing Phase
Tasks of the research management service:
1. Justification to the different sponsors.
2. Justification in case of audits and controls.
3. Elaboration of Databases.
4. System of evaluation of project results and its effective application, as well as the internal
procedure of intellectual property and exploitation results.
5. Claims presented by sponsors of the different projects and if it was the case the financial
repercussion of them and the person or unit that takes the expense.
A consultation that had relation with the opinion that Research Vice President had of
researchers fulfillment (periods, costs) was added.
1-Does periods and requirements fulfillment of the established documentation, exist on the
side of the researcher?
2.4. Analysis of research groups
The research groups that took part in this study were selected. For that purpose, a size of
sample of 20 % of the stratum was designed, which means 956 research groups. A designed
sample was chosen since by a previous study the reticence that researchers had to answer
this type of instruments was known. The minimal sample limit marked by statistic was of 355
show units, for a confidence of 95 % and a mistake of 5 %, but a major sample was gave
guarantee of the validity of results.
In relation to the number of research groups that are inquired by university, the same
percentage that represents every university inside the universe was applied to the size of the
sample.
The selection of show units was carried out by means of simple random sampling. According
to the table of random numbers.
The specification table of the survey of research groups from public universities is shown
Table 2.
382
Population universe
Research groups 20 public university.
Perimeter of research
Spain
Modes of data collection
Personel mail intercept survey and personel in
home survey.
Date of accomplishment
Begin: 1 de February de 2007
End: 15 de May de 2007
Sample size n
956 (Designed sample)
Rate of response
46% (university = 407,Centers and institutes =
36, Total
corrected sample 443)
1,9 % sample be not valid
Mistake
3,7 %
Confidence level
95% , P = Q
Types of sampling
Sampling stratified with election of sample
units by means of random simple system
distributed in proportional form by university.
Table 2: Specification sheet of the public universities survey.
2.5. Questionnaire of research groups
It aims to know the opinion of the interviewed person about the processes of project
management that Vice President Research carry out. These questions that then must be
contrasted with the answers of Vice President Research organisms, were designed in the
following way:
Aim: To evaluate the effectiveness of the research support services the university that.
Management service of research
Initial Phase
1. Dissemination of public calls, information on requirements, conditions and terms.
2. Advice, taking into account of every research group.
3. Support on search of interested companies.
4. Collaboration in contracts negotiation.
Execution Phase
5- Information to researcher on economic situation of Project research.
6- Information on system of assistant staff and scholarship holders selection.
7- Legal advising.
8- Support the use of computer science applications and electronic communication channels.
383
Closing Phase
9- Internal rules to regulate intellectual property and explotation results.
10- Claims presented by sponsors of different projects.
Researcher Management
11- Do you fulfil the targets and term stablished in the contract?
12- Do you deliver partial and final informs on time?
13- Do you fulfil the laws and statutes that affect research?
3. Results
The results are compared, following the order Project Life Cycle Tabla 3. The analysis of the
data picked up from vicepresidents in the starting phase indicates that:
o The dissemination process is evaluated as effective by all the universities, opinion
that share researcher evaluating their average with 4,78.
o About the process of individual advising to research groups, the Vice Presidents
indicate that it is only carried out when it is asked for or “sometimes” and researchers
indicate in a 42.5% that the support is regular and 35.8% values it as bad.
Average
Average
Vice
President
Frequency of the
Process
Researcher
Evaluation of the
process
Dissemination
4,0
Always
4,78
Good
Needs
2,91
3,47
Regular -Bad
Contracts negotiation
3,09
3,23
Regular
Search of companies
2,45
2,91
Bad
Personal Information
3,55
Always
3,71
Regular -Good
European Projects
3,27
Always
3,13
Regular -Bad
Computer
application
2,55
3,16
Regular -Bad
Regular -Bad
sciencie
When it is asked for
Sometimes
Legal Advising
3,64
Always
3,20
Tracking Information
2,64
Always
**
Economic Tracking
2,82
Always
4,17
Results projects
1,64
Never
**
Registry
nonconformities
1,82
Never
2,58
Good
I do not know
Table 3: Average of the processes of projects management in public Universities.
** Processes nonconsulted to research groups
384
o
Related to the support in the search of companies, Vice Presidents Research indicate
that it is only carried out when it is asked for; however the activity was valued as bad
by a 41% of researchers. It was analyzed, in addition, if there was difference of
opinion with respect to the search of companies process by a researchers it existed,
following the amount of budget that the group managed, Table 4.The results indicated
that those groups that managed a major budget estimated on 300,000 Euros, had
equal or worse opinion that those than handled a lower budget.
Search of companies to research groups
Annual
R&D&i
Total
It does
not
exist
I do not
know
Bad
Regular
Good
5
10
12
4
3
34
6001-300.000
euros
38
55
138
64
20
315
300.001-600.000
euros
4
6
12
9
3
34
600.0011.500.000 euros
2
5
2
4
0
13
1.500.00130.000.000
euros
0
1
3
1
1
6
Sobre
30.000.000
1
1
0
2
1
5
50
78
167
84
28
407
Budget 0-6000 euros
Total
Table of contingency 4: Annual budget R+D+i. Search companies groups public universities.
o Referring to the collaboration in the contract negotiation a 41.7% of the Vice
Presidents indicate that the activity carried out is, nevertheless researchers have a
great dispersion of opinions, 27.5% consider it regular, 19.2% of them consider it
good and 19.4% consider it bad. The Table of contingency 5 was carried out, to know
if the contract processes that were well evaluated by researchers corresponded to
contracts with companies.
The analysis of data indicated that only 14% of the research groups that worked with
companies evaluated the management well since the other 5.2% corresponded to the
opinion of the research groups that managed public projects. Demonstrating this way
that the process of contract support with companies is evaluated of regular to bad by
researchers.
385
Negotiation of the contract for the group
Budget finance 0%
by companies
Between 1%
and 20%
Between 21%
and el 50%
More of 50%
I do not know
Total
It does
not
exist
I do not
know
Bad
Regular
Good
23
80
29
33
21
186
4
15
30
39
32
120
5
3
13
25
12
58
5
2
5
11
11
34
1
0
2
4
2
9
38
100
79
112
78
407
Total
Table of contingency 5: Budget financed by companies * Negotiation contract group. Public universities.
In the phase of execution of the project they carried out six consultations whose answers are
given next:
With respect to the support to the researcher for the selection of staff and
scholarship holders, Vice President Research in a 66.7% of the sample, indicated
that it was an activity that carried out was. Researchers, however almost 65.1% as
evaluate it a service of regular to good.
o
o
As far as the legal advising on intellectual property and results of explotation is
concerned, a 66.7% of Vice President Research indicate that it is an activity that is
always carried out opposed to a 44.2% of research groups that consider the activity
of regular to good showing a clear disagreement with the process.
o
In relation to the support in the management of European projects a 50% of
the sample of Vice President Research indicate that they carry out realise the activity.
The rest of universities execute it only, when it is asked for, nevertheless, a 53.6% of
researchers think that the service is only of regular to bad evaluating in only a 12.8%
as good.
o
As far as the support to researcher in the use of electronic communication
channels to postulate to projects is concerned a great dispersion of answers exists
among universities, Table 6, where it is shown that a part of universities only carry out
this process when it is asked for.
The opinion of research groups is that support is evaluated as regular or bad by a
58.7% of the sample.
386
Truthful
Frequency
Percentage
Truthful
Percentage
Accumulated
Percentage
Never
2
18,2
18,2
18,2
When it is asked
for
4
36,3
36,3
54,5
Sometimes
2
18,2
18,2
72,7
Always
3
27,3
27,3
100,0
11
100,0
Total
Table 6: Computer science applications. Public universities.
o
In relation to consultations regarding the sending to researcher in a permanent
way of the economic situation of the project, a 91% of Vice President Research
indicate that an economic tracking of projects exists and 81.8% (only) reveal that
information about tracking project is carried out, being only economic tracking. As far
as research groups is concerned 47.2% of research groups consider good the
economic tracking and 33.2% as regular.
In the closing phase the two processes consulted to Vice President Research are not carried
out, being the system of evaluation of the project results the least used, as it is shown on
Table 7 of frequencies. In relation, to the processes of nonconformities registry research
groups do not receive information on the complaints made by sponsors, therefore, a 50.6%
do not know if this information exists against a 7.9% who consider it as good.
Results
projects
Registry
nonconformities
No
Sometimes
Yes
Frequency
7
1
3
Percentage
63,6
9,1
27,3
Frequency
6
1
4
Percentage
54,5
9,1
36,4
Table of Frequencies 7: Vice President Research public universities.
The analysis of these results in the context of the changes that Spanish universities live
make evident that weaknesses and problems exist in the model of management of the
centralized research that universities have adopted.
The evaluation of Vice President Research management lets us know that the organism
owns a better opinion of the services that it gives than the one researchers do. But also the
belief exists that they must improve support services, lack of personnel specialization,
rotation and deficiency of resources are factors that Vice President Research indicate as
variables that affect the projects management that they carry out.
