Real-Life Knowledge Management

Transcription

Real-Life
Knowledge Management
Lessons from the Field
Abdul Samad Kazi
Patricia Wolf
Real-Life
Edited by:
Dr. Abdul Samad (Sami) Kazi and Dr. Patricia Wolf
Copyright © 2006 KnowledgeBoard
Some rights reserved. This book (and its digital version that is available online through
http://www.knowledgeboard.com) is released under a Creative Commons AttributionNonCommercial-NoDerivativeWorks license. Under this license, you are free to copy, distribute,
and perform the work presented in this book under the following conditions: you must give the
original author credit for the work; you may not use it for commercial purposes; you may not alter,
transform, or build upon this work (for more details on this license, visit
http://creativecommons.org/licenses/by-nc-nd/2.5/). For other permissions, you should contact the
author(s) of the case study in question.
Although all care is taken to ensure the integrity and quality of this publication and the information
contained herein, no responsibility is assumed by the publishers, editors, or the authors of individual
case studies for any damage to property or persons as a result of operation or use of this publication
and/or the information contained herein.
Published by: KnowledgeBoard in collaboration with,
VTT –Technical Research Centre of Finland (www.vtt.fi)
ISBN 952-5004-72-4
Printed in Finland
Real-Life Knowledge Management: Lessons from the Field
Table of Contents
Table of Contents
iii
Preface
vii
Industrial Experiences and Communities of Practice
Case 1:
Learn, Exchange, Develop
5
Peter Troxler, ETH VDI, [ k n w l d g ], the Netherlands
Kristina Lauche, TU Delft, the Netherlands
Case 2:
The Knowledge Management Journey of Israel Aircraft Industry
23
Rony Dayan, Israel Aircraft Industry, Israel
Edna Pasher, Edna Pasher and Associates, Israel
Ron Dvir, Innovation Ecology, Israel
Case 3:
Communities of Practice – A Case Study from the Automotive
Industry
47
Patricia Wolf, Hochschule für Wirtschaft (HSW) Luzern, Switzerland
Abdul Samad (Sami) Kazi, VTT -Technical Research Centre of Finland
Case 4:
Sharing through Social Interaction: The Case of YIT
Construction Ltd.
63
Abdul Samad (Sami) Kazi, VTT – Technical Research Centre of Finland
Anssi Koivuniemi, YIT Construction Ltd., Finland
Case 5:
Addressing Communication Silo’
s using Complexity Techniques
and Social Network Analysis
81
Sonja Blignaut. IBM Business Consulting Services, South Africa
Case 6:
Beneath the Surface of Conscious Patterns: Using Narrative to
Characterise the Culture of Innovation at a Leading R&D
Organisation
103
Gillian A. Maree, South African Council for Scientific and Industrial Research
Dirk J. Roux, South African Council for Scientific and Industrial Research
Mario A. Marais, South African Council for Scientific and Industrial Research
iii
Table of Contents
Industry Analysis and Business Modelling
Case 7:
Applying Business Modelling to Knowledge Management
129
Lale Citipitioglu Eidal, DNV Maritime, Norway
Flemming Hjorth, DNV Maritime, Norway
Jørgen Kadal, DNV Maritime, Norway
Case 8:
From Knowledge Management to Business Intelligence
145
Joël Van Hoolandt, Magma Consulting, Belgium
Case 9:
Enabling Cross-Project Knowledge Creation through
“Knowledge Oriented Project Supervision”
163
Waltraud Grillitsch, University of Klagenfurt, Austria
Alexandra Müller-Stingl, University of Klagenfurt, Austria
Robert Neumann, University of Klagenfurt, Austria
Case 10:
Sharing Knowledge between Generations: a Case Study about
Managing Apprenticeship in Italian Districts
181
Marco Ruffino, IRSEA and University of Bologna, Italy
Case 11:
Knowledge Based Methods in Cost Reduction Processes
195
Roberta Cuel, University of Trento, Italy
Case 12:
Supporting Mathematical Modelling by Guiding and Monitoring
Huub Scholten, Wageningen University, the Netherlands
Adrie J.M. Beulens, Wageningen University, the Netherlands
iv
213
Design and Implementation of KM Solutions
Case 13:
Knowledge Management for Organisations Dealing with Critical
Incidents
235
Andrew Lewis, France
Ben Diamond, West Midlands Fire Service, United Kingdom
Case 14:
prodela: On-Demand Training for Public Administrations
253
Liza Wohlfart, IAT, University of Stuttgart, Germany
Dorothee Frielingsdorf, IAT, University of Stuttgart, Germany
Case 15:
Building KM @ Patni
269
Shashi Kadapa, Patni Computer systems Ltd., India
Case 16:
Identifying Key Skills and Competencies Across the Enterprise
287
Gary Cairns, AMEC NNC Limited, United Kingdom
Case 17:
Open Source Content Management System for a Sports Centre
301
Arthur Visser, I.C.E. Europe, Belgium
Case 18:
Virtual Collaboration in the Brazilian Mould and Die Making
Industry
321
Rolando Vargas Vallejos, Universidade de Caxias do Sul, Brazil
Patricia Wolf, Hochschule für Wirtschaft (HSW) Luzern, Switzerland
v
Preface
Background and Introduction
There is a body of knowledge and literature around the broad theme of knowledge management.
While some is grounded in scientific theories, others offer guidance on how to undertake successful
knowledge management initiatives. There is however, very limited material on real-life cases that
one can actually learn from. The aim of this book is to tap into the experience of members of the
KnowledgeBoard community and to then share this experience through a structured set of hands-on
case studies.
Case studies offer an opportunity for reflection and learning through close examination of the
domain or industrial sector presented within a case study. While some may argue that as opposed to
scientific methods and theories, case studies cannot be replicated, they serve as a valuable
mechanism for possible adaptation. They serve as an effective means for the capture of good and
bad practices in a simple and reader-understandable manner. It may be argued that case studies are
tacit in nature. This is true to the point that many rely on the experiences of those engaged in the
case study. Thereby, they offer an opportunity to access the true source of knowledge. In the words
of Albert Einstein, “The only source of knowledge is experience”1.
In early 2005, we were time and again asked by different members of the KnowledgeBoard
community for case studies on knowledge management. We explored the idea further and realised
that a book on real-life knowledge management experiences could serve as an excellent mechanism
for creating a portfolio of case studies from different domains and industrial sectors. Based on more
than thirty expressions of interest and more than twenty case study submissions, we set about on the
task to create KnowledgeBoard’s first book. Our main selection criterion for the cases was that they
must be practical and easy to understand and together, should cover a broad set of domains and
industrial sectors. Each of the eighteen case studies in this book has been peer-reviewed and where
required, its contents improved.
Real-Life Knowledge Management: Lessons from the Field is about the art and practice of
knowledge management in real life settings. Knowledge is both created and shared within a context.
Based on the hands-on experiences of the authors’where the knowledge was created and valuable
lessons learned, this book provides the context for the sharing of this knowledge through a set of
case studies spanning several domains and areas of practice.
Through the eighteen case studies presented in this book, we learn about practical experiences in
consultancies, financial institutions, fire fighting, global project management groups, public
administrations, research and development organisations, small and medium enterprises, sports
centres, the aerospace sector, the automotive industry, the construction industry, the IT sector, the
maritime industry, the mould and die making industry, the nuclear industry, the oil and gas industry,
and the water management sector. The case studies draw from a pool of experience spanning
Austria, Belgium, Brazil, Finland, Germany, India, Israel, Italy, Norway, South Africa, the
Netherlands, and the United Kingdom.
1
http://uk.encarta.msn.com/text_761562147___4/Einstein_Albert.html
vii
Preface
Case Study Structure
All eighteen case studies presented in this book have been organised around a common content
presentation structure to foster ease of reading and understanding. Structured around industrial
experiences and communities of practice, industry analysis and business modelling, and the design
and implementation of knowledge management solutions, each case study features sections on
industrial context, problem definition, learning objectives, approach, case analysis, key findings,
business impacts, a set of conclusions, and most importantly, offers a valuable set of practical tips
and lessons learned.
Background
Industrial Context
Problem
Background information on
the case study from an
industrial perspective
Description of the problem(s)
that the case study addresses
Learning Objectives
What the reader will learn from the case study
Approach
Tools, techniques, methodologies used for the
investigation and analysis of the case study
Case Analysis
Detailed analysis and
discussion of the case study
Results and Business Impacts
Key Findings
Business Impacts
Main findings from the case
analysis including barrier &
enablers of success/failure
Verified and/or potential
business impacts based on
case analysis
Conclusions
Description of the issues addressed, reflections
on the solutions identified, and identification of
important emerging issues for further exploration
Practical Tips & Key Lessons
Practical tips and lessons learned from the case study.
These are the main takeaways and nuggets for the reader
Case Study Structure and Content Flow
viii
Portfolio of Case Studies
Knowledge is co-created and shared within a context. This book serves as such a context for sharing
knowledge. Hopefully you will use it as a basis for new knowledge creation/adaptation within your
organisations, and use it to get in touch with the authors to share your experiences and co-create
new knowledge. We hope you enjoy this collection of real-life case studies in knowledge
management as much as we and the authors did in presenting and compiling some of the great
experiences we had in exploring, implementing, and being part of knowledge management
initiatives in different industrial settings.
Let us now explore what we can learn from each of the case studies.
ix
Preface
Industrial Experiences and Communities of Practice
1
Peter Troxler and Kristina Lauche
This case study is about how three organisations in the oil and gas
industry implemented communities of practice to foster learning, sharing
and development of knowledge. In this case study, we recognise that onedimensional, tactical implementations of communities of practice fail to
deliver all potential. We learn the various characteristics of communities
of practice that need to be addressed when designing such communities
and understand that one purpose of communities of practice is to build
capacities and that therefore their implementation needs a strategic
approach.
Peter Troxler
Kristina Lauche
The Knowledge Management Journey of Israel
Aircraft Industry
Rony Dayan, Edna Pasher and Ron Dvir
This case study is about the journey of Israel Aircraft Industries into
knowledge management. It describes how the plan was embedded in the
company’s comprehensive change program which focuses on four core
principles: customer; people; innovation and technology; and one
company. In this case study, we learn how to need to balance well
planned and emergent initiatives, central and local resources,
comprehensive strategies and step-wise implementation.
3
Communities of Practice
–A Case Study from the
Automotive Industry
Patricia Wolf
Abdul Samad (Sami) Kazi
x
2
The Knowledge
Management Journey of
Israel Aircraft Industry
Rony Dayan
Edna Pasher
Ron Dvir
Communities of Practice – A Case Study from the
Automotive Industry
Patricia Wolf and Abdul Samad (Sami) Kazi
This case study describes how upon a merger of two automotive firms, a
project aimed at facilitating co-creation of knowledge across different
functional units was implemented. It provides an overview of the
conception, implementation, and optimisation of the project. The case
study provides reflections and key lessons learned on how crossfunctional communities of practice can be developed, nurtured, sustained,
and as to what radical changes they may bring into existing organisational
decision making processes.
Sharing through Social Interaction: The Case of YIT
Construction Ltd.