387
4. Conclusions
The valuation of service management of Vice President Research from public universities
that researchers carry out “is bad”, they feel alone and devoid of the necessary resources to
manage projects. Vice President Research however do not value their management bad and
are conscious of a series of problems which must be considered nowadays as:
o
Deficiency of human resources described in relation to the activity to develop.
o
Lack of budgets or lack of economic resources to manage research.
o
Exponential increase in the number of public calls in projects in the last years.
o
The demands for the control and audit of aid.
o
Little clear norms of sponsoring organisms as far as eligible expense and term
(in some autonomic funds) is concerned.
o
The general norms of Public Administration, sometimes too rigid, that makes
difficult the research management.
References
Bricall, J.M.; 2000. Informe Universidades 2000. Conferencia de Rectores de las
Universidades
Españolas.
Recuperado
el
20
octubre
2005,
de
http://www.crue.org/presentacion.pdf
Delgado López-Cozar , E.; Jimenez Contreras, E.; de Moya Anegón, F., 2003,.The evolution
of research activity in Spain: the impact of the National Comisión for the Evaluation of
Research Activity (CNEAI), Research Policy vol. 32, 123-142
Etzkowitz, H. and Leydersdorff, L. 1997. Universities and the Global Knowledge Economy:
Atriple Helix of University-Industry-Government Relations. London: Pinter
Gibbons, M., Limoges, C., Nowotny, H., Schwartzman, S., Scott, P. and Trow, M. 1994. The
newproduction of knowledge. London: Sage
Malecki, E.J. 1991. The dynamics of Local, Regional and National change. Technology and
Economic Development. Essex: Longman.
Sanyal, Bikas C. 1995. Innovations in University Management, Paris: UNESCO. Publishimg:
IIPE. International Institute for Educational Planning
Sanz Menéndez, L. 2003 La investigación en la universidad española: la financiación
competitiva de la investigación, con especial referencia a las Ciencias Sociales y
Económicas Publicado en Revista Española de Investigaciones Sociológicas, 109: 181-218,
Enero-Marzo 2005.
OECD 2002. Science, Technology and Industry Outlook 2002. Paris: OECD.
Sánchez Ferrer, L. 1996 Políticas de reforma universitaria en España; 1983-1993. Tesis
doctoral. CEACS-Instituto Juan March, Madrid, España
388
Dra. Marinka Varas Parra
Ingeniería Industrial. Facultad de Ingeniería
Universidad de Antofagasta
Av. Jaime Guzmán E. s/n
Antofagasta. Chile
389
CONSTRUCTION RISK MANAGEMENT: APPLICATION AND
EDUCATION
El-Dash, K.
College of Technological Studies, Kuwait
Abstract
Construction risk management emphasizes the application of risk management in the
construction industry, focusing on large scale and innovative projects. The application of risk
management in construction projects is still less than its true value. The present study
compares the level of the needs of the construction industry and the current level of acquired
risk management knowledge and skills. The study demonstrates the response of many
industrial foundations in Kuwait that contribute to construction projects, focusing on
infrastructure projects. The responses were used to deduce the major elements required to
implement construction risk management in the market. Moreover, the current levels of
knowledge and skills of those elements were assessed, also based on the collected data.
The requirements to narrow the gap between needs and actual education level are
discussed. Finally, factor analysis methodology was used to categorize and evaluate the
parameters that affect the implementation of risk management in the construction industry in
Kuwait.
Keywords: Construction, risk, education, management, Kuwait
1. Introduction
Construction management education has been of interest for more than forty years. Most
university civil engineering departments offer fundamental construction management courses
for undergraduate students.Some departments offer specialized courses, such as contracts,
specifications, equipment, scheduling, and/or computer applications. Meanwhile, the subject
of construction risk management has received little attention in undergraduate curricula. On
the other hand, more universities offer graduate courses that focus on risk management in
the construction industry (Banik, 2004; Niewoehner, 2005).
Risk management is a complicated process that interrelates with many other processes in
the construction industry and on construction projects (Alexander, 1998; Chapman, 1997;
Grey, 1995; Holland, 2006). Investigating project risks includes studying potential events that
may affect the scope, cost, time, or performance of the project's objectives. Investigating
potential risks requires the collaboration of all disciplines contributing to the project.
Technical, managerial, financial, and administrative departments of the participating firms
390
need to cooperate to identify and respond to expected risk events. This integrative process
needs practical experience to adapt the required environment (Edwards, 1995; Sherer, 2004;
Smith, 1999). This is the main reason for the reduced risk management content taught to
undergraduate students.
The target of the present study is to compare the current level of knowledge and skills with
that required to implement risk management in the construction industry based on the
opinions of construction practitioners in Kuwait. The study considered the following steps:
• identifying the knowledge required to conduct the process of risk management in the
construction industry and on construction projects;
•
assessing levels of awareness and implementation of risk management;
• assessing the level of risk management knowledge required to improve its
implementation;
• comparing the required knowledge to the current status of implementing risk
management (comparing the previous two items); and
• determining the major parameters that contribute to progress in implementing risk
management in construction projects and the construction industry.
2. Method of study
A simple model is proposed for the parameters required to implement the risk management
process. The model was prepared based on preliminary discussions with construction
practitioners and academics. A questionnaire was developed based on preliminary
discussions and the proposed model. Participants with more than 15 years of experience in
construction projects and currently working on mega projects were selected to ensure a good
understanding of risk management requirements. The items of the model were discussed
with 52 participants to clarify any misunderstandings regarding the issues under
investigation. The sample included 32 participants from the government and 20 from the
private sector.
The questionnaire included two categories of questions:
1. The current level of awareness and implementation of risk management concepts and
tools in construction projects; and
2. The target level of awareness and implementation.
The questions included evaluations of the required level of knowledge and skills for junior
engineers with up to five years of experience and senior engineers with more than five years
of experience for each of the previous two categories. The levels found for junior engineers
show a trend that satisfaction is strongly related to undergraduate education level. On the
other hand, senior engineers' satisfaction with implementation is strongly correlated to
postgraduate education, training programs, and interactive experience.
3. Risk management processes
The risk management requirements model presented in Figure (1) illustrates the main
391
requirements to implement the process of risk management. The risk management process
is divided into four subprocesses, which are identification, analysis, response, and review.
3.1. Risk planning and identification
Identification of potential risks is usually the first practical process and the most crucial one in
the risk management process. The risk identification process usually is a part of the project
plan and is based on the collection of institutional policies and regulations that place many
constraints on the process of risk management, such as budget contingencies, time delay
thresholds, human resources rules of hiring and firing, rewards and penalties, performance
control limits, reporting systems, and procurement policies.
A risk management environment is subject to the support of higher authorities in the project,
the organization, and the construction industry. The support of governmental, educational,
and industrial institutions helps to improve the awareness of risk management. This
awareness leads to a better collaborative environment that improves risk planning and the
identification process.
Risk identification depends on many other parameters, such as project management
knowledge and skills, and technical, finance, accounting, economics, and legal capacities.
Analysis
SWOT
Monte
Carlo
Grids
Insurance
Risk
indices
Statistics
EMV
Training
Executives
support
Reliable
management
Sponsor
support
Planning &
Government
support
PM
Finance
Accounting
Law
Economics
Response
Creativity
Organizational
Policies
Management
Capacity
Auditing
Review
Identification
Figure 1: Risk management requirements.
3.2. Risk Analysis
Risk analysis usually includes qualitative and quantitative analyses. Qualitative analysis is
concerned with determining the prioritization of the expected risk events based on the
anticipated possibility and impact of each. Quantitative analysis is concerned with numerical
determination of the consequences of these risk events.
392
Many tools may be used in the qualitative and quantitative analysis, such as SWOT analyses
(strengths, weaknesses, opportunities, threats), risk indices, grid analyses, Monte Carlo
simulations, EMV analyses (expected monetary value), and statistical calculations.
3.3. Risk Response
The risk response process includes planning for the required action to be considered in case
a risk event occurs. It also includes taking planned action if required and following up with the
consequences of these actions to ensure that a risk plan results in the required outcome.
Risk response is determined by proposing several alternatives to eliminate or mitigate an
expected risk and assigning the optimum alternative as a response. Most risk responses are
based on contractual agreements, insurance, and managerial solutions.
3.4. Reviews
Knowledge
Risk
Management
Decision making
Creativity
Value
Engineering
Economics
Scope
Management
Communication
Management
Contract
Management
HR
Management
Insurance
Law
Finance
Accounting
Quality
Statistics
Management
Time
Cost
Management
Management
Experience
Figure 2: Risk management body of knowledge.
Reviewing the risk plan is a continuous process that starts with initial planning and continues
into identification, analysis, and response processes. However, the review of the overall risk
management lifecycle is fundamental to the subsequent step of whether to change the risk
plan or to plan for another risk event in the same or different project. Auditing is the usual
form of review at the end of the risk management lifecycle, while other forms like checklists
could be used at particular times during the process.
4. Risk management body of knowledge
A construction project risk management body of knowledge integrates with that of the
management of other projects. Figure (2) shows schematically the constituents of a body of
393
knowledge. These items are distributed in the figure according to proposed requirements of
knowledge and experience. The proposed distribution is configured specifically for
construction engineers. These constituents can be rearranged with respect to level of
education as follows:
4.1. Undergraduate level
a.
Time management
b.
Cost management
c.
Quality management
d.
Scope management
e.
Statistics
4.2. Postgraduate level (including professional training and practice)
a.
Human resources management
b.
Communication management
c.
Contract management
d.
Finance
e.
Accounting
f.
Law
g.
Economics
h.
Value engineering.
5. Results and analysis
The data collected from the questionnaire was analyzed using statistical descriptive analysis
to compare the results for each parameter and then analyzed using the factor analysis
method to facilitate categorization of the considered parameters.
5.1. Statistical descriptive analysis
The data collected considered the requirements for risk management in terms of level of
education and skill required for application. The questionnaire developed included three
categories for each item as follows (Table 1):
a. The current implementation level, which represents the awareness level of the
specified item;
b. The required level for a junior engineer involved in risk management, which
represents the education and skills of freshly graduated engineers; and
c. The required level for a senior engineer involved in risk management, which
represents the education, knowledge, and skills of engineers with more than five years of
experience in the construction field.
394
Each item was evaluated using a scale from one (lowest level) to five (highest level). The
number of choices from the participating 52 practitioners was multiplied by the chosen rank
(from one to five) and the sum divided by 10. The calculation process could be presented in
the mathematical form as follows:
WS =
1 * n1 + 2 * n 2 + 3 * n3 + 4 * n 4 + 5 * n5
10
(1)
Where WS is the weighed sum for the specified item and ni is number of participants who
voted (i) for this item. The results of these calculations are presented in Figures (3)–(6) in the
form of a comparison among the weighed sums for the item's implementation level, required
level for junior engineers, and required level for senior engineers. The data collected were
categorized in the following four categories.
5.1.1. Risk management awareness
The first part of the collected data considered the general awareness of construction
management and construction risk management specifically. The results, shown in Figure
(3), indicate that the current level of knowledge is reasonable with respect to junior
engineers. The allocated required level for freshly graduated engineers is less than the
general current implementation level. On the other hand, the required level of senior
engineers is higher than the implementation level. All categories show the same trend for the
senior engineers, especially when it comes to strategic risk management, program risk
management, and project risk management.
5.1.2. Risk analysis tools
Qualitative and quantitative risk analyses include many parameters that could be utilized.
The items shown in Table (1) and Figure (4) were chosen because they are common tools in
construction project risk management. Figure (4) shows the gap between the actual practice
in using these tools and the required level of education and skills for the tools considered.
Practitioners pointed out that CPM is the most necessary tool to be utilized in construction
projects. Statistical calculations and EMV follow CPM in implementation importance. Figure
(4) also shows that the difference between the required level of knowledge for junior
engineers and the required level of knowledge for senior engineers is marginal. This slight
difference could be attributed to the fact that the participants in the study were senior
engineers and they considered that these technical processes are usually carried out by
junior engineers while the seniors only review the processes. The other tools considered in
the study (decision trees, grids, risk indices, and Monte Carlo simulation) received slightly
less interest from the participating practitioners. The results in this category demonstrate the
need to improve the level of the application of risk analysis tools in undergraduate courses
and the corresponding practical training during study or during the early periods of junior
engineers' construction careers.
395
Implementation Junior need Seniors need
Construction management awareness
CM experience
18.6
12.6
24.5
RM experience
15.3
9.2
18.9
Strategic Risk management
10.9
6.7
19.7
Program Risk Management
11.2
8.6
19.0
Project Risk management
13.3
12.6
20.3
EMV
10.4
17.6
19.3
Statistical sums
12.0
17.9
21.6
Monte Carlo
8.8
14.8
13.9
Decision Trees
9.6
16.2
17.3
Risk indices
9.1
15.5
15.1
Grids
12.5
17.2
17.9
CPM
18.4
21.6
24.3
Insurance
10.1
9.7
18.0
Policies
13.7
13.0
19.3
Management
13.9
15.2
22.5
Auditing
14.1
12.5
17.4
PM experience
16.5
9.2
24.7
Finance
13.8
11.5
20.4
Accounting
13.7
10.6
18.2
Equipment management
15.1
11.9
18.6
Law
Economics
13.7
9.9
18.1
12.9
13.8
18.9
Risk analysis tools
Risk response capacity
Risk management knowledge
Table 1.Risk management items considered in the questionnaire and the corresponding weigh sum
(WS).
396
Figure 3.Evaluation of construction risk awareness level versus requirements.
Figure 4.Evaluation of risk analysis tools implementation versus requirements.
5.1.3. Risk response capacity
Risk response is a cyclical process that includes planning, execution, and control. This
process requires a high level of managerial knowledge and skills. These requirements are
evaluated quantitatively in the results of the collected data. Figure (5) shows the results of
this partial study. The most necessary knowledge area is found to be the capacity to manage
project risks by utilizing acquired knowledge and skills. Next to management come
organizational policies, such as distribution of roles and responsibilities, risk management
strategy, and risk threshold. The results show that the required level of knowledge of
insurance is slightly below that of organizational policies and slightly higher than that of
auditing processes. The insurance culture is still weak in the construction field. The
manipulation coverage of insurance in construction is usually limited to physical losses like
accidents, injuries, fatalities, and construction damage. Project managers and sponsors
seldom exploit insurance for other managerial risks like liquidation damages, cost overruns,
and schedule delays. The assignment of appropriate premium and determination of the
conditions required for the insurance agreement vary from one project to another depending
on the complexity, managerial capacity, experience, and stakes of the project. These are the
major barriers to expanding insurance to cover similar risks. Establishing a systematic
approach to calculating an insurance premium and insurance contract clauses could be
performed in a well-documented construction environment but not in developing countries.
397
Figure 5: Evaluation of risk response capacities implementation versus requirements.
Figure 6: Evaluation of risk knowledge areas implementation versus requirements.
5.1.4.
Risk management knowledge
Construction risk management body of knowledge includes a diversity of knowledge areas
(Figure 2) that are difficult for an individual engineer to accumulate. Hence, risk management is
a cooperative process that must be carried out by a management team that acquires the
required knowledge. The study considered the major required branches of knowledge, including
project management, finance, economics, accounting, equipment management, and law.
Figure (6) illustrates the collected data for the implemented and required levels of risk
management body of knowledge.
As can be seen in Figure (6), project management knowledge and skills are currently the most
commonly implemented knowledge areas, followed by equipment management. The other
knowledge areas, considerably behind the previous two areas, are at a similar level. The
importance placed on equipment management could be attributed to participants' background
as senior engineers. The results (Figure 6) show that the required knowledge level for project
398
management is most necessary for senior engineers. The engineering practitioners affirmed
that finance is the second most important area and economics is the third. The lack of
importance reported for law reflects low awareness of the very strong relation between
construction risk on one side and contract clauses, construction changes, and dispute
resolution on the other. The gap between the actual, practical level and the requirements for
senior engineers is significant for all areas considered, especially project management. The
required improvement in project management is reported mainly as risk management and
contract management capabilities, which was concluded from individual notes and discussions
with the participants. The self-assessment of the senior practitioners in this regard is greatly
appreciated. It reflects awareness of the great need for more effective acquisition and
implementation of management knowledge.
The required level of knowledge for the junior engineers was satisfying, as shown in Figure (6).
However, participant practitioners noted that acquiring higher levels of knowledge and skills by
juniors is usually attained through practical implementation. These results direct the strategies
of the construction industry in developing countries toward improving the capabilities of the top
and middle management personnel to attain the required competition through long training
courses by international experts, gaining experience from benchmark firms, or intermingling
with well-established firms in local projects.
25.0
WS
20.0
15.0
Governm ent
Private
10.0
5.0
Motivation
Reward
Change
Funds
Management
Security
Positions
Policies
Training
Quality
Innovation
RAM
Standards
Culture
Sponsors
0.0
Figure 7: Positive contribution to risk management processes.
5.2. Factor analysis
The results of the factor analysis show three main factors that contribute to the variability of the
data collected. These three factors and the corresponding parameters are discussed below in
sequence that accord to the variability encountered in the factor analysis process. Figure (7)
presents the weighed sums (WS) for these parameters. The responses in this regard were on a
scale from one to five, where a score of five meant the implementation of the item was the
highest positive contribution. It could be noticed from the figure that the financial aspects and
security are the main contributions in the governmental sector. On the other hand,
Standardization and the support from senior management are the main contribution in the
399
private sector. The three parameters encountered by the deduced factors are introduced in the
following discussion:
5.2.1. Culture of the organization
The first factor considers the parameters correlated to organizational culture. Table (2)
introduces theses parameters and the correlations among these parameters. All the
correlations have positive values which points to the strong correlations among these
parameters. The highest value is between Standardization of the organization and
implementing quality standards. The relationship between quality and standardization is
obvious and convincing.