Abdul Samad (Sami) Kazi and Anssi Koivuniemi
This case study introduces the concept of social processes and
intervention points as effective instruments for the capture and sharing of
knowledge in project-based industries such as construction. We learn the
use of social processes in solving complex problems through crossfunctional teams and are exposed to new forms of knowledge
visualisation through a combination of building product models and
knowledge sources.
Addressing
Communication Silo’
s
using Complexity
Techniques and Social
Network Analysis
Sonja Blignaut
5
4
Sharing through Social
Interaction: The Case of
YIT Construction Ltd
Anssi Koivuniemi
s using Complexity
Techniques and Social Network Analysis
Sonja Blignaut
This case study brings forth experiences from the information
management division of a large financial institution in South Africa that
was struggling with functional silos that prohibited their ability to
leverage the strengths of their different areas of specialisation. In this case
study, we are reminded not to forget the context in which an SNA was
run. If run to determine the knowledge network, it should not be used to
extrapolate the graphs to provide cultural insights.
Beneath the Surface of Conscious Patterns: Using
Narrative to Characterise the Culture of Innovation
at a Leading R&D Organisation
Gillian A. Maree, Dirk J. Roux and Mario A. Marais
This case study provides insights into the use of narrative techniques to
uncover behaviours, themes and archetypes beyond everyday conscious
patterns of recognition within and R&D organisation. We learn of ten
interesting archetypes for innovation, namely the: Clumsy Puppy, Couch
Referee, Courageous Captain, Bright-Eyed-Bushy-Tailed Researcher,
Inspiring Coach, Intellectual Maverick, Intrepid Explorer, Meticulous
Bureaucrat, Narrow-minded Nitpicker and Willing Victim.
6
Beneath the Surface of
Conscious Patterns: Using
Narrative to Characterise
the Culture of Innovation
at a Leading R&D
Organisation
Gillian A. Maree
Dirk J. Roux
Mario A. Marais
xi
Preface
Industry Analysis and Business Modelling
7
Applying Business
Modelling to
Lale Citipitioglu Eidal
Flemming Hjorth
Jørgen Kadal
Applying Business Modelling to Knowledge
Management
Lale Citipitioglu Eidal, Flemming Hjorth and Jørgen Kadal
This case describes how one organisation, DNV Maritime, discovered
and applied a unique approach to business modelling for developing and
communicating effective knowledge management solutions within a
complex environment of a wide range of processes, technologies, and
competencies. We learn how business modelling can help describe and
visualise all aspects of a business environments for use in identifying
knowledge management needs and solutions.
From Knowledge Management to Business
Intelligence
Joël Van Hoolandt
This case study discusses the changes that occurred in a global
knowledge management program within a big financial services
company. We learn of the strategic knowledge management market
segmentation within an audit firm and are informed that management
shifts in knowledge management strategies are not accidental. The case
study informs that international knowledge management projects embed
changes within themselves, going from a technical stage to a people focus
and, finally, a business intelligence stage.
Enabling Cross-Project
Knowledge Creation
through “Knowledge
Oriented Project
Supervision”
9
Waltraud Grillitsch
Alexandra Müller-Stingl
Robert Neumann
xii
8
From Knowledge
Management to Business
Intelligence
Joël Van Hoolandt
Enabling Cross-Project Knowledge Creation through
“Knowledge Oriented Project Supervision”
Waltraud Grillitsch, Alexandra Müller-Stingl and Robert Neumann
This case study describes the concept and process of a successful
institutionalisation to foster communication and sharing among through
the use of knowledge oriented project supervision meetings. We learn of
the reasons and pre-conditions for implementing knowledge management,
the key concepts and processes of effective knowledge oriented
supervision meetings, and how to institutionalise knowledge management
within organisations in the long-run.
Sharing Knowledge between Generations: a Case
Study about Managing Apprenticeship in Italian
Districts
Marco Ruffino
10
Sharing Knowledge
Between Generations: a
Case Study about
Managing Apprenticeship
in Italian Districts
Marco Ruffino
This case study concentrates on how effective knowledge management as
opposed to knowledge transfer can be used as an effective learning
mechanism for apprentices within SMEs. We learn of the typical
knowledge transmission problems in SMEs-apprentice relationships and
of strategies aimed at improving learning processes from both formal and
non-formal learning perspectives.
11
Knowledge based
Methods in Cost
Reduction Processes
Roberta Cuel
Knowledge Based Methods in Cost Reduction
Processes
Roberta Cuel
This case study presents how consultants manage knowledge in order to
effectively produce value from their activities. It primarily focuses on
social aspects, namely the relations among individuals that allow people
to improve their productivity in terms of expertise in the service activity
and innovation of products and solutions. We learn how cost management
consultants through re-alignment of their strategies become knowledge
brokers amongst producers and vendors and enable co-innovation
amongst them.
Supporting Mathematical Modelling by Guiding and
Monitoring
Huub Scholten and Adrie J.M. Beulens
This case study discusses how a knowledge base can be used to solve
complex multi-disciplinary problems through a model based approach in
the water management sector. We learn how successful execution and
completion of multi-disciplinary complex projects can be supported
through a knowledge-based system. This system facilitates
communication between team members and helps them to report what has
been done, formatted for various purposes and audiences.
12
Supporting Mathematical
Modelling Management
by Guiding and
Monitoring
Huub Scholten
Adrie J.M. Beulens
xiii
Preface
Design and Implementation of KM Solutions
13
for Organisations Dealing
with Critical Incidents
Knowledge Management for Organisations Dealing
Andrew Lewis and Ben Diamond
This case study discusses the development and implementation of
knowledge management tools and processes for use by fire fighters. We
learn how critical incident managers can integrate outside knowledge into
their decision-making without experiencing information overload. The
case study also describes how KM methods and tools can gain the trust
necessary for their deployment in domains where lives are on the line and
where trust is generally built through long-standing relationships.
Andrew Lewis
Ben Diamond
prodela: On-Demand Training for Public
Administrations
Liza Wohlfart and Dorothee Frielingsdorf
This case study discusses how an on-demand training solution was
developed to help German public administration keep up their
modernisation process through translation of theoretical concepts into
real-life practice. We learn how by combining e-learning, traditional
seminars and knowledge management in an innovative mixture can
enable on-demand support and strengthen community aspects within
public administrations.
15
14
prodela
– On-demand Training for
Public Administrations
Liza Wohlfart
Dorothee Frielingsdorf
Building KM @ Patni
Shashi Kadapa
Building KM @ Patni
Shashi Kadapa
xiv
This case study describes how a knowledge management initiative at
Patni (global IT services provider) was used to build and sustain a
focused KM solution to meet the challenge of generating knowledge
value for a highly mobile global 10,000+ workforce, serving Fortune
1000 customers. We learn how to build a brand within an organisation
around a knowledge management initiative and as to how to harness
various drivers such as market, people, technology and others to make the
knowledge management initiative sustainable and successful.
Identifying Key Skills and Competencies Across the
Enterprise
Gary Cairns
This case study describes the design and implementation of a tool for
identifying qualifications, skills, and experiences of employees within an
enterprise engaged in the UK nuclear sector. We learn how to create a
taxonomy of skills and competencies and build a database for enterprise
sharing of this information. The case study furthermore provides
guidance on how to apply change management principles to gain buy-in
from all staff involved in the implementation and roll out.
17
Open Source CMS for a
Sports Centre
Arthur Visser
16
Identifying Key Skills and
Competencies Across the
Enterprise
Gary Cairns
Open Source Content Management System for a
Sports Centre
Arthur Visser
This case study presents an insight into how the content management
needs of a sports centre in the Netherlands could be satisfied through the
implementation of an open source content management system. We learn
of the key benefits and pitfalls involved in implementing a content
management system for sports centres (or even SMEs). The case study
also discusses the differences and financial implications between
implementing an open source content management system and a
proprietary one.
Virtual Collaboration in the Brazilian Mould and Die
Making Industry
Rolando Vargas Vallejos and Patricia Wolf
This case study discusses how competitors in the Brazilian die and mould
making industry tried to establish stable relationships through the concept
of virtual organisation breeding environments. The case study informs
how through joining forces it was possible to have access to new
technologies, to share information and knowledge, to qualify employees,
to contract research projects and consultancy work. We learn that through
the establishment of such collaborative networks, new co-created
innovations may emerge.
18
Virtual collaboration in
the Brazilian Mould and
Die Making Industry
Rolando Vargas Vallejos
Patricia Wolf
xv
Preface
Acknowledgements
First and foremost, we would like to acknowledge and appreciate the enthusiasm and contributions
from the numerous authors who have shared their experiences and lessons learned in this book. This
book would not have been possible had it not been for them. We are grateful to Ron Dvir who
voluntarily took up the task to add a visual touch to this book by engaging the artistic skills of his
father Ayre Dvir who has done wonders with the artwork in this book. We would also like to
acknowledge the support of the European Commission and in particular its IST programme for their
financial support for KnowledgeBoard.
We would like to thank you, the reader, for taking the initiative and time to explore and learn from
the vast experience presented in the case studies in this book. We certainly believe that this will lead
to new opportunities for the co-creation and sharing of knowledge.
Ed Mitchell, the outgoing editor of KnowledgeBoard has been one of the main driving forces
behind this book project. His constant probing and bubbling enthusiasm has provided an extra level
of motivation in preparing this book. It is to Ed, that we humbly dedicate this book and look
forward to remaining in his circle of friends.
Dr. Abdul Samad (Sami) Kazi, VTT – Technical Research Centre of Finland
Dr. Patricia Wolf, Hochschule für Wirtschaft (HSW) Luzern, Switzerland
KnowledgeBoard, April 2006.
xvi
Dedicated to
Ed Mitchell
(outgoing editor of KnowledgeBoard)
for his passion, enthusiasm,
and jolly nature that help
foster communication
and interactive dialogue.
Industrial Experiences
and Communities
of Practice
1
Peter Troxler
Kristina Lauche
Dr. sc. techn. Peter Troxler, dipl. Ing. ETH VDI, [ k n w l d g ] ([email protected])
Dr. Kristina Lauche, Dipl.-Psych., TU Delft, Faculty of Industrial Design Engineering ([email protected])
Abstract
Communities of Practice have become almost a standard answer to
the quest of organisations to foster learning, sharing and development
of knowledge. However, they still pose the classic challenges of
Knowledge Management of finding the middle ground between
technology-driven and people-driven approaches, between systematic
solutions and mere fads, and between forced and self-directed
participation.
This chapter reports how three companies in the same industry
implemented Communities of Practice. The examples show three
fundamentally different approaches to the challenge, each with its own
strengths and weaknesses.
A meta-analysis of discussions among Knowledge Management
practitioners and academics serves to contextualise the three
examples and to develop a set of issues to be considered and
questions to be addressed when implementing Communities of
Practice. The chapter concludes that ‘
implementing Communities of
Practice is all about designing practice’
. It drafts a design strategy
based on the assertion that Communities of Practice ‘
have to dance
between freedom and purpose’ and have to be allowed to (re-)
negotiate their direction and circumstances.
Keywords: Expert Systems, Communities of Practice, Current Trends and Developments, Case
Study, Oil and Gas Industry
Background
Organisations are keen to create a culture that furthers learning from past experience, exchanging
learning among employees, and developing the knowledge gained. Lave and Wenger found such
cultures when describing learning and apprenticeship in groups as diverse as meat cutters,
midwives, tailors, naval quartermasters and non-drinking alcoholics (Lave and Wenger, 1991).