Sponsors
Culture
Standards
RAM
Innovation
Quality
Sponsors
1.000
0.589
0.424
0.333
0.143
0.530
Culture
0.589
1.000
0.699
0.387
0.549
0.407
Standards
0.424
0.699
1.000
0.593
0.475
0.804
RAM
0.333
0.387
0.593
1.000
0.209
0.741
Innovation
0.143
0.549
0.475
0.209
1.000
0.486
Quality
0.530
0.407
0.804
0.741
0.486
1.000
Table 2: Correlation matrix for parameters considered by the first factor.
1. The contribution of sponsors.
2. The culture of the construction organizations. Standardization of the organization.
3. Responsibility assignment matrix.
4. Encouraging innovative attitude.
5. Implementing quality standards.
5.2.2.
Motivation and development
The second factor considers the parameters correlated to the motivation and development of
the personnel who are supposed to participate in risk management processes. Table (3) shows
the correlations among these parameters. The correlation values are even higher than those
obtained for the first factor. The highest correlation is between sufficiency of training programs
and support from direct and senior management staff. Both parameters show that the
participants in need for more technical and managerial support to enhance their performance.
1. Sufficiency of training programs.
2. Encouraging organizational policies.
3. Availability of appropriate positions.
4. Adequate job security.
400
5. Support from direct and senior management staff.
Training Policies Positions Security Management
Training
1.000
0.609
0.572
0.459
0.763
Policies
0.609
1.000
0.641
0.375
0.591
Positions
0.572
0.641
1.000
0.483
0.303
Security
0.459
0.375
0.483
1.000
0.670
Management
0.763
0.591
0.303
0.670
1.000
Table 3: Correlation matrix for parameters considered by the second factor.
5.2.3.
Risk management retention.
Table (4) shows the correlations among these parameters. The correlation values in this set of
parameters are less than the previous two sets. The highest correlation is between rigidity of
change rejection environment and sufficient reward system. There is no distinguished
relationship between these two parameters and the other two parameters.
Funds
Change
Reward
Motivation
Funds
1.000
0.197
0.316
0.390
Change
0.197
1.000
0.406
0.223
Reward
0.316
0.406
1.000
0.110
Motivation
0.390
0.223
0.110
1.000
Table 4: Correlation matrix for parameters considered by the third factor.
1. Availability of funds required for development.
2. Rigidity of change rejection environment.
3. Sufficient reward system.
4. Enthusiasm and ambition.
6. Conclusions
The following are the conclusions from this study, which was conducted to assess the current
level of, and the improvements required to enhance the implementation of, construction risk
management in Kuwait.
The current level of construction project management knowledge at B.S. level is
satisfactory based on the multidisciplinary interaction required to acquire an advanced level
of risk management skills.
More practical training is required for college students and junior engineers to acquire
know-how in utilizing risk management tools.
401
More awareness is required for senior engineers working in construction project risk
management with respect to the following fields:
-
Insurance;
-
Organizational policy planning and implementation;
-
Project management and general management skills; and
-
Finance, economics, law, and accounting.
Both government and private construction organizations need to be reengineered to
reflect the importance of risk management in their systems.
- Top management must encourage innovative solutions to enhance earned value
through risk management processes.
- Organizations' policies need to include clear motivational clauses to encourage
employees to develop themselves and to acquire new techniques.
- Organizations must construct and enforce a standardized system for all managerial and
technical processes in their work flow.
- Construction organizations need to define well-organized responsibilities and authorities
for all stakeholders, by which risk management processes can be established.
- Training programs should be controlled by monitoring the outputs of these programs to
achieve the required added value.
Reward systems should be provided for employees who acquire knowledge of risk
management through academic or training programs. In addition, there should be a reward
system for the implementation of this knowledge and skills.
References
Alexander, C, 1998. Risk Management and Analysis, Volume I, John Wiley & Sons Ltd.,
Chichester, England , 1998.
Banik, G. C., 2004. Teaching Real-Life Management to MS Construction Students, Proceeding
of ASEE Annual Conference & Exposition, Salt Lake City, Utah.
Chapman, C. and Ward, S., 1997. Project Risk Management: Processes, Techniques, and
insights, John Wiley & Sons Ltd., Chichester, England.
Edwards, L. 1995: Practical Risk Management in the Construction industry, Thomas Telford
Publications, London, UK, 1995.
Grey, S., 1995. Practical Risk Assessment for Project Management, John Wiley & Sons,
Chichester, England.
Holland, J. K., 2006: Risk Management & Contract Guide for Design Professionals, Arden
Publications, Vienna, VA, USA.
Niewoehner, R., 2005. Implementing and Teaching Risk Mitigation in Project Courses,
Proceeding of ASEE Annual Conference & Exposition, Portland, Oregon.
Sherer, B., 2004: Portfolio Construction and Risk Budgeting, Risk Books, Spain.
Smith, N. J., 1999. Managing Risk in Construction Projects, Blackwell Science Ltd, Oxford, UK.
402
Karim El-Dash
Civil Engineering Department
College of Technological Studies
POB 42325, Shuwaikh
Kuwait 70654
Phone: +965-99310261
Fax:
+965-22314533
403
Area: Training in Project Engineering
TOPIC AREA: TRAINING IN PROJECT
ENGINEERING
405
SUPPORT TO CREATIVITY IN THE SEMESTER PROJECT USING
TRIZ
Aguilar-Zambrano, J.
Valencia, M.
Pontificia Universidad Javeriana – Cali
González-Cruz, Mª C.
Abstract
This article shows a qualitative analysis of the effectiveness of the use of TRIZ, a systematic
creativity strategy, as a tool to support the creative competence within the Semester Project in
the curriculum of Electronic Engineering. The analysis is based on an experience carried out
with second year students from the Javeriana University in Cali, Colombia. The Semester
Project is a pedagogical activity seeking to boost different student competences. The project is
based on the formulation of a problem that must be solved by using more than one curriculum
course currently taken by the students. The project is evaluated by a set of professors
throughout the semester. The results obtained show that the proposal of ideas using TRIZ is
higher compared to those generated by trial and error without any creativity technique. TRIZ's
potential for engineering work is corroborated because it uses the accrued knowledge of
designers over the years helping new generations attain this knowledge. It addresses, from
the curriculum, the formation of professionals with abilities to face the new oriented to project
enterprises.
Keywords: Creativity, TRIZ, formation competences, integrated curriculum.
1. Introduction
The student formation processes in engineering are currently focused in the development of
competences, known as the know-how in a context (Aguilar & Rivera, 2005). On the other
hand, it is required that the curricular structures explicitly respond to the new working
challenges of the surroundings (Yorke et al., 2006) and that the curriculum must be taken into
account as a whole, as opposed to an aggregate of subjects along time guided by academic
interests (Koksal & Egitman, 1998). Context recognition and application of pertinent knowledge
in this context calls for new teaching methodologies as well as new attitudes of professors and
students for these purposes. The Semester Project is one of those new methodologies which
seek to look at the curriculum as a whole and form professionals with skills to deal with problem
solving, with specific experiences carried out at the university (Aguilar, 2001).
407
Creativity is a predominant element in product development because it allows the generation of
ideas that can then be materialized for the improvement or the generation of new products.
There are various techniques that enhance creativity, some focused on a personal internal
search for inspiration (brainstorming, analog thinking, lists of attributes, SCAMPER, and many
more) and others, such as TRIZ, which invite the design engineer to use a set of knowledge
already developed by other designers to boost inspiration (Ogot & Okudan, 2006).
The article begins with the presentation of the Semester Project, then describes in a general
way the strategy of systematic creativity - TRIZ and finally presents the pilot experiment
conducted with Electronic Engineering students from second year from the Pontificia
Universidad Javeriana in Cali, Colombia.
2. The Semester Project
Since the year 1996, the Electronic Engineering undergraduate program of the Pontificia
Universidad Javeriana sectional Cali, Colombia, has worked with a pedagogical strategy known
as the Semester Project. The Semester Project is a strategy in the training of future engineers
that from the enunciation of a practical problem, which is related to the subjects that the student
is enrolled in the semester, seeks to develop a set of competences that will be of professional
usefulness both in the disciplinary and labor field. The Semester Project is done each
semester for three years of the programme, from the fourth to the eighth semester. In the
fourth semester, it only involves one subject and it is the preparation of the student for this type
of learning, based on project-based learning. In higher semesters it will have a larger number
of subjects associated to it.
The Semester Project seeks to promote the following types of competences: the academic,
aimed to strengthen the disciplinary formation; the argumentative, which enables the student to
propose and defend an idea; the behavioral, which permits the student to recognize the
different contexts in order to act properly in them; the communicative, which lets the student
interact properly through different means of expression; the conceptual, which recollects the
fundamental elements of the phenomena to make proper use of them; and the creative one
which enables the student to propose new ideas for finding solutions (Aguilar & Rivera, 2005).