Their (initially descriptive) concept of Communities of Practice soon became more of a tool for
organisational design (Wenger, McDermott and Snyder, 2002; Wenger, 1998) and the seemingly
number one answer to the question of learning, exchanging and developing knowledge in
organisations.
Industrial Context
The studies reported in this chapter are all situated in the energy industry, and more specifically in
larger engineering service providers to the oil and gas industry. These companies operate globally
as contractors to major operators. They employ up to several ten thousands of employees in
multiple locations around the world. Multiple locations do not only refer to geographical separation
7
but also to working for different clients, often in the clients’offices. Engineering services comprise
of both, turnkey solutions that involve a suite of engineering services to provide a comprehensive
solution to the customer from design to project management and including fabrication of equipment
and integration and commissioning, and customized services to meet specific needs of a client.
Projects are typically secured through a competitive tendering process. Organisational
characteristics include flat structures, regular reorganisations, and a high degree of specialisation.
In terms of number of employees, the oil and gas industry is actually quite small. Therefore there is
an unwritten rule, particularly among contractors, that “you don't burn bridges”in this industry. It is
very likely that the same people and organisations will have to work together again on a later
project. This creates a certain atmosphere of familiarity in the industry, which is quite unique. A
second relevant characteristic is the cyclical nature of the petroleum market which companies
readily follow. This leads to a hire-and-fire climate in the industry forcing employees in and out of
the business, offering no steady lifelong opportunity. Thirdly, the workforce is rapidly aging, in a
world steadily demanding more oil and gas, and the industry has not been too successful in
attracting new employees.
In many regions of the world, the oil and gas industry has reached a high level of maturity with
experience in exploration, production and processing gained over decades. The industry is heavily
knowledge based; it has developed and applied systems, policies and practices to manage this
knowledge over at least two decades. The oil and gas industry has led the way in Knowledge
Management for many years, and operators like Shell or BP are often referred to as flagship
companies when it comes to the development and application of Knowledge Management in
general and the idea of Communities of Practice in order to foster learning and to improve the
exchange and development of knowledge.
For service providers, however, it is often not only the inherent benefits of learning or of the
exchange and development of knowledge that alerted companies to devote more attention to
Knowledge Management, but the fact that competitors who started Knowledge Management
activities earlier are able to offer their services at lower cost. One informant for example explained
that the trigger for their new approach was ‘an incident that involved the company losing an
international engineering project to another company.’ The alarming surprise was that the
competitor bid for half the price without underselling their performance.
The three case studies in this chapter investigate individual engineering companies and their
approach to Knowledge Management. All of them work in the same industry –the upstream oil and
gas industry –yet they cover different specialties.
Problem
The discipline of Knowledge Management has since its inception in 1986 (Wiig, 1986; Wiig, 1988)
been plagued not only by the constant rivalry of technical and people approaches, but by also the
quest to help, cause or oblige knowledge workers to share and develop their knowledge (Drucker,
1954). To foster knowledge sharing, Knowledge Management employs technical, organisational or
motivational means, such as mandatory meta-data capture, over-instrumentalised descriptive
concepts borrowed from disciplines such as ethnography, or even simply cheap awards. Questions
of control and ownership of knowledge are largely left out in discussions; an assumed business
rationality seems to presuppose that it is right that employees add their very knowledge as a surplus
value to the ‘intellectual capital’claimed to be owned by the corporation and its share-holders and
that they thus eventually contribute to their own alienation and self-exploitation.
The problem in Knowledge Management practice is obviously finding the middle ground between
technology and people, between fads and methods, between forced and self-directed participation.
8
The three case studies and the discussions of Communities of Practice among practitioners and
academics illustrate this drive for best practical solutions.
Learning Objectives:
•
Recognise that one-dimensional, tactical implementations of Communities of
Practice fail to deliver all potential benefits.
•
Understand that one purpose of Communities of Practice is to build capacities
and that therefore their implementation needs a strategic approach
•
Learn about the various characteristics of Communities of Practice that have to
be addressed when designing communities.
•
Appreciate that Communities of Practice can contribute to participation as well
as to establish and reinforce control.
•
Acknowledge that Communities of Practice need to be given the freedom to
(re-)negotiate their own purpose.
Approach
This chapter compares Knowledge Management strategies and their implementations found in three
engineering companies in the upstream oil and gas industry. The three companies, however, have
each a different approach to systems, policies and practices of Knowledge Management; they
represent three exemplar approaches.
The first case study was set in a company that developed a technology-driven Knowledge
Management approach using knowledge engineering methods and solutions. Consequently, their
Knowledge Management system was composed of a rule-based expert system and three knowledge
bases that were linked together, providing easy access to the most valuable knowledge to the
engineer workforce worldwide and linking engineering-related and other knowledge. Yet the
company showed little to no recognition of the people aspects of Knowledge Management beyond
knowledge elicitation and distribution.
The second case study was based in a company that focused on the implementation of Communities
of Practices as an extension to existing organisational structures. Initiated by senior management as
a response to outside challenge, it built on experience from other subsidiaries of the company.
Previously dominated by information management, the new approach embraced ‘people to
solutions’as well as ‘people to people’paradigms. New communities were initiated and supported
by central support staff; their tools were embedded in a web-based portal environment.
The third case study showed a yet another approach in which Communities of Practice again played
a central role when the company’s former, database driven Knowledge Management system needed
replacement. However this company did go one step further in an attempt to substitute all existing
organisational structures such as departments and locally based competence centres by
Communities of Practice that operated and communicated on a global scale, supported mainly by
Internet and telecommunications technology.
The chapter then contrasts the three cases with key issues from the ongoing discussion among
Knowledge Management practitioners and academics on the topic of ‘Communities of Practice’,
particularly on the use of technology, the multiple design decisions required when implementing
Communities of Practice, the questions of failure and stimulation of communities, and the dilemma
of contrived vs. natural communities.
9
Case Analysis
Case 1: Global Oilfield Services Provider
The first case study was set in the context of a global oilfield services provider that has its
headquarters in Houston, Texas. It employs about 25,000 people in over 90 countries. The actual
case study was located in one of their divisions, which is highly specialized in one crucial step of
upstream oil production and has its global operations base in Aberdeen, Scotland.
This specialist field was not only relatively new but it also involved a wide set of skills from
mechanical engineering, geology, physics, and other disciplines. Therefore, the community of
skilled engineers was small. The company was operating worldwide. Its engineers were allocated,
on a flexible basis and as business demanded it, to either of the two major sections responsible for
the western and eastern hemisphere respectively. These engineers shared all the characteristics of a
work-related Community of Practice.
To streamline operations, to share past experience and to avoid running into similar problems twice,
the company developed a Knowledge Management approach that was heavily reliant on database
technology. The Knowledge Management system was composed of a rule-based expert system and
three knowledge bases that were linked together – a set of performance guidelines, an online
training system, and a repository of lessons learnt.
Figure 1: Development of Knowledge Management and Communities of Practice (case 1):
Triggered by working constellations and early success with an expert system a technologyheavy approach was pursued with little user involvement and little consideration for adequate
employee training.
10
The repository had been built using a knowledge engineering approach that included semantic
modelling of the knowledge domain (Preece, Flett, Sleeman, Curry, Meany and Perry, 2001). Then,
intensive knowledge acquisition campaigns were conducted with a small number of expert
engineers. The knowledge acquired from each campaign was formalised in the semantic model, but
also written-up in a natural-language knowledge book that was used to validate the accuracy of the
knowledge captured and then to distribute the knowledge throughout the company. Later, the
repository was ported to the company’s database driven intranet and extended with the performance
guidelines and the online training system. A central support team maintained the whole Knowledge
Management system.
The three knowledge bases were operated as knowledge storage and retrieval systems. They were
regularly updated with lessons learnt from project reviews, with general reviews of the project
planning and execution process (primary process). Occasionally the Knowledge Management
support team carried out knowledge acquisition campaigns to gather in-depth knowledge e.g. from
strategically important projects or across all projects for a certain client.
The knowledge bases themselves did not have the capability of making suggestions based on stored
knowledge. Actions were entirely left to the engineers. Still, they were designed to provide all
engineers with easy access to the most valuable knowledge. Particularly they allowed easy linking
between engineering-related and other knowledge from the projects.
We carried out a series of qualitative interviews with key engineering staff, a domain expert and a
knowledge engineer, and a web-based survey with 39 engineers to analyse the use of the knowledge
bases. The rule-based expert system was excluded from that study since it was only used by a
couple of engineers at their central office. The study investigated frequency and ease of use of the
knowledge-based systems and general user satisfaction. It showed that the systems were not being
used to their full potential.
The performance guidelines were the part of the Knowledge Management System used most. 90 %
of users reported to be confident using that part of the system. Particularly more experienced users
felt the guidelines were useful to their job.
The repository of lessons learnt was used at a similar level as the guidelines. About half of the users
felt the lessons learnt were reasonably useful, and they reported a good, very good or excellent
satisfaction with the system. However, the introductory training did not enable users to report new
cases without central support. This central support was not always available due to lacking
resources.
The online training system was the least used of the three systems. Additionally user satisfaction
was lower: the training system relied heavily on the involvement of tutors. Users complained that
tutors were slow to respond, if at all, to the learners’coursework. They also reported having not
enough time to complete the coursework.
Overall, the users reported easy access to and successful re-use of lessons learnt across geographical
areas and across clients.
Case 2: Engineering Contractor
The second case study was situated at an engineering contractor that delivers design, construction,
maintenance and operation services to the oil and gas and other industries in over 100 countries.
The company employs about 85,000 people. Its headquarters are located in Houston, Texas. The
case study itself is situated in one of their main European bases in Aberdeen, Scotland, which is part
of their midstream and downstream operating segments. This business unit focused on the local
implementation of Communities of Practices, initiated by senior management.
11
Introducing Communities of Practice had been a response to outside challenge, ‘an incident that
involved the company losing an international engineering project to another company’, as an
informant put it. The alarming surprise was that the competitor bid for half the price without selling
their performance under price. The initiative mainly incorporated experience from other subsidiaries
of the company. Previously, their focus had been on information management. The Communities of
Practice approach embraced new paradigms, such as ‘people to solutions’and ‘people to people’.
The Communities of Practice approach was very much driven by a central support team who
initiated and supported new communities. Within half a year the number of communities quickly
rose to 23, with topics ranging form electrical engineering to human resources, performance
measurement and new graduates.
The central piece of communication technology was a web portal, which included a ‘skill-finder’
database, a tool for threaded discussions, and any documents and tools or gadgets the communities
choose to have on their portal. The intention was to replace peer-to-peer email habits with one
central communication hub for a range of tasks.
Figure 2: Development of Knowledge Management and Communities of Practice (case 2):
From existing communities the idea was spread as a local initiative, making use of a generic
portal product and systematic training for the champions who were in charge of leading a
individual Community of Practice.
Each community had a champion who promoted it and facilitated the internal discussion.
Champions received a half-day training course, which provided background understanding into
Communities of Practice and stages of their development, an update of the technological features of
the web portal and a brief introduction to facilitation skills. Employees were encouraged to join
communities as they wished; all were open to anyone with access to the company’s intranet.