2.1 Characteristics of the Semester Project
The Semester Project stems from the collective work of professors of each academic semester.
These professors, before starting the semester, define the problem to be solved and elect a
professor as the Semester Project coordinator. This professor is responsible of coordinating
the project throughout the semester and acts as the information contact with students on all
issues pertaining to the project. Professors act jointly, like a Board of Directors, with the
students in three different moments of the project to receive the results. Professors act on
individual basis throughout the project in guiding students from their area of knowledge. The
following characteristic elements of the Semester Project are recognized:

Formulation of a project, of average complexity, by the group of professors of the
semester.

Planned solution of the problem.

Collective assessment of reports on a regular and programmed basis.
408

Formal documentation of technical reports submitted by students.

Practical and oral verification report of the project results.
Table 1 shows how the Semester Project is evaluated based on the instruments used and on
the aims of the academic activity.
Activities
Instruments
Schedule and
Presentation of
proposed solution
to the problem
First and Second
Project
Submission
Written
document
Final Submission
Communicative,
argumentative,
Behavioural
Competences
Academic
Competence
Conceptual
Competence,
Research
Methodology
X
X
X
X
Oral
Presentation
Paper
X
X
X
X
X
Document
X
X
X
Oral
Presentation
X
X
Written
document,
Table 1: Assessment characteristics of the Semester Project.
3. The strategy of systematic creativity -TRIZ
The theory of inventive problem solving TRIZ was developed by Genrich Altshuller and his
colleagues in the former Soviet Union in the year 1946. This methodology, which combines
principles and algorithms, is now used worldwide. The investigation of TRIZ assumes that
there are universal principles of invention which are the basis for creative innovations that
promote the advancement of technology (Salamatov, 1999). When these principles are
identified and codified, the invention processes can be more predictable. The Altshuller’s
research team has analyzed through 50 years over two million patents. They have classified
them by their level of inventiveness and they have analyzed them to identify the principles of
innovation. The key concepts in TRIZ theory are Ideality, ARIZ, Contradictions, SubstanceField Analysis, Laws of systems evolution, and Knowledge Base of the Effects and Inventive
Principles (Goel & Singh, 1998). At present, there are various works of effective use of TRIZ
with other design engineering methodologies such as Axiomatic Design (Kang, 2004; Shin &
Park, 2006) that help engineers in problem solving.
409
Altshuller’s work presents a set of tools to solve problems where the notion of contradiction is a
key aspect in TRIZ theory. It is affirmed that through a careful analysis of a problem its inherent
contradictions can be identified. The tools that TRIZ provides should be applied every time a
contradiction has been identified (Cavallucci, 2002).
Figure 1 shows an overview of
Altshuller’s work.
Sources
Analysis
Analysis of patents
Classification
Essential notions
Analysis of inventiveness
of solutions
Analysis of psychological
behaviour, biographies
and creativeness of
inventors
Tools to avoid or
prevent psychological
inertia
Analysis of
technical system
evolution’s history
Analysis of existing
methods and tools
Analysis of scientific
and technical
literature
Solving tools
Results
•Psychological inertia
•Ideality
•Contradiction
•Resources
•9 screens
•Miniature men
•Alternative systems
•Size, time, cost operators
•Vépoles and standards
•Matrix for solving technical
contradictions
•Method for solving
phsysical contradictions
•ARIZ
•Data base of scientific
effects
Figure 1: Fields analyzed by Altshuller and the results (Cavallucci, 2002).
Using TRIZ methodology, a design team changes its specific design problem into a general
design TRIZ problem (see Figure 2). The general TRIZ design problem derives from the
analysis and classification of a large number of problems in various engineering fields and
suggests a set of general design solutions. From the general design solutions, the design team
can derive particular solutions for their specific problem. Therefore, the power of TRIZ is in its
inherent ability to bring solutions from different and seemingly unrelated fields of knowledge to
use them in a particular design problem of innovative characteristic.
Salamatov [6] states that "engineers usually think concretely but not systemically. After
establishing a problem often they focus their attention on a particular object that must be
improved. For example: if the problem describes a tree, they only consider the tree. In systems
thinking, it is necessary to imagine not only the tree itself but simultaneously the forest
(supersystem) and separate the branches and leaves (subsystems). Possibly, he must consider
the wood, climate, the cells of the leaves too ". Regarding this aspect, which doesn’t permit to
have a broad vision of the problem, appears the concept of mental inertia which refers to the
internal and external characteristics of people which don’t permit them to overcome a
410
predetermined way of conventional thinking to have access to non-conventional mental
operations.
Figure 2: Characteristic of TRIZ methodology for solving a problem (Cavallucci, 2002).
Figure 3 shows the nine windows diagram or multi-screen that gives a broader vision of a
problem and it is a strategy that is suggested by TRIZ theory to overcome mental inertia. The
horizontal axis is associated with time, where the problem is analyzed, and the vertical axis with
the structure of the problem. It is suggested to follow the strategy according to the numbers
indicated in the windows so that the contradictions associated with the problem can be
identified (Cavallucci, 2007).
411
Figure 3: Nine windows diagram (Cavallucci, 2007).
3.1 The Contradiction Matrix
A contradiction is a situation in which a parameter is affected negatively when wanting to
improve another parameter in the analysis of a problem. For example, if one wants to make a
very fast measuring device, the accuracy of the measurement is affected by the traditional
methods. In this case, the parameter wanted to be improved is speed, but the one which would
negatively be affected is accuracy.
Altshuller suggested a matrix of parameters that allows finding inventive principles of problemsolving from the identification of contradictions in those parameters resulting from the analysis
of a problem. The parameters originally proposed were 39 and the inventive principles, 40. The
parameters can be classified into three groups (Savransky, 2000): Common physical and
geometric parameters, negative parameters independent of the technique and positive
parameters independent of the technique. Contradictions are classified into technical
contradictions and physical contradictions. A technical contradiction is one that has two
different parameters and a physical one, the one that has two equal parameters.
To use the Contradiction Matrix, the Main Function of the system which is going to be designed
must be identified first and the contradictions arise from the analysis of that function. When the
contradictions have been identified, the cell which contains the inventive principles suggested
to overcome such contradictions is identified in the matrix. The cell is the intersection of the
parameter that is improved (column 1 in Figure 4) with the parameter that is worsened (row 1 of
Figure 4). The nine windows allow the identification of both the Primary Function of the system
as well as the contradictions of the problem. Contradictions are found analyzing state
transitions at the time. For example, between window 1 and 9 (see Figure 3), it is possible to
find a contradiction because to reach a new state (window 9) one or more contradictions must
be overcome from the previous state (window 3).
412
With the contradictions identified, particular ideas must be generated from the inventive
principles suggested. Figure 4 shows an example of the use the Contradiction Matrix. A
contradiction is found analyzing a problem that could be the design of a sturdy table of low
weight. The parameter you want to improve is the strength (14), but the parameter that
worsens is the weight of the stationary object (2) for the case of a traditional table. In the cell
of the intersection are the principles suggested to overcome the contradiction and help in the
design process. For example, the suggested principles are: 40, 26, 27 and 1 for the use of
composite materials (40), Use copies (26), disposable Objects (27), and Segmentation (1). The
order of the suggested principles is associated with the highest use frequency in similar
situations in other types of problems. One possible solution could be to build a table with
composite materials like fiberglass which has high resistance and is of low weight, taking into
account the suggestion of the 40th principle.
Figure 4: Example of the use of the Contradiction Matrix.
4. The use of TRIZ in the Semester Project
A pilot experiment was conducted in order to assess TRIZ’s potential in supporting the creative
competences within the Semester Project. There wasn’t any previous work with creativity
strategies and students used intuitively knowledge drawn from textbooks, Internet and
brainstorming. The first experience was carried out with second year Electronic Engineering
students who for the first time are participating in this pedagogical experience and only involves
one subject, Electric Circuits. The proposed project was the manufacturing of equipment which
could pour and weigh ground coffee in definite weights. The problem arises from a real
necessity of small organic coffee producers in Colombia, who require of automatic systems for
413
this type of work and who have few monetary resources to buy an automatic machine for such
purpose.
Students participated in a short TRIZ training, basically in the use of the Contradiction Matrix
and the nine windows, and the instruction was complemented with the use of CREAX
Innovation Suite © software. The training was done in a remote way through video conference
from Valencia, Spain.
To guide the work in the use of the Contradiction Matrix, students were asked to carry out an
activity consistent with the scheme of problem-solving by TRIZ following these steps:
•
Problem Clarification. Each group had to adjust the requirements of the proposed system in
the Project Semester.
•
Reformulation of the problem. Identification of contradictions. Each group had to identify
three parameters that could favor the fulfillment of the requirements of the problem and
from them the ones that will be worsened. It was suggested the use of the nine windows in
order to broaden the vision of the problem and identify contradictions.
•
Generic Solution. From the contradictions, the groups had to find the solution principles
using the Contradiction Matrix.