Individuals determined the content of their skill profile and updated it themselves.
12
We carried out semi-structured interviews with eleven of the 23 champions and issued a web-based
questionnaire to hundred members (89 returned) and hundred non-members of communities (26
returned). In the interviews, the champions reported a healthy mixture of users but a slow take-up of
the idea of communities, and a number of usability issues of the web portal. Also, some voiced
concerns over the long-term management support such as ‘other initiatives have fallen due to
funding being withdrawn’or problems of ‘getting my boss interested’.
The most active communities were those that already existed as professional groups with
established social structures and regular meetings, and company-wide initiatives, which utilised the
portal as a new communication channel. To our surprise, most interviewees perceived the rationale
for Communities of Practice and the benefits so far as mainly local. A training day for community
leaders and the experience of using the portal had began to raise awareness for the benefit of
breaking down geographical boundaries to develop communities in a global context.
The questionnaire data showed that 83.2 % of the respondents saw Knowledge Management as a
priority and the majority was reasonably satisfied with how problems of knowledge transfer were
addressed. However, 39.3 % felt it could be improved. The portal technology was evaluated as fit
for use but employees felt there was too much focus on virtual meetings and not a strong enough
vision of the future.
Most participants showed an acceptable understanding of Knowledge Management, mainly centred
on sharing of knowledge; there was no statistical difference between members and non-members.
We did find statistically significant differences for the expected duration and the integration of
Communities of Practice into the culture of the company: Members were more positive towards the
integration into the culture, and expected Communities of Practice to last for five to ten years.
Case 3: International Project Management
The third company also delivers international project management and services to a multitude of
industries. As such it is a direct competitor to the company in case 2. The company has its global
headquarters in London. Worldwide, it employs over 44,000 people in about 40 countries. The case
study is located in their Oil & Gas office that caters for North Sea and worldwide oil and gas
installations.
The company had had a formal Knowledge Management system for six years prior to our study.
This system was essentially a database that could be accessed via the Internet by customers and
employees alike. Additionally, there existed a few Communities of Practice, albeit in an informal
manner.
In 2002 the company implemented a new, global Knowledge Management strategy, which adopted
a completely new organisational structure based on a very formal Communities of Practice
approach. This new structure that went across countries, business units and industries effectively
abolished the former, departmental structure of the organisation and regrouped everybody into
Communities of Practice.
The communication between community members took place mainly via the company’s web-based
portal in so-called ‘e-rooms’. Employees could log on to these ‘e-rooms’and communicate by
email and chat with colleagues around the world. Still community members also communicated on
a more personal network, via telephone and in occasional meetings. Directors or department
managers were appointed to manage each Community of Practice and each community was given
an annual budget and was responsible for its allocation, alongside the hiring and firing of
employees.
13
Figure 3: Development of Knowledge Management and Communities of Practice (case 3): An
existing Knowledge Management approach (informal groups, Internet-enabled database) was
replaced by a global Communities of Practice approach across countries, business units and
industries where Communities of Practice replace the more traditional departments as
organisational units.
The case study we carried out was designed to help the company to reinvigorate their Communities
of Practice. Of particular interest was to study if and how the Communities of Practice worked,
what attitudes employees held towards them, whether they were viewed as problematic by
community members or leaders, how they compared to past Knowledge Management programmes
and what problems or advantages they held for communication throughout the company.
The study was carried out using a web-based questionnaire that was distributed via the community
leaders to which 121 employees responded. These included a cross section of employees in each of
the organisation’s communities, plus the community leaders themselves.
The results of the study showed that the majority of employees found the information they gained
from the communities reliable. Communities met typically daily, and employees indicated they used
their community portal for about an hour a month, mainly for networking, knowledge sharing and
problem solving, often as the first port of call when a problem arises. The majority of employees
rated the community portal as more effective for obtaining information, compared to individual
methods.
However, it was obvious from some of the answers that a small but significant number of
employees had no knowledge of the community portal. The overwhelming majority of employees
were not subscribed to any e-room, and even more worryingly, a number of employees reported in
the open-ended questions that they had never heard of e-rooms and did not know what they were.
Furthermore, the small number of employees who did subscribe virtually never used e-rooms, and
stated that the training available for e-rooms was very inadequate.
14
Overall employees had a positive opinion towards knowledge sharing, however they felt that too
much emphasis was placed on e-mail and found that virtual communication was very impersonal.
Finally, the majority felt that problems with knowledge transfer were not being effectively dealt
with in the organisation, and over a third stated that the methods for obtaining information needed
to be improved.
Overall, this study suggested that, on the whole, the Communities of Practice strategy was being
accepted by the employees and functioned well on a variety of levels, although there seemed to be a
potential for the company to introduce global standards across the communities, ensuring that time
and money was spent equally in developing all of the communities so they could be equally as
effective.
As the Communities of Practice had been enforced rather than developed organically, educating of
employees, promoting of best practices and establishing organisational global consistency seemed
key issues for the company. On top of this, the technology supporting the communities needed to
become part of people’s daily routines and habits, ensuring all employees are involved regularly. A
problem seemed to lie in the technology itself and it was suggested to re-evaluate the technology
used.
Practitioners’Issues
The three companies, although belonging to the same industry, showed fundamentally different
approaches to the problem of finding a ‘middle ground’ in Knowledge Management practice.
Before we attempt to distil challenges and best practices from these examples, we wanted to
validate whether the particular issues and approaches resonate with the general debate on the
implementation of Communities of Practice. For this purpose, we carried out a meta-analysis of
discussions by Knowledge Management practitioners and academics, held in an online special
interest group on Communities of Practice, initiated by KnowledgeBoard’s Diane Moult and
Miguel Cornejo (Moult and Cornejo, 2005).
Figure 4: Number of replies and reads per discussions topic in an online special interest group
on Communities of Practice by Knowledge Management practitioners and academics.
15
The online discussions took place over three years and covered topics in areas such as introduction
of Communities of Practice (21 discussions), benefit of communities (9 discussions), technology
support (7 discussions), definitions of ‘Community of Practice’(7 discussions), industry specific
issues (3 discussions), topics inherent to academia, such as calls for conferences etc. (9 discussions),
and a variety of more divers subjects (19 discussions). From all these discussions, six attracted a
high number of reads and replies (see figure 1). These six discussions will be reported below.
Discussion number 1 started out comparing email, online forum technology and weblogs as means
of supporting knowledge transmission, and it also listed other Internet technology that could be
used to support communities. The discussion highlighted in several posts that email was seen as
easy to use, quick, light, and needed minimum connection time, whereas forums and similar tools
were said to be more expensive. An allusion to individual-centric and organisation-centric aspects
of support technologies was not explored fully.
Discussion number 2 looked at another tool to support communities, namely wikis (a website that
allows people to post and edit content and keeps track of the changes). It mainly listed the various
possibilities of this type of technology and pointed to a number of examples, with people also
exposing their own experience in using wikis, e.g. in a technical sales team ‘as a way of capturing
the bang-up-to-date status of configurations and specifications for the systems’(Duckett, 2005), or
in architectural and engineering design practices. The general opinion in this discussion, however,
seemed to be that ‘larger rims with more hierarchical cultures seem to find the idea of giving
everyone write access rather frightening’(Bartholomew, 2005).
Discussion number 3 was an attempt to define several types of Communities of Practice, initially
triggered by an APQC report on the subject (Leavitt, Hasanali, Lemons, Peltier, O'Dell, Hubert and
McDermott, 2000). The report suggested four types of communities: Knowledge Stewarding
Community, Helping Community, Best Practice Community, and Innovation Community. Other
distinctive parameters were suggested, such as size; emergent vs. management initiated (or natural
vs. contrived); formality (formal vs. informal); border (loose, including externals vs. tight); virtual
vs. real (meaning mediated communication vs. face to face meetings); national vs. international;
local vs. distributed; volunteer vs. mercenary participants; influence, power and corporate
integration vs. informal peer-focused collaboration; core or marginal status. While such a discussion
could not lead to a formal conclusion, it highlighted the multiple aspects of Communities of
Practice that would have to be considered when designing effective communities that ‘have to
dance between freedom and purpose’(Ash, 2004).
Discussion number 4 evolved around the topic of failed Communities of Practice and the failure of
certain practices or methods in the context of communities. There was a general agreement that
learning from failures would be a valuable exercise and although ‘the conference circuit and
publishing world sometimes allows spin doctoring to flourish’ (Chaundy, 2004), some actual
failures were also being reported. However, many of the posts were a mere expression of frustration
about ‘hierarchy based on age, nepotism, authority’(Gioacchini, 2004), the ‘fight for old privileges,
(Gioacchini, 2004) or the feeling of a ‘lack [of] a real KM culture’(Guglielman, 2004).
Discussion number 5 focused on the question how participation in Communities of Practice could
be stimulated. Suggested methods included posting an interesting subject, holding kick-off
meetings, changing the leadership, getting a moderator to post news, setting up events or projects. A
recurring theme in the discussion was that ‘the real reason to participate is having some good reason
to do it’(Cornejo, 2004), i.e. creating a good (business) reason to use the Community of Practice,
and eliminating obstacles to doing so.
Discussion number 6, finally, discussed quite controversially contrived vs. natural communities.
The discussion started with the assertion that ‘efforts to establish what I call “contrived CoPs”(i.e.
sponsored, supported by and accountable to management) will, for the most part, never yield the
same richness found in naturally occurring CoPs’(Nickols, 2002) which generated some agreement
(‘a “forced CoP”would be indeed, a contradiction in terms’(Por, 2002)) and some disagreement
16
(e.g. quoting Etienne Wenger ‘Whether a community starts spontaneously or is seeded by the
organization, its success depends on the initiative of people who have a connection to the domain,
recognize the potential for the community, and are in a position to build on it.’(Por, 2002)), and
even denial (‘I don’t believe spontaneous CoPs exist. At all.’(Cornejo, 2002)). The discussion then
continued on the relationship between a community and ‘management’or ‘the organisation’and in
what way Communities of Practice could be an instrument to attain tangible business goals. One
participant, himself a former senior executive posted, ‘that performance is about capability and that
knowledge is about building capability’ (Saint-Onge, 2002). He suggested building a
‘comprehensive knowledge strategy that will clearly contribute to building higher levels of
sustainable performance’(Saint-Onge, 2002). Rather than imposing business objectives on the
communities being formed, but Saint-Onge believed ‘that an organization can sponsor communities
in an intelligent way’by focusing on the purpose of capacity building, by establishing a charter for
each community, by supporting the communities with regard to their purpose and purview, and by
‘keep[ing] managers from taking over these communities and convert them as instruments of their
interest’(Saint-Onge, 2002) since these ‘feudal landlords of the modern organization’(Saint-Onge,
2002) had to learn how to facilitate the horizontal, knowledge-based ‘conductivity’ of the
organisation, not the traditional, vertical transactions. Another contributor, Richard McDermott
confirmed this necessity to ‘keep the long-term capability building vision present’ instead of
diverting the community into a short-term goal oriented task-force. Further, he suggested to ‘let the
community shape its overall direction (even if it is a negotiation with management)’(McDermott,
2002) and thus to strike a balance between the ‘contrived’and ‘natural’poles set out at the start of
the discussion.