•
Specific solution. With the principles identified, each group had to generate at least two
ideas per group for the solution of the problem.
5. Results
Sixteen students, organized in four groups, participated in the experience voluntarily. The
groups carried out the proposed tasks, identified the contradictions based on the analysis of the
problem, selected the principles suggested by TRIZ and generated a set of solutions to their
problem.
To illustrate the results, one of the groups presented the analysis that is shown in Table 2.
Some of the students’ ideas generated using the principles were:
● Principle 2. Separation. The circuit must be separated by functions of power supply,
switch control, signal reception of the balance scale, visualization, and reception of other
signals of administration and control.
● Principle 5. Combination. A solenoid valve must be used that works together with data
from the balance scale to control the flow of the ground coffee.
● Principle 35. Parameter Changes (physical states). For an industrial implementation, the
material which the prototype should be built must be stainless steel and also a drying system
must be added in order to prevent the coffee from accumulating and thus impeding the normal
flow through the main funnel.
414
Favourable
factors
Measurement
accuracy (28)
Ease of
operation (33)
Reliability (27)
Harmful factors
Loss of
substance (23)
Time
Consumption
(25)
28,34
10,16
Ease of
manufacturing
(29)
6,35
10,28
2,28
2,5
10,30
10,35
Table 2: Use of the Contradiction Matrix by a group of students in experience with TRIZ in the Semester Project.
After the groups submitted the results of the experience with TRIZ, a discussion group was held
in order to assess their experience. The discussion was focused on the following aspects:

The understanding of the functioning of the TRIZ Contradiction matrix
The assistance of TRIZ to the Semester Project work.
The difficulties in using the TRIZ principles.
The results show that there is an easy understanding by engineering students of the strategy of
systematic creativity TRIZ, particularly of the Contradiction matrix. Regarding the support that
the tool gives to the Semester Project, the students consider it worthy but that they require
more time to fully understand the extension of the principles. According to this, it was verified
that students restricted the use of the principles to generate ideas for aspects associated with
electric circuits rather than the overview of the problem.
Likewise, the results have helped generate a reflection about the importance of providing tools
to enhance creativity in engineering students such as TRIZ. However, a training course with
more examples is required in this particular case of the Semester Project so that students can
understand the meaning of the inventive principles and can use better its potential, since in the
results mental inertia was detected. Despite the fact that some authors consider that little
knowledge makes less effective the use of these techniques in first year students, the authors
of this paper consider that they should be used to favor creative competence.
Currently, the authors are conducting empirical studies with TRIZ with multidisciplinary design
teams, which will analyze the specific characteristics of manipulation of the nine windows and
the identification of contradictions in these teams.
6. Conclusions
In a qualitative way, comparing this experience with the experiences of authors in past years
with the Semester Project, it is verified that the proposition of ideas using TRIZ is higher and of
greater variety than with the absence of techniques to promote creativity. Students using the
TRIZ methodology make better use of information they have obtained from the review of similar
systems and information available on the Internet and a greater variety of proposals between
the different groups involved can be achieved.
The knowledge and use of strategies for generating ideas in engineering courses help
strengthen the creative competence of students. An additional asset, when using TRIZ, is the
415
change that occurs in the way of generating new proposals for solutions in a guided and
convergent manner that deviates from the traditional form of trial and error.
For engineering students, accustomed to procedures and techniques, TRIZ is a strategy of
easy assimilation that allows them to generate new concepts using the experience of many
inventors throughout history. However, unlike other strategies of more intuitive creativity, TRIZ
requires a process of training accompanied by examples to appreciate better its convenience.
The Semester Project has been enriched with the use of this methodology that fulfills an
absence detected within the creative competence that was not explicitly worked with students.
The integrating educational activities such as the Semester Project, which have clearly defined
its working structure and are supported with adequate tools for their execution, address from
the curriculum the formation of professionals with skills to face the new oriented to project
enterprises.
References
Aguilar, J and Rivera L. 2005. Elaboración de un currículo basado en competencias
proceedings XXV Conferencia Nacional de Facultades de Ingeniería ACOFI. Colombia
In
Aguilar, J., 2001. El Proyecto de Semestre: una Actividad para la Formación de Líderes
Académicos. In proceedings XXI Conferencia Nacional de Facultades de Ingeniería ACOFI.
Colombia, pp. 123-128
Cavallucci, D., 2002. TRIZ, the Altshullerian approach to solving innovation problems. In
Engineering Design Synthesis. Amaresh Chakrabarti (Ed), pp. 131-148
Cavallucci, D., Lutz, P and Thiébaud F., 2002. Methodology for bringing the intuitive design
method's framework into design activities. In Proceedings of the Institution of Mechanical
Engineers, Part B: Journal of Engineering Manufacture, pp. 1303-1307
Cavallucci, D., 2007. Theoretical grounding and principles of TRIZ. Course Handouts,
Strasbourg. Insa.
Goel, P and Singh, N., 1998. Creativity and Innovation in Durable Product Development.
Computers Industrial Engineering Vol. 35 Nos. 1-2, pp 5-8
Kang, Y., 2004. The method for uncoupling design by contradiction matriz of TRIZ, and case
study. In Proceedings of Third International Conference on Axiomatic Design, Seoul, pp. 21-24
Koksal, Gulser and Egitman, Alpay., 1998 Planning and Design of Industrial Engineering
Education Quality. Computers Industrial Engineering. Vol 35, Nos. 3-4, pp 639-642
Ogot, M and Okudan G., 2006. Systematic creativity methods in engineering education: a
learning styles perspective. International Journal Education. Vol 22, No.3, pp. 566-576
Salamatov, Y., 1999. TRIZ: The right solution at the right time. Insytec B.V.
Savransky, S., 2000. Engineering of Creativity. CRC Press.
Shin, G. y Park, G. 2006. Decoupling Process of a Coupled Design using Triz Modules, In
Proceedings of Fourth International Conference on Axiomatic Design. Firenze.
Yorke, M. and Knight, P., 2006, Curricula for economic and social gain. Higher Education Vol.
51, pp. 565-588.
416
Jaime A. Aguilar-Zambrano
Science Engineering and Production Department
Pontificia Universidad Javeriana - Cali
Calle 18 No. 118-250 Cali, Colombia
[email protected]: [email protected]
417
PROJECT MANAGEMENT TEACHING IN ENGINEERING HIGHER
EDUCATION: A NEW PERSPECTIVE FOR DEVELOPING
COMPETENCIES
De los Ríos, I.
Ortíz, I.
Díaz-Puente, J.M.
Universidad Politecnica de Madrid
Abstract
This paper sets out the methodological basis of a cooperative model for interdisciplinary
training focused towards developing competencies for the certification in Project Management.
The model is being gradually phased in a new way by the GIE-project Group at graduate and
post-graduate educational levels (in an International Erasmus Mundus Master’s, an
international PhD course, and a Post-Graduate Seminar). The educational cooperation strategy
is directed towards obtaining a validation of competencies as to knowledge, experience and
attitudes in line with the standards of the International Project Management Association (IPMA).
The results show how phasing in teaching aims across different educational levels facilitates a
gradual training in the 46 elements of competencies required to obtain the IPMA’s certification
in Project Management. The educational strategy is conceived as a new teaching dimension
within the framework of the EHEA, taking the course projects - a Preliminary course plan of the
subjects, a Final Degree Course Project and a Final Master’s Course Project- as an educational
component suited to generating a "pre-work experience" to link teaching activity to the business
and industrial environment.
Keywords: Competent Project Management. Cooperative Education. Project-Based Learning.
International Certification. International Erasmus Mundus.
1. Introduction: Certification of competencies
We are currently witnessing a qualitative change in the concept of what constitutes a job that
goes beyond skills and qualifications to focus on competencies. According to this new
concept, job competencies form the foundations of the job world. What business is demanding
nowadays are competent professionals rather than experienced experts.
As part of the framework of the European Higher Education Area emphasis is placed on the
fact that one of the measures needed to become employable shall be through the
development of transversal skills and competencies such as communication and
languages, an ability to handle information, solve problems, work as part of a team and perform
418
well socially (Convención de Instituciones Europeas de Enseñanza Superior, Salamanca,
2001). Competence is an extension of the concept of skill and qualifications, the result of the
way the organisation of work and planning activities are rapidly evolving technologically
(Grootings, 1994). Vocational action competence (Delcourt, 1999) is thus referred to and
may be understood as the sum of the worker’s competencies that are essential if a work task is
to be done well (Delcourt, 1999). Other authors define vocational action competence as the set
of knowledge, procedures, attitudes and abilities that a person possesses and are necessary to
effectively tackle –with the level and quality of performance required- the tasks demanded by a
profession in a particular job to solve any problems that may arise in an independent and
creative manner, and contribute to organising the work and the social working environment
(Echeverría, 2002). Numerous studies throughout the world are proposing this new focus on
competence-based education (Parsons, Caylor and Simmons, 2005), as a process that must
integrate knowledge, skills and values (Mulcahy, 2000). This new approach has led to
certification taking on a more important role in all disciplines both within the sphere of
engineering and other professional fields (Nelly, 2007). In this context different models of
competence certification are appearing (EMCI, 2007; ASCE, 2005; CHEA, 2001; CONOCER,
1997; Long, and Kishchuk, 1997) with different approaches in their outlines, processes and
certification systems.