Key Findings
The first case study is a salient example for an approach that lacks consideration of the sociotechnical character of a Knowledge Management system (Coakes, 2002) both for implementation
and use. Learning occurred only slowly because sufficient resources had not been allocated.
However, the company could benefit from shared knowledge through the use of this system.
The second case study can be seen as typical for the early stage of implementing Communities of
Practice. It also is a representative of a ‘me-too’strategy in which Communities of Practice were
introduced mainly because many of the competitors were doing it. Therefore, it does not come as a
surprise that those communities that in actual fact existed before the introduction of the paradigm
performed best, alongside with company-wide initiatives that mainly benefited from electronically
supported communication.
The third case study is particularly interesting since the company decided to use Communities of
Practice as the main shaping principle of its organisational structure. Communities therefore
became not only networks of learning but the basic founding blocks of managing the company.
Alarmingly, the workforce did not seem to embrace this shift of paradigm as wholeheartedly as
management expected, and even some of the respondents in our study quite frankly negated to be
aware of some of the building blocks of that paradigm.
The technology-driven approach to Knowledge Management in the first case achieved sharing of
knowledge but encountered problems with people not being able to contribute to the system as
planned initially. Learning only took place slowly because of lack of sufficient resources. However,
the company could benefit from shared knowledge through the use of this system.
The community-driven approach supported by portal technology in the second case achieves
knowledge sharing but encounters problems with technology and the transfer from teams to the
17
organisation as a whole. It supports best existing communities and company-wide, global
initiatives.
The radical approach of the third case study seems prone to fail on two fronts. Firstly any
organisational change bears the risk of being resisted and delayed. More importantly the instrument
of Communities of Practice is probably not suited to deliver on all fronts or for all issues of
managing an organisation.
The early Knowledge Management initiatives as presented in the first case were mainly technology
driven but failed to capture people's tacit knowledge. So the more recent initiatives in Knowledge
Management as presented in the second case are directed at behaviour and culture. This second
approach relies strongly on the concept of Communities of Practice, originally derived from
anthropological studies of how communities share and pass on expertise. By participating in a
‘communal memory’, the individual can do their job without having to remember everything
themselves.
When the concept of Communities of Practice was introduced, it was intended to conceptualise
already existing organisational practice (Lave and Wenger, 1991). Turning it into a management
approach to improve Knowledge Management, however, meant that the people had to be
encouraged to participate in what initially was voluntary and spontaneous. This poses some
fundamental questions: To what ends does an organisation apply the concept of Communities of
Practice? Does this increase renewal, participation and democracy in the organisation? Is the
concept applied to establish and reinforce control?
Business Impacts
There are three core messages that seem to emerge from the case studies and the discussions
reported in this chapter:
1. Institutionalised Communities of Practice have become a business reality. However, there are
numerous questions that have to be addressed when implementing Communities of Practice,
ranging from their strategic purpose to their practical relevance and the various issues related to
their day-to-day running.
2. Technology to support the interaction particularly of distributed communities is equally a fact. As
with the concept of Communities of Practice itself, technology selection and implementation has to
consider strategic relevance as well as the practicalities of the every-day use and how technology
fits with the organisational and individual practice and the habits of its users.
3. The evidence from our case studies and the reflections of experienced practitioners leads us to
conclude, that the implementation of Communities of Practice is all about designing practice.
Practice here means and includes directed activities of transforming tangible or intangible objects
into an outcome by the application of tools; practice is directed by motives (i.e. a business purpose,
a strategy, but also personal motives); and practice is situated in a social context with rules and a
certain division of labour. Essential, it seems, is the notion that this whole activity system and its
part need to be (re-)negotiable to a certain degree in order to sustain its ability of capacity building.
Conclusions
A lot of research and discussion has gone into the development of Communities of Practice, since it
was popularised as a tool for interaction and communication in an organisational environment. Onedimensional approaches to the introduction of Communities of Practice do not seem to deliver the
full benefits and performance the concept is believed to be able to deliver. However there are
numerous success stories of Communities of Practice, if they are not, as sometimes suggested, the
18
product of the conference circuit and publishing world. As Communities of Practice ‘have to dance
between freedom and purpose’to sustain their purpose of capacity building, it appears relevant to
allow them to (re-)negotiable their direction and circumstances. However, more theoretical
conceptualisation and empirical research is needed to fully explore this aspect of Communities of
Practice and their inception and management.
Practical Tips and Key Lessons:
•
One-dimensional, tactical implementations of Communities of Practice don’
t
lead to immediate failure. But they are not likely to deliver the full results and
business impact. This means: If there is an immediate benefit of people
working together in a Community of Practice it is worthwhile to explore this
instrument. To leverage their real benefit, the introduction of Communities of
Practice needs to look at long-term impact, too.
•
Implementation of Communities of Practice needs a strategic approach that is
aimed to build capacities as an essential prerequisite to boost performance in
organisations.
This means: The long-term effect of Communities of practice is that a company
builds capacities –such as information exchanges stored in computer systems,
experience shared between employees, and the culture and organisational
ability to draw on information, knowledge and experience across the whole
workforce. These capacities will lead, eventually, to better performance.
•
Communities of Practice are shaped by a number of characteristics: intent;
size; formality; volunteer vs. mercenary participants; influence; power and
corporate integration vs. informal peer-focused collaboration; core or marginal
status, to name a few. Creating a Community of Practice requires conscious
decisions about these ‘
dilemmas’
.
This means: There is no single recipe for “
the right form and shape”of
Communities of Practice. The historical, commercial and cultural context of an
organisation and the purpose of Communities of Practice need to be taken into
account when deciding on the appropriate format.
•
Implementation of Communities of Practice has to address the questions: To
what ends does an organisation apply the concept of Communities of Practice?
Does this increase renewal, participation and democracy in the organisation? Is
the concept applied to establish and reinforce control?
This means: Communities of Practice can be used to control employees. Or they
can be used to give them the chance to participate in the development of a
company. Either way, when introducing Communities of Practice, one has to be
clear what is intended to avoid creating false hopes among participants.
•
Communities of Practice ‘
have to dance between freedom and purpose’
. Part of
this dance is for Communities of Practice to have the freedom to (re-)negotiate
their own purpose.
This means: If Communities of Practice are too focused on a narrow purpose
they are likely to fail to deliver beyond those boundaries defined initially. As a
business and its environment change and as Communities of Practice mature,
they need, at least, to readjust their purpose. Moreover, Communities of
Practice have the inherent potential to “
think beyond boundaries”and to
instigate creativity and innovation –if they are allowed to do so.
19
Acknowledgements
We would like to acknowledge the support of Elaine Blows, Lyndsey-Anne Borrowman and Joanne Henrietty,
who collected data for the individual case studies as part of their degree studies at Aberdeen University. We
are also grateful to our contact persons in each of the companies and all participants for their time and
support.
References1
Ash, Jerry, 2004, Re. Different Types of CoPs. Available online at
http://www.knowledgeboard.com/cgi-bin/forum.cgi?comment=2155, last accessed
30.08.2005.
Bartholomew, David, 2005, Experience with wikis. Available online at
31.08.2005.
Chaundy, Claire, 2004, Failed KM practices. Available online at
30.08.2005.
Coakes, Elayne, 2002, Knowledge Management: A sociotechnical perspective, in Elayne Coakes,
Dianne Willis, and Steve Clarke, eds.: Knowledge Management in the SocioTechnical world
(Springer, Berlin).
Cornejo, Miguel, 2002, Re: "contrived" vs. natural communities. Available online at
http://www.knowledgeboard.com/cgi-bin/forum.cgi?comment=687, last accessed 30.08.2005.
Cornejo, Miguel, 2004, Something to discuss about. Available online at
30.08.2005.
Drucker, Peter F., 1954. The Practice of Management (Harpers and Row, New York,).
Duckett, Tim, 2005, A very effective tool. Available online at
http://www.knowledgeboard.com/cgi-bin/forum.cgi?&comment=2779, last accessed
30.08.2005.
Gioacchini, Raniera, 2004, it's a fight for old privileges. Available online at
Gioacchini, Raniera, 2004, oral vs. written. Available online at
30.08.2005.
Guglielman, Eleonora, 2004, Italian organizations and KM culture. Available online at
30.08.2005.
Lave, Jean, and Etienne Wenger, 1991. Situated Learning: Legitimate Peripheral Participation
(Cambridge University Press, Cambridge).
Leavitt, Paige, Farida Hasanali, Darcy Lemons, Neil Peltier, Carla O'Dell, Cindy Hubert, and
Richard McDermott, 2000, Building and Sustaining Communities of Practice, (American
1
Note: due to a relaunch of the KnowledgeBoard website, all online references at http://www.knowledgeboard.com can
now be found at http://forum1.knowledgeboard.com by simply replacing the leading www by forum1 in the addresses
given in the list of references.
20
Productivity & Quality Center (APQC)). Available online at
http://www.apqc.org/portal/apqc/ksn?paf_gear_id=contentgearhome&paf_dm=full&pagesele
ct=detail&docid=100577, last accessed 31.08.2005.
McDermott, Richard, 2002, "contrived" vs. natural communities. Available online at
Moult, Diane Le, and Miguel Cornejo, 2005, KnowledgeBoard: the European Knowledge
Management (KM) Community - SIG - Communities of Practice, (Knowledge Board).
Available online at http://www.knowledgeboard.com/community/zones/sig/cp.html, last
accessed 14 September 2005.
Nickols, Fred, 2002, "contrived" vs. natural communities. Available online at
Por, George, 2002, Re: "contrived" vs. natural communities (1). Available online at
Por, George, 2002, Re: "contrived" vs. natural communities (2). Available online at
Preece, Alun, Alan Flett, Derek Sleeman, Daivd Curry, Nigel Meany, and Phil Perry, 2001, Better
Knowledge Management through Knowledge Engineering, IEEE Intelligent Systems 16, 3643.
Saint-Onge, Hubert, 2002, Re: "contrived" vs. natural communities (1). Available online at
Saint-Onge, Hubert, 2002, Re: "contrived" vs. natural communities (2). Available online at
Saint-Onge, Hubert, 2002, Re: contrived vs. natural communities (3). Available online at
Wenger, E., R. McDermott, and W. Snyder, 2002. Cultivating Communities of Practice: A guide to
managing knowledge (Harvard Business School Press, Boston).
Wenger, Etienne, 1998. Communities of Practice: Learning, Meaning, and Identity (Cambridge
University Press, Cambridge).
Wiig, Karl M., 1986, Management of Knowledge: Perspectives of a New Opportunity, Conference
on Expert Systems (ILO, Gottlieb Duttweiler Institute, Rüschlikon/Zürich).
Wiig, Karl M., 1988, Management of Knowledge: Perspectives of a new opportunity, in U.
Fellman, and T. Bernold, eds.: User Interfaces: Gateway or Bottleneck (North-Holland, New
York, Amsterdam).
21
Authors’Biographies
Peter Troxler works as a Knowledge Management expert with management
consultancies in Europe. He supports organisations in the private and public sector to
building management systems for the knowledge economy. Peter has worked in
academia at ETH Zurich, Switzerland, at the University of Applied Sciences Solothurn
Northwestern Switzerland, and at the University of Aberdeen, Scotland. His main
interests are cross-disciplinary issues at interface of industrial psychology, IT and
engineering, and management science. Peter Troxler holds a Dr. sc. techn. in
Industrial Management from ETH Zurich, Switzerland. He is a member of many
professional organisations, notably the Swiss Associoation of Industrial Psychologists
(SGAOP), the International Council of Systems Engineering (INCOSE), and the
German Association of Engineers (VDI). He is a member of and a regular contributor
to KnowledgeBoard, the online community of the European KM Forum.
Kristina Lauche is an Assistant Professor in Design Methodology at the Technical
University of Delft, the Netherlands. Her research focuses on team interaction in
complex work environments. Research topics include organisational change and
learning, product innovation and participatory systems design. Kristina received her
PhD in Psychology from the University of Potsdam, Germany. In her research career
she has worked at the University of Munich (Germany), ETH Zurich (Switzerland), the
University of Aberdeen (Scotland), and the University of Helsinki (Finland). She is a
member of the Society for Industrial and Organizational Psychology (SIOP), the
German Society of Psychology (DGPs), and the International Design Society.
22
2
The Knowledge
Management Journey of
Israel Aircraft Industry
Rony Dayan
Edna Pasher
Ron Dvir
The Knowledge Management
Journey of Israel Aircraft Industry
Rony Dayan, Israel Aircraft Industry ([email protected])
Edna Pasher, Edna Pasher and Associates ([email protected])
Ron Dvir, Innovation Ecology ([email protected])
Abstract
Firstly, we tell the story of the journey of Israel Aircraft Industries (IAI)
into Knowledge Management. We show how the plan was embedded
in the company’
s comprehensive change program which focused on
four core values: “Customer”, “People”, “Innovation & Technology”
and “One Company”. Knowledge Management has been originally
part of the “Innovation & Technology”value – but in time managed to
relate to, and support all four values. We then briefly describe the KM
plan of the company and its 12 chapters. Several organisational
mechanisms to support the implementation of the KM program are
discussed, including performance measurement, the KM handbook
and the distributed organisational structure of the program. We chose
to describe in details one of the modules of the program – Fostering
the knowledge of core Competence Centers. A systematic process to
define such centers is outlined, and a list of many different ways to
support and nourish such competence centers is proposed.
This case is concluded with a list of several tips which are especially
valuable for large and complex organisations that attempt to address
systematically the challenges of effective Knowledge Management.
For example, we elaborate on the need to balance well planned and
emergent initiatives, central and local resources, comprehensive
strategies and step-wise implementation.
Keywords: Knowledge Management, Competence Center, Community of Practice, Change, KM
Implementation, KM Measurement, Aerospace
Background
Industrial Context
It all started many years ago, before the Knowledge Management program in Israel Aircraft
Industry (IAI) began to take shape. At the beginning of our collaboration, Edna Pasher &
Associates consulted the IAI on the process of turning production and service units (such as the
training unit) into autonomous profit centers.
The concept of "Profit Centers", which helped to create a business culture, had a problematic side
effect: it also created internal competition. As a result of the pressure to show profit, the different
business units of IAI began to compete with one another, sometimes over the same customer, and
the culture that emerged was characterized by lack of knowledge sharing. It became clear that
something had to be done about this. IAI was still perceived as one united company in the market,
and the IAI had to overcome the lack of cooperation between the different units in order to improve
and stay competitive in the changing markets of the 21st century.
As a result of this emerging need for unity, in 1997 the head of the Electronic Group at IAI turned
to Edna Pasher & Associates as consultants, asking for ways to strengthen the subject of knowledge
25
sharing within the group. This will be further discussed in details. In addition, IAI started a Change
Program in February 2000 aiming to define and implement various actions to support the four
values established:
•
Customers
•
People
•
Innovation and technology
•
One Company
Knowledge Management was added as one of the actions of the "Innovation and Technology"
value.
Back in May of 1997, Edna Pasher & Associates1 met with the head of the Electronic Group. He
felt there was a need to find ways in which to encourage and accelerate an atmosphere of
cooperation and knowledge sharing among the 17 different "planets" of the group. The
recommendation made was to hold a series of "knowledge café" events. Knowledge cafés are
multiple-participants events in which a special technique is used to enable meaningful and fertile
conversations for the establishment of knowledge sharing and for the encouragement of innovative
group thinking. The events were advertised throughout the Electronic Group, and were opened to
any of the workers who wished to come.
The first event dealt with several topics, such as "the reciprocal relations between the company –
the group –the division", "intellectual capital management in the company" and others. Following
were seven other events, which focused on a variety of issues: starting with "ways for enlarging the
division's profitability", through "business intelligence" and "intellectual property". Meanwhile, as
mentioned above, the change program of IAI started to evolve and emerge. That, along with the
success gained from the "knowledge café" meetings, the IAI decided to invest in knowledge
management as a milestone in the company's cultural change.
In March 2001, IAI established a steering committee whose objective was to define the knowledge
management actions. It included experts in strategic thinking and processes, along with IT experts,
and experts from the field of industrial engineering. The committee employed a team of three
consultants (Edna Pasher & Associates, Ron Dvir of Innovation Ecology and Moria Levy of ROM
Knowledgeware) who began with a comprehensive diagnosis process. This process included a
survey of around 400 IAI employees, along with one hundred interviews of employees in numerous
occupations and various levels of management. The diagnosis was limited by the steering
committee to three subjects –engineering, production and maintenance.
The results of the diagnosis pointed at several issues of knowledge management that should be
addressed:
1
•
A gap was found between whom the employees cooperate with and whom they should be
cooperating with, in order to achieve the best results.
•
There was significant interest in communities of practice (CoPs).
•
There were “competence centers”at IAI that showed expertise performance in technologybased development or at giving specific kinds of service, and those "knowledge islands"
needed nurturing and investment.
•
Employees gave a lot of credit to the method of best practices.
With the contribution of Michal Goldberg
26
•
A need for a computation system that would enable better documenting and retrieving of
knowledge was discovered.
It took the team a couple of months to establish and to summarize all the information collected
during the extensive diagnosis process into a fully planned knowledge management strategy.
Beyond the three activities mentioned above, the steering committee initiated a pilot for knowledge
management implementation in one division, inclusive of all fields. However, opposed to the plan,
the Electronics group wanted to begin implementing a knowledge management program of its own
and unrelated to the company's effort.
Sounds like a problem? Not necessarily. We believe that almost all problems, including this one,
could be turned into opportunities. And so, following the activity of this unit, we used the situation
as a learning ground on what to do, as well as what to avoid, for the systematic implementation of
knowledge management. The lessons we learned from the pilot became a valuable part of our
diagnosis report, and later-on, of our strategic knowledge management plan.
Problem
This case presents several issues with which we dealt at the strategic level and at the
implementation level.
•
The most basic problem with which the whole project began, which we have already
described, was the lack of knowledge sharing within the company. That was the result of the
decentralized organizational structure of IAI. As a huge organization that includes over 30
units and around 14,000 employees, IAI is naturally susceptible to the lack of knowledge
sharing.
•
The biggest obstruction for knowledge sharing, as we learnt from the diagnosis process,
wasn't the perception of knowledge = power, as found in numerous organizations. The
obstruction of knowledge had to do with the "Not Invented Here" approach, which means:
"it came from another organizational unit/division, so it's probably not relevant to us".
•
Since IAI started a vast change process that involved the four values described above,
another challenge it faced was the integration of the knowledge management along with it.
The two programs, held within a two-year period, were in some aspects, addressing similar
values ("One Company", "Innovation", and others), but in a different way and through a
different focus.
•
Another challenge at IAI had to do with "knowledge dripping". Many of the company's
retiring employees were, after years of working, experts who held allot of valuable
knowledge, which they took with them when they left.
•
There was a problem of "re-inventing the wheel" found throughout the company. It was
even more meaningful due to the fact that the work at IAI is usually done in the form of
projects, typical of a vertical structure. We found that there was a need to improve the
amount of learning that was done from one project to another. Recently, IAI has been trying
to deal with this problem by managing its projects using the Integrated Product Team (IPT)
method.
•
Possibly the biggest challenge found was innovation. We discovered that although there
were strong innovative powers in the company, they were all local powers. A massive and
systematic emphasis on the encouragement of innovation was needed.
27
•
Integration the knowledge management program to the existing company
values, strategy and processes.
•
Establishing an implementation plan that combines the different aspects of
knowledge management into a coherent conceptual framework, which covers
the "life cycle" of knowledge.
•
Creating a system for monitoring and passing down the knowledge
management plan to all hierarchal levels.
•
Establishing a framework for measuring the progress and achievements of the
implementation process.
•
"Competence centers" as a procedure that encourages knowledge collection,
knowledge sharing and innovation.
•
Establishing "communities of practice" in a distributed company with separated
organizational units.
•
Creating a support structure.
Approach
The analysis of the IAI case is based on a “participants as observers”approach. One of the authors
was a member of the KM steering committee at the earlier phase of the project, and was then
nominated as the company Chief Knowledge Officer. The other two authors took an active part
throughout out the KM program creation and introduction process, as process and content
consultants, workshops facilitators and analyzers in the three parts of the case described in this
chapter, namely strategy formation, Communities of Practice building, and Competence Center
creation.
Conclusions were based on a relatively large number of “mini-cases” – about 15 CoPs and 38
Competence Centers that were explored.
For the analysis we used a large body of documentation:
•
The KM handbook.
•
A comprehensive set of documentation referring to the program and to its implementation
in IAI
•
The metrics related to KM, documented in the company overall implementation
management system –the Policy Deployment Model (PDM).
•
The detailed documentation and planning forms of each of the Competence Centers, based
on a standard template.
For the purpose of this case study, we focused our attention on a general analysis of the program
and on one specific application –the Competence Centers. The reason for choosing this particular
application out of 12 procedures covered in the IAI KM program was that this application surfaces
many of the human, cultural, business and technological challenges related to KM in a company
with the characteristics of IAI.
28
Case Analysis Part I: The Strategic KM Program
Knowledge Management as a Comprehensive Program
The Change program of IAI started in February 2000 with the definition and implementation of the
various actions backing up the four values established:
•
Customers
•
People
•
Innovation and technology
•
One company
Knowledge Management (KM) was established as one of the actions for the “Innovation and
technology” value. In fact Knowledge Management turned out to be part of other defined actions
within the program (as the process of identifying core competence centers within the various
divisions, the “Customer intimacy” action – intended to create a sense of closeness between the
customer and IAI employees at all levels, or the action to implement debriefing as a regular way of
proceeding after any meaningful event, created within the context of a “learning organization”)
(Dayan, 2003).
The purpose of Knowledge Management has been to foster innovation and to create and share
knowledge, in order to promote business goals. IAI, therefore defines knowledge management as
the process of identifying, capturing, leveraging and creating knowledge to deliver value to our
customers.
Various frameworks represent knowledge management implementation in different companies.
Some are minimizing and suffice with an Intranet having the task of sharing knowledge; some are
implementing sharing through the ever popular "Communities of Practice". We believe Knowledge
Management should be viewed as a comprehensive program including all phases in its life cycle, so
we have presented it for implementation divided into four chapters:
•
Knowledge Capture and Documentation
•
Knowledge Retrieval for Re-Use
•
New Knowledge Creation
•
Knowledge Sharing
The ultimate goal of the Knowledge Management action being to achieve a required Competitive
Advantage, we have phrased out for each chapter specific procedures to perform the required
activity and to measure their influence and effect on business results.