The concept has also taken deep root in the working environment for selecting staff and in
“management by competencies” when evaluating human capital and even in the form of
competence-based remuneration (Echeverría, 2002).
2. Competence certification in the sphere of Project Management
The basic terms for certification in the area of project management derive from the international
standard ISO/IEC 17024 “General requirements for organisations that issue individual
certifications” [8]. Within this international scope of competence certification, competence is
defined as the aptitude shown for applying knowledge and/or skills, and when applicable,
proven personal attributes (UNE-EN ISO/IEC 17024, 2003).
In the area of competence in project management the IPMA model (AEIPRO-IPMA, 2006)
proposes a certification system based on three areas of competencies: technical, behavioural
and contextual competencies. The IPMA developed these competencies (ICB-IPMA
Competence Baseline) from the National Basic Competencies (NCB). In the case of Spain,
AEIPRO, a member of the Spanish section of the IPMA, draws up the Requirements for
Competence in Project Management (NCB) that set out the main work experience for the
IPMA’s universal certification system in combination with the 40 IPMA member associations
spread across the world.
For many years the qualifications and competence programmes have been a matter of great
importance for project management associations throughout the world. Apart from the IPMA
model that has been around since 1965, there are other project management certification
Institutes: the Project Management Institute defines a certification system based on areas of
functional knowledge (Project Management Institute, 2000) while Austria’s Project Management
system refers to people’s knowledge (explicit) and experience (Gareís, 2002) In the sphere of
engineering projects the cooperative education programmes are presented as an approved
option for students to improve their acquisition of competencies (Parsons, Caylor and
Simmons, 2005; Bartkus, 2001; Hackett, Martin and Rosselli, 1998).
419
3. A model of de cooperative education for competence in Project Management
Although certification, education and training need to be completely separate [8], candidates
need to receive education and training in project management if they are to be successful in the
certification process (AEIPRO-IPMA, 2006). This is why higher education institutions oriented
towards training future professionals in the area of project management must enhance the
chances of their pupils, students or course participants obtaining a specific certificate that
reflects their personal attributes, education, training or work experience.
Therefore, the UPM’s Educational Innovation Group, part of the GIE-Project formed by a
group of teachers from the UPM’s Project Engineering area is implementing a cooperative
education model directed towards student competence in project management. This model is
geared to generating a new teaching dimension around the Course Projects – a Preliminary
course plan of the subjects, a Final Degree Course Project and a Final Master’s Course
Project- as an educational component suited to generating a "pre-work experience" to develop
new learning methods within the framework of the EHEA to link teaching activity to the business
and industrial environment (GIE-Project, 2007).
The cooperative education model is gradually developing competencies in the different
educational levels in the field of engineering in accordance with the IPMA’s international
standards: two degree subjects, two post-graduate programmes and an international PhD.
Cooperative education (co-op) is developed using a multidisciplinary group of Engineering
Projects teachers from Madrid Technical University (UPM) and through agreements with
numerous Public Institutions –the Agricultural and Rural Development Department of the
Madrid Regional Government and the Regional Institute of Safety and Hygiene at Work
(IRSST) of the Madrid Regional Government– Local Action Groups, development project
managers and Business. These educational agreements between the UPM and the different
public and private institutions jointly define the scope of the projects, the partners and the
interested parties and the requirements and goals of the projects.
The projects are carried out by a team in a real context. Each group of students with the help of
a teacher-tutor must solve the problems set in the real context of their work, structure their
project, decide the work to be handed in, plan the project stages and schedule them according
to the time available and the team’s resources, manage the changes, periodically inform the
teacher-tutor and the promoter of the progress being made and finally document the results
obtained and verbally inform the representatives of the institutions collaborating of these
results. In some cases -as with the subject called 5th course rural development Projects- the
teams have some financial resources at their disposal to develop the projects. In many cases
these projects are the object of new phases, and may even be actually put into practice.
These tasks off cooperation are the mechanism for gradually facilitating access to the three
spheres of competencies: technical, behavioural and contextual. The teaching programmes
designed allow evaluating knowledge concerning the basic elements of Project Management.
The strategy is being designed in collaboration with the Spanish Engineering Projects
Association (AEIPRO), a part of the IPMA since the beginning of the 90s; this ensures that the
competencies in the teaching programmes are up-to-date and are incorporated into the
international corpus of knowledge on project management (Kerzner, 2000; Huemann, 2000;
Crawford, Pollack and England, 2006; Crawford, 2005; Morris, Patel and Wearne, 2000). Table
1 summarises the programmes of the different levels of education and the external institutions
that are cooperating on this educational innovation experience.
420
Programme
Levels
Programme Name
ECTS
Institutions UPMETSI
Grade
Projects (1 group of
students)
3
DPPR-ETSI
Agricultural
Scientists
Mexico Postgraduate
School
Integrated Rural
Development Projects
3
DPPR-ETSI
Agricultural
Scientists
DG Agriculture and
Rural Development
(Madrid School)
Agricultural
Scientists
Forestry
INFODAL.
4th course
Grade
5th course
International
Master’s
Postgraduate
Postgraduate
Seminar
International Master’s
Erasmus Mundus
60
Industrial
Engineers
Certification in Project
Management:
Requirements for the
Competence
2
Agricultural
Scientists
Industrial
Engineers
External Institutions
U. Cork; CNEARC
EPTISA, IDOM
IDOM Engineering
Company
CPV-Project
Management Company
Civil Engineers.
PhD
Sustainable planning and
management of
Development Projects
60
Agricultural
Scientists
Forestry
Mexico Postgraduate
School
Industrial
Engineers
Table 1: Programmes and subjects of the different degree and postgraduate levels.
All the educational actions of the programmes are based on the method of learning by doing
that the UPM’s Department of Projects and Rural Planning and the Madrid Regional
Government have been developing since 1989 for advising on the design of Engineering
Projects in rural areas (Cazorla, 2006; Cazorla, De los Ríos, 1998, 1996;
www.grupogesplan.es). The degree programme courses -4th and 5th course Project subjects
and the Final Degree Course Project- aim to prepare students for a possible IPMA Level D
certification (Expert in Project Management).
The Erasmus Mundus Master’s programme (www.agrismundus.eu/agris-mundus/ ) and the
Postgraduate Seminar in Project Management (www.grupogesplan.es) seek to ensure that
those taking part are well-prepared to pass the IPMA Level D certifications, and if they can
prove work experience in simple projects they can opt to do Level C (Project Management
Professional). Students with work experience who can prove satisfactory use of the elements of
competence in complex projects will be able to opt for Level B certification (Project Manager).
For Level A (Programme Manager or Project Managers) satisfactory use of the elements of
competence in programme or portfolio coordination must be proved. It must be borne in mind
that education and training cannot replace a person’s having the appropriate level of
experience needed to pass the IPMA’s Levels C, B and A certification, which means that at
these levels in no case will access be possible without possessing the required experience.
Figure 1 shows an outline of these educational cooperation strategy levels aimed at validating
421
individuals’ competence in project management within the IPMA system. 46 elements of
competence for project management are gradually approached. The first training levels look at
elements of technical competence, while the postgraduate levels approach the elements of
contextual competence, connected with the context of the project and organisation, and the
elements of the sphere of behavioural competence related to the attitudes and personal skills
required for project management.
+ WORK EXPERIENCE
=
Experience
+
Attitude
+
Knowledge
CO
N
BE TEX
HA TU
Co VI A
mp OU L a
ete RA nd
nc L
e
Knowledge
+
International PhD
Sustainable Project Planning and Management
UPM Postgraduate Seminar
Project Management Competence Requirements
1 ECTS
International Master Erasmus Mundus
UPM Master of Science
TE
Co CH
mp NIC
et AL
en
ce
IPMA certification
certification system
system
IPMA
Competence
60 ECTS
Project Subjects UPM Degree
4th Course + 5th Course +
Final Degree Course Project
Figure 1: Levels of the training model focused on the development of competencies in project management in line
with the IPMA system.
As we have seen, the IPMA proposes a Project Management certification model based on three
areas of competence: technical, behavioural and contextual. The following table lists the 46
elements of the three spheres of competence that are gradually approached in the educational
cooperation strategy presented.