We have made our utmost to create within the implementation of the program an open mind for IAI
employees to view their own world as part of a much larger environment, including customers,
partners, and suppliers, within the context of the “extensive enterprise”, but with competitors as
well, to bring in Knowledge from the outside world and transform it into Knowledge based
products and services.
The program approved by IAI management includes the appointment of a full time Director of
Knowledge, who operates under the hospices of the Vice President for R&D and Strategy, and in
addition, the appointment of a part time Knowledge Manager for each and every division and
headquarter organization to initiate and coordinate the activities within the division (about 30
people all together).
Upon initiation, a steering committee was instituted with representatives from all IAI groups as well
as from the corporate organizations. The various tasks of the committee included the update of the
29
Knowledge Management strategy, setting priorities for the implementation of the program, and
controlling the program budget management.
The Knowledge Management Vision
New Knowledge
Will be created,
Shared,
And reused,
As part of IAI personnel’
s common practice,
For the achievement of IAI’
s vision
And the fulfillment of its goals
Figure 1: The Knowledge management vision
The Knowledge Management Chapters
This is the basis of Knowledge Management. Without it, there is no room for retrieval and re-use.
IAI employees gather a great deal of knowledge along their work –professional Knowledge coming
from experience, knowledge about processes, about projects, about the external environment
(customers, suppliers, opportunities), or about company products. In many cases, this knowledge
remains with the expert and is not available to others. Knowledge capture and documentation
enable the extraction of this knowledge to become a strategic asset by replacing tacit with explicit
and personal with organizational.
Retrieval and re-use of existing knowledge prevents us from reinventing the wheel time and again,
prevents us from reiterating previous mistakes and enables the duplication of successes; thus,
fostering professionalism, making processes more efficient, and diminishing development cost and
time to market.
In a time in which competition constantly increases and in which time, quality and budget are of
essence, knowledge creation and innovation are a vital necessity of any organization and the capture
and re-use of existing knowledge are no longer sufficient to cope with the rate of growth necessary
for the organization to stay ahead of the competition.
Knowledge Sharing
In a competitive market, in which time-to-market is constantly shortening and competition is about
alertness, price, innovation and professionalism, knowledge sharing within the organization is an
essential condition to its sustained success.
30
Implementation
The Knowledge managers convene once a month, each time in a different division, to get insight
about local successes and difficulties and also to discuss common issues related to the
implementation of the program. At one of the first of these conventions, the Knowledge managers
voted for a set of procedures to support the four above mentioned chapters and they are:
•
Critical Knowledge capture
•
Lessons learned extracted from debriefings and from day-to-day activity
•
Content management
•
Fostering the Knowledge of core competence centers
•
Knowledge in price proposals
•
Establishing a business Knowledge base
•
Establishing a technological Knowledge base
•
Knowledge extracted from the innovation process
•
Knowledge created along the new product initiative process
Knowledge Sharing
•
Communities of practice
•
Generating best practices
•
Using portals to share Knowledge
A general overview of the program is shown in Figure 2.
31
Content
management
Knowledge
capture and
documentation
Technological
data-base
Knowledge
retrieval
for reuse
Business
data-base
Knowledge
Capture
Lessons learned
Knowledge
in price
proposals
Project
Knowledge
New
knowledge
creation
Intranets
sharing
Knowledge
sharing
Good
Practices
Competence
Centers
Knowledge
from
Innovation
Communities
of
Practice
Figure 2: The program's chapter and procedures
The KM Handbook
Given the comprehensive nature of the Knowledge Management program, Knowledge managers are
required to carry out at least one procedure out of each of the above chapters.
For the purpose of having a standard set of procedures describing the various processes mentioned
above, a Knowledge Management handbook was written and published within IAI with the help of
Edna Pasher & Associates. Its purpose has been to commonly answer across IAI questions as to
why, what, who, how and when do we do anything to support the full life cycle of Knowledge
Management. Each and every procedure is described in details though the handbook as a whole
gives a complete picture of the program to enhance its comprehensive nature. For each procedure
we have included the following sections:
32
•
Introduction and definition of terms
•
The requirement
•
Purpose and goals
•
Process description
•
Technology and other resources backing
•
Organization
•
Measurements
•
Cultural content and training required
Monitoring
The process used by IAI for the management of its change program (inclusive of the Knowledge
Management program), is based on the Hoshin Kanri method 2 (Kondo 1998). Hoshin Kanri is a
Japanese technique for deploying company strategy down to an individual’s annual objectives. Thus
a logical link is made between the CEO's intent and the individual daily actions. Hoshin Kanri can
be literally translated as Policy Management, but in the anglicized world, “Management” has
become “Deployment” (Palmer 2004). The Hoshin Kanri process is first, a systematic planning
methodology for defining long-range key entity objectives. These objectives are planned for the
following year and do not change within the year. Second, the Hoshin Kanri process does not lose
sight of the day-to-day "business fundamental" measures required to run the business successfully.
This two-pronged approach provides an extended period for the organization to focus its
breakthrough effort while continuously improving key business processes on a day-to-day basis.
Hoshin Kanri ensures that everyone in the organization is working toward the same end. The plan is
hierarchical, cascading down through the organization and to key business-process owners.
Ownership of the supporting strategies is clearly identified with measures at the appropriate level or
process owner within the organization.
In the Hoshin Kanri process, strategic planning is systematized: The format of the plans is unified
via standards. The standardization provides a structured approach for developing and producing the
organization's strategic plan. The structure and standards also enable an efficient linkage of the
strategic plan through the organization. This ultimately leads to an organization-wide understanding
of not just the plan but also the planning process.
The hierarchical linkage attribute of the Hoshin Kanri plan occurs because of the passing-down
process of the plans at each succeeding level. This is the cascading attribute of the Hoshin Kanri
planning process. It is a very important step in empowering the organization. As each succeeding
level accepts its portion of the plan, it has been involved in the plan's development by adding detail
where it can best contribute and add value. This is also, how the organization buys into the plan; it
now has some ownership of the plan itself. The Hoshin Kanri methodology is a strategic planning
process with the built-in ability to empower the organization.
Measurement
The Knowledge Management program made it a point to continuously stay linked with the
operational and business measures of the division. We are now measuring practically everything we
do within the change program in IAI and the Knowledge Management program is different only in
that it didn't limit itself to plain performance measurement, but went an extra step in defining three
levels of measurements:
•
Performance measure
•
Throughput measure
•
Result measure
To clarify the measurement process, we will use the "communities of practice” procedure as an
example to describe the measure levels:
•
2
Performance - we are measuring the number of people joining the community, the frequency
of their meetings, the members’attendance, and the amount of knowledge shared.
The subject has been addressed in a workshop at the KM-Asia Summit in Singapore in October 2005.
33
•
Throughput - we have required communities of practice within IAI, to self determine goals.
The achievements of these goals, or the generation of best practices out of the meetings, are
throughputs to be measured.
•
Result - we are aiming at finding a relationship between actual business results and
Knowledge Management activity. Measuring those results that are directly or indirectly
accountable to throughputs of the community are a result measure.
IAI is using a method of self-assessment for the implementation of various processes across the
company. The knowledge management implementation is self-assessed as well. The method is
based on the well established capability maturity matrix (CMM) that grades in 5 levels the depth to
which an organization has arrived in the implementation of a given process:
•
Awareness
•
Training
•
Understanding
•
Commitment
•
Habit
At least twice a year, a committee internal to the division and appointed by its GM assesses the
level achieved in the implementation of KM in the division. This is done using a detailed set of
questions about every aspect of the program.
The questions seek an appraisal of the locality of the implementation, as opposed to processes
characterizing the whole organisation, the occurrence of throughput of it and the awareness to its
affect of the operational and business results. Another issue being assessed, is the level of the source
of initiative to the program –does it stay at corporate level or does it go down to the division, to the
directorate or maybe even to the personal level. The leveled approach to assessing the maturity of
the assimilation of a process is a matter of essence. The mere awareness of what "is done" in this
process can not be but the very beginning of the implementation. The next stage is the
methodological one in which the details of "what is done" are now being trained and implemented,
creating a common language. Only when you understand a process, you can fully implement it, not
only by the word, but also by the spirit. Then comes the commitment phase and only latter the
process implementation becomes a habit and is not considered a process anymore.
The internal assessment is moderated on a yearly basis, by a team external to the division that
includes IAI’s director of knowledge, a representative of the KM steering committee, and a
knowledge manager from another division. Those results are negotiated with the division’s
management whose purpose is to learn from the assessment in order to correct its implementation
method and improve its results. Building on the cultural content of the KM program, the selfassessment results are also positively published on the intranet and are another source of internal
competition within the company to give it a catching effect.
Senior Management and Employee Support
Senior management is involved in the program in a few ways:
•
34
The General Managers of the divisions are considered as the customers of the program.
Once a year, during the period in the fourth quarter allocated for planning for the following
year, they set their goals and targets. These are picked by the knowledge manager who then
establishes the KM program for the following year, choosing procedures to better achieve
the division's goals. These constitute the basis of the plan for the following year program.
•
Communities of practice in IAI are induced self-organizations. That is to say, management
is involved in creating the opportunity for the communities to form, and then, their
participants are left to decide by themselves if they want to join the community. To each
community a sponsor belonging to the top management level of the company, is assigned.
The members are required to vote on a set of common goals they are committed to, but the
sponsor approves those goals. This process creates an involvement of management that
helps both ways.
Case Analysis Part II: Competence Centers
Background
Out of the many modules of the comprehensive KM program of IAI, the company chose to focus in
2005 on “competence centers”. The CKO realized that in an engineering based organisation, this
mechanism could bring about significant results.
During the period of April 2005- November 2005 we ran 38 workshops with 38 competence centers
– and each was based on an in depth conversation with key players of the existing or emerging
center. The analysis in this section is based on these workshops.
What is a Competence Center?
IAI uses the following definition:
“A competence center is an employee or group of employees which are responsible for a specific
knowledge domain (that includes databases, information and expertise) and are capable to turn
this knowledge into value to the company and its customers. A competence center can be a
technological, operative or production group which provides the company with a business and
competitive advantage and is critical to the nourishment of existing business lines or to the
development of new ones”.
What do Competence Centers do?
Each competence center has a clearly defined mission, which is related to the provision of specific
business related services to the internal and external customers of the company. In addition to its
main mission, a competence center is expected to fulfill several roles which support the creation,
documentation and sharing of knowledge related to the center’s expertise area:
•
Mapping the current knowledge situation of the organisation and identify emerging
knowledge needs of the center’s internal and external customers..
•
Turning tacit knowledge of the individual experts into tangible documented information
which is accessible to more employees.
•
Continuously developing the expertise of the center and the organisation and ensure it
remains as a market leader and is well updated with the state-of-art knowledge.
•
Identifying of emerging and disruptive technologies as well as global trends.
•
Sharing the center’s expertise with other functions in the organisation.
•
Distributing information about the center’s expertise and knowledge.
•
Exploiting systematically the center’s competencies to the needs of the various units of the
organisation.