422
Grade
Postgraduate
PhD
Final
TRAINING-EDUCATION MODEL
COMPETENCES
4th
course
5th
course
Degree
Course
E. Mundus
Master’s 3
Postgraduate
Seminar4
International
PhD
Project
1. Technical competences
1.01 Project Management Success
1.02 Interested Parties
1.03 Project requirements & objectives
1.04 Risks & opportunity
1.05 Quality
1.06 Project organisation
1.07 Teamwork
1.08 Problem resolution
1.09 Project structures
1.10 Scope and deliverables
1.11 Time & project phases
1.12 Resources
1.13 Cost & finance
1.14 Procurement & contracts
-
-
-
-
1.15 Changes
Grade
Postgraduate
PhD
Final
TRAINING-EDUCATION MODEL
4th
course
5th
course
COMPETENCES
Degree
Course
E. Mundus
Master’s 3
Postgraduate
Seminar4
International
PhD
Project
1. Technical competences
1.16 Control & reports
1.17 Information and documentation
1.18 Communication
-
1.19 Start-up
-
1.20 Closure
2 Behavioural competencies
2.01 Leadership
-
2.02 Engagement & motivation
-
2.03 Self-control
-
-
-
2.04 Assertiveness
2.05 Relaxation
-
-
-
-
423
2.06 Openness
2.07 Creativity
2.08 Results orientation
2.09 Efficiency
2.10 Consultation
2.11 Negotiation
-
-
-
-
-
-
-
2.12 Conflicts & crisis
2.13 Reliability
2.14 Values appreciation
2.15 Ethics
3 Contextual competences
3.01 Project orientation
3.02 Programme orientation
3.03 Portfolio orientation
-
3.04 Project-programme implementation
-
3.05 Permanent organisation
-
3.06 Business
-
3.07 Systems, products and technology
3.08 Personnel management
3.09
Health,
environment
security,
safety
-
&
3.10 Finance
3.11 Legal
Competence partially developed in the training programme
Competence fully developed in the training programme
FDCP: Final Degree Course Project;
4. Conclusions
The educational strategy goes back to 1989 as part of the scope of a 5th course Project subject
as part of an agreement with two Engineering Firms IDOM and CPV, with the backing of
AEIPRO with the Madrid Regional Government to advise on the design of Engineering Projects
in rural areas. During the 2004-05 course, it was validated as a pilot experiment in the UPM
within the framework of the European Higher Education Area (EHEA). After a new educational
innovation project, the following course took a further step by involving one of Mexico’s major
postgraduate institutions, the use of ICTs, a system of virtual tutorials and the use of video
conferences Educational cooperation was extended to the international sphere by an
International Erasmus Mundus Master’s –in cooperation with 6 EU Universities and another 9
from outside the EU- and a joint PhD with the Mexico Postgraduate School. Since 2006 the
424
strategy has been complemented by a Postgraduate Project Management seminar in
collaboration
The educational strategy generates a new teaching dimension, taking the course projects as an
educational component suited to generating a "pre-work experience" within the framework of
the EHEA to link teaching activity to the business and industrial environment. 46 elements of
competencies are gradually approached. In the first courses 20 elements of technical
competence are taken into account that are basic for managing projects. In the postgraduate
levels the sphere of contextual competence is examined in greater detail where 11 elements
related to project context and organisation are covered. Finally, within the sphere of
behavioural competencies 15 elements are taken into account that cover the attitudes and
personal skills required for project management.
References
AEIPRO, 2006. Internacional Project Management Association (IPMA). Bases para la
Competencia en Dirección de Proyectos Versión 3.0.
ASCE, 2005. American Society in Civil Engineering. “Certification” http://www.asce.org/
professional/certification/ (October).
Bartkus, K. R. 2001. Skills and Cooperative Education: A conceptual Framework. Journal of
Cooperative Education, Vol. 36, No 2001, pp. 17-24.
Cazorla, A., 2006. Planificación para la Sostenibilidad: Proyectos de Ingeniería en un ámbito
Local-Rural. ETSIA. UPM.
Cazorla, a., De los Ríos, I. 1996. La enseñanza del desarrollo rural y la planificación en su
nueva dimensión: una estrategia metodológica. II Jornadas Nacionales de Innovación en las
Enseñanzas de las Ingenierías. Universidad Politécnica de Madrid. Instituto Ciencias de la
Educación ICE. Madrid. Págs. 564-574.
Cazorla, A., De los Ríos, I., 1998. La enseñanza del desarrollo rural: una estrategia
metodológica. El Ingeniero Agrónomo en el contexto de la nueva Política de Desarrollo Rural.
Editorial: Fundación Premio ARCE. Colección Jornadas Agronomía. Madrid.
CONOCER., 1997. Sistema Normalizado y de Certificación de Competencia Laboral, Consejo
de Normalización y Certificación de Competencia Laboral. CONOCER, México, D.F.
Convención de Instituciones Europeas de Enseñanza Superior, 2001, Salamanca. Perfilando el
Espacio Europeo de la Enseñanza Superior.
Council for Higher Education Accreditation (CHEA), 2001. Glossary of Key Terms in Quality
Assurance and Accreditation. http://www.chea.org/international/inter_glossary01.html
Crawford, L. Pollack, J and England, D., 2006. Uncovering the trends in project management:
Journal emphases over the last 10 years. International Journal of Project Management. Volume
24, Issue 2 , Pages 175-184
Crawford, L., 2005 Senior management perceptions of project management competence.
International Journal of Project Management. Volume 23, Issue 1, Pages 7-16
Delcourt, J. 1999. Nuevas presiones a favor de la formación en la empresa. Formación
Profesional, 17, 3-14.
425
Echeverría, B. 2002. Gestión de la competencia de acción profesional. Revista de
Investigación Educativa, 20 (1), 7-43.
EMCI (2007). Engineering Management Certification International. A Guide to the Engneering
Management
Certification
Body
of
Knowledge
(EMC-BOK
Guide).
http://www.engineeringcertification.org/
Erasmus Mundus Master’s. 2006. Sustainable Development in Agriculture Master of Science “Rural
Local
Development”.
http://ec.europa.eu/education/programmes/mundus/
projects/2006/11.pdf ; http://www.agrismundus.eu/agris-mundus/
Gareís, R. 2002. Pm baseline, Knowledge Elements for Project and Programme Management
and for the Management of Project-Oriented Organisations, Versión 1.0 2001. Project
Management Austria. 2002.
GIE-Project. 2007. Grupo de Innovación educativa de la UPM. Objetivos del Grupo.
http://www.upm.es/innovacion/cd/06_gie/grupos/consolidados/ GIE27.pdf
Grootings, P., 1994. De la cualificación a la competencia: ¿de qué se habla?, Revista Europea
de Formación Profesional, 1, 5-7.
Hackett, R.K., Martin, G.R. and Rosselli, D.P. 1998. Factors Related to Performance Ratings of
Engineering Students in Cooperative Education Placements. Journal of Engineering Education,
Vol. 87, No 4, 1998, pp. 445-458.
Huemann, M., 2000. Individual Project Management Competences- The Need for Project
Management Knowledge and Experience. 15th IPMA World Congress on Project Management.
Kerzner, H., 2000. Project Management. A Systems Approach to Planning, Scheduling and
Controlling. John Willey &Sons, New York.
Long, B. and Kishchuk, N. 1997. National Council of the Canadian Evaluation Society.
Professional Certification. CES.
Morris, P. W. G. Patel M. and Wearne S.H., 2000 Research into revising the APM project
management body of knowledge. International Journal of Project Management. Volume 18,
Issue 3, Pages 155-164.
Mulcahy, D. 2000. Turning the contradictions of competence: competence-based training and
the beyond. Journal of Vocational Education and Training, 52 (2), 259-280.
Nelly, W. 2007. Certification and Accreditation in Civil Engineering. Journal of Professional
Issues in Engineering Education and Practice. July 2007. pp. 181-187.
Parsons, C., Caylor. E. and Simmons H. 2005. Cooperative Education Work Assignments The
role of Organizational and Individual Factors in Enhancing ABET Competencies and Co-op
Workplace Well-Being. Journal of Engineering Education. pp. 309-316.
Project Management Institute, 2000. A guide to the Project Management Body of Knowledge.
Newtown Square. Project Management Institute.
UNE-EN ISO/IEC 17024 2003. Conformity assessment - General requirements for bodies
operating
certification
of
persons.
http://www.iso.org/iso/iso_catalogue/catalogue_tc/
catalogue_detail.htm?csnumber=29346
426
Acknowledgements
We should like to acknowledge the collaboration of the institutions participating in the
International Erasmus Mundus project entitled, “Rural Local Development” Master’s Course, of
the International PhD between the UPM and the Mexico School of Postgraduates in
“Sustainable Planning and Management in Rural-Local Development” and of the postgraduate
Seminar, “Certification in Project Management: Requirements for Competence”, for their
contribution to the results of the paper.
Correspondence (For more information contact):
Ignacio de los Ríos
GESPLAN (Grupo I+D+i Planificación y Gestión del Desarrollo Rural-Local)
GIE-Project. Grupo Innovación Educativa.
Sustainable Development IMR&LD Agris Mundus Director UPM
Dpto. Proyectos y Planificación Rural.
28040 Madrid. Spain
Phone: + 34 (0) 91 336 58 38 / 610 877 699
Fax: + 34 (0) 91 336 58 35
E–mail : [email protected]
www.grupogesplan.es
www.agrismundus.eu/agris-mundus/
427