35
•
Marketing the center’s marketing (internally and externally) and actively developing the
demand for its knowledge and services.
•
Managing, creating, exploiting and protecting the Intellectual Properties created by the
center.
•
Nourishing the “next generation”and systematically coaching and training new experts in
order to ensue continuity of the center’s competencies.
Characteristics of Competence Centers
We came across diverse types of competence centers in IAI which can be described along the
following dimensions:
Lifecycle stage: Few centers are groups which exist for many years and seem to be beyond their
peek. Others are mature centers, and few of the centers are just emerging.
Focus: the centers are categorized into technological, operative and production centers. Clearly, the
most important ones in IAI are the technological ones thus most efforts were invested in these
centers. However, in some cases this clear cut distinction is not as clear.
Size: Most competence centers include about 5-10 members. However, there was an exception of a
one person competence person.
Organizational composition: some centers have been homogenous and included members from a
single organisational unit; some were based on members from different units, sometimes from
different plants and divisions. However, it was decided not to include experts from outside the
company. In the case of such resources, they were considered as external strategic interfaces and
sources of knowledge.
Distribution of knowledge: in some centers the same body of expertise is common to most
members, in others there are clear distinctions and expertise is well distributed between the
members. In some centers there is a clear top expert who is clearly more knowledgeable than the
others in all expertise areas of the center.
Challenges
Some of the challenges which were addressed by the competence centers were common to all of
them (and indeed, are common to other aspects of KM as well). Other challenges were relevant only
to a few centers. When you read the following list, you will probably identify interlinks between the
listed challenges:
36
•
Lack of time – many experts claimed that they are overloaded with their day to day work
(either routine processes or fire fighting) and don’t have available time to invest in activities
dedicated to knowledge sharing, for example.
•
Lack of resources –the members of many competence centers pointed out that resources for
KM fostering activities such as training courses, participation in conferences and sometimes
even investment in information technologies are scarce.
•
Lack of management attention – members of several competence centers felt that the
awareness of their top managers to the criticality of investing in nourishing competence
centers is limited. In these cases, management focuses mostly on achieving the operational
goals of their units.
•
Internal competition – as a result of IAI's divisions being profit centers, the internal
competition between the businesses units is unavoidable. As a result, barriers for knowledge
sharing between experts from different divisions and plants are considerable. In the recent
years, the value of “one company” has been promoted and there is some progress in
knowledge sharing, e.g. in the case of company wide Communities of Practice. However,
internal competition is still a limiting factor to full knowledge sharing.
•
Aging workforce – like the case of similar organisations in aeronautics and defense, the
average age of IAI employees is relatively high. In many competence centers there are no or
only one young employee. This poses two challenges: A. continuity of the competences
base: what will happen to the knowledge of the experts, when they retire in few years? B.
low energy levels in some cases, limited openness to new ideas, and lack of fresh
perspectives.
•
Mind set – in some competence centers we observed minds sets which are not compatible
with a knowledge sharing and exploration culture. For example: “each expert needs to know
only about his own expertise area. There is no need to organize regular knowledge sharing
meeting of experts belonging tour multi disciplinary competence center”. And also: “There
is nothing we can learn from the equivalent competence center in plant X. We are far more
progressing”. And also: “Our field is mature, there is no technological progress and no
need to explore new directions”.
•
Lack of learning and development opportunities – in some cases, the only real channel to
acquire expertise in new technologies is through R&D projects. However, scarce resources
for self funded research activities limit the options of some competence centers to prepare
for the introduction of new technologies.
•
Disappearing technology – in the case of one center, the expertise is focused on a
technology which will disappear in a few years. The competence center is retained in order
to support old products owned by the company customers. Would this lead to the
degradation of the center?
37
The Implementation Process
The paths IAI took to implement the approach of competence centers are described in the following
chart:
IAI Strategic KM outlined. Center of Competence is
one of the modules
2003
IAI KM Handbook created and published,
incld. A chapter on competence centers
2004
Approx. 40 competence centers during 2005
2005
(Pilots)
Implementation of
Competence
Center is added to
the company PDM
measurement
system
Focused efforts on other modules
e.g. CoP
IAI chooses to introduce the Competence
Center approach during 2005
Each Plant knowledge manager, assisted by the
CKO, chooses two pilot competence centers
An introductory mini-workshop (three hours) to each competence center
Participants: Plant knowledge manager, key/all members of the existing/emerging
competence center , sometimes a senior manager, facilitator
Content:
A. Introduction presentation and conversation: what is a competence center,
objectives, roles, players (based on the handbook).
B. Co-creation of the center’
s profile, using a template which is completed during the
workshop. Template covers the following issues:
vRoles –leader and content manager
vMission
vExpertise areas
vMissing expertise (knowledge gaps)
vList of experts (and specific expertise areas of each)
vInterfaces with customers, strategic knowledge collaborators.
vCompetitors (and relative knowledge advantages)
vIntellectual property and other information assets
vCurrent knowledge processes (learning, sharing, documenting, etc.)
vWork plan for next year –practical actions that can progress the competence center
vInitial structure of the competence center’
s portal;
Output: awareness, profile of the center, draft work-plan
Competence center meets several weeks after the workshop with its managers
to refine work plan and start initial implementation of the agreed action items
2006
(Full
Implementation)
Introducing the approach through the
workshop to more/all competence
centers in each unit of IAI
Intensive implementation
of the work plan of each
center
Figure 3: The implementation process
38
The Players
We identified six groups of stakeholders and function related to a competence center:
The competence center leader –this is the person responsible for the KM related activities of the
center, e.g. knowledge sharing, continue learning etc. In some cases it is the group manager, in
other cases it is one of the experts who is interested to pursue this challenge.
The content manager – this is the person responsible to coordinate the processes of documenting
and making available the center’s knowledge. Typically, he or she is also responsible for the
center’s portal and other information technologies assets.
The EXPERTS –these are the core players. In a typical competence center there are 5-10 experts.
While we didn’t use a formal definition, we linked “expertise”with extensive experience, excellent
skills in a specific area, hands-on competencies to deliver and a distinct capability to solve problems
and support others who seek advice, help, guidance and second opinion.
The top manager –this might be the plant or directorate manager, who is expected to support the
center by publicly recognizing its importance, approving its work plan, providing resources,
highlighting priorities and monitoring performance and progress.
The local KM manager – this is the division, plant or directorate knowledge manager, who is
expected to support the competence center throughout its lifecycle.
External players – internal customers, local customers, members of other competence centers
within and outside IAI.
Figure 3 outlines a typical constellation of a competence center:
Internal and external
customers
Community of
Practice
(domain X)
Top Manager
Competence center:
•Leader
•Content manager
•Experts
Other domain X competence
centers in IAI in other plants
External strategic
knowledge collaborators
e.g. consultants and
academics
CKO
Competence centers in IAI
in complementary domains
Figure 4: A typical constellation of a competence center
33 Ways to Nourish a Competence Center
In this section we report on many practical activities that can enhance the competencies of the
Competence Center and support it in actively “managing its knowledge”. All of these ideas were
39
suggested by the center’s members in the competence centers workshops, and some are already
implemented. Clearly, some of these ideas are applicable to all centers, and others are relevant only
to few of them.
In order to
Upgrade, enhance,
update and
deepen the
center’s
competencies
Provide access to
others to the center’s
expertise and
knowledge
Develop a
new
generation
Effectively
exploit the
center’s
competencies
A competence center can pursue some or all of the following actions
proposed in the 38 workshops:
Periodical multi
disciplinary
meetings (experts
from different
domains)
A regular meeting of
the center –to share
ideas, problems, etc.
(e.g. monthly two-hour
meetings)
Establishing a
Community of Practice
based on similar
competence centers
from different units
Mapping all IAI
experts in the
domain, creating
“yellow pages”
IT
Creating a portal for
internal and external
knowledge sharing
Sharing
Establishing a national
community of Practice
Documenting tacit
knowledge in “lessons
learned”, “problems and
solutions”, “rules” and
“tips”sheets
Locate experts
in proximity (one
physical space)
Analysis and
improvement of the
unit’s current KM
processes
Documenting tacit
knowledge in handbook
Learning & Exploration
Participation in the main
professionals
conferences
Visit suppliers of
materials,
equipment,
technologies
Analyze
failures and
successes
Publish papers
Visit other IAI plants
or other companies
Membership in professional
associations
Subscription to the
relevant professional
magazines
Formal certifications to
its members
Visit customers
Rotation of responsibilities between experts
Internal marketing and
exposure to potential
customers (video,
Self funded R&D projects
brochure, demonstrationto develop competencies
meetings
and new domains
Moe focus on
patent
registration
Enhance
availability of It
resources (more
computers)
Formal certifications to
its members
Actively identify emerging
Deepen the theoretical
technologies and explore
knowledge e.g. academic (visits, internet etc)
courses)
Internal courses
Business intelligence
–learn about
(potential0
competitors
Creating a shared
directory in the network
Exploitation
Internal recruiting new
young employees and
develop as the next
generation
Systematic process to
train new experts
(including personal
coaching by a senior
expert)
Next generation
Figure 5: 33 ways to enhance capabilities of a competence center in IAI
40
Key Findings
Three years into the program, all divisions of IAI have implemented KM in different degrees.
One of the syndromes, preachers of Knowledge Management encounter is the response by proposed
practitioners that they have been managing their knowledge (without capitals) all along, though they
do not use the term KM. Well, it does make a difference to do it intentionally and methodologically.
The framework of the program, the specific procedures in place or even the details of their
definition are less important than the process of implementation and it being systematic (Frank,
2001).
We have dedicated the first and second years of implementation to lecturing, teaching and even
tutoring the details of the procedures using the Knowledge Management handbook, published on
the Intranet and with the help of the Knowledge Managers. This has been done at all levels, starting
at the management level of the division. This was performed by the company's Director of
Knowledge himself together with the local Knowledge Manager who gave the home touch with
recognizable examples. It then proceeded to middle management by the Knowledge Managers who,
in order to get the attention of their audience at this stage had to be very specific about what needs
to be done and what's in it for them and their departments. Finally, it was carried out by the General
Managers to the personnel level at periodical communication meetings at which they present their
policy and point out subjects to concentrate on.
On the second year of implementation we focused on the consciousness of the link between the
division's goals and the chosen KM procedures. We did that using the attributes of the Hoshin Kanri
method. This is particularly important to get management at all levels buying into the program. It
indeed delivered as we had more and more GMs associating with the program and KM procedures
better applied to local situations, environment and goals.
On this year we also concentrated on the Community of Practice (CoP) procedure. We have defined
CoPs in IAI as organizations having a common subject which is of the interest of the company, you
volunteer to be part of (but once you do, you are committed to it), and which has goals community
participants have voted for (typically goals matching their own needs or the needs of their
departments). We have established already over 15 technological communities of practice (dealing
with a specific technology that maybe implemented in quite a few divisions), about 5 procedural
communities (dealing with a common basic procedure), and another 5 organisational communities
(dealing with people from a similar organization in various divisions). The major worry was about
the technological communities that required people from different divisions that were competing
against each other up to now to cooperate in order to mutually improve their capability. As long as
we kept the community at the technological level and didn't venture into the business level, it has
worked beautifully, and these communities are now flourishing while the ties between those people
that started as technological grew into social ties that sometimes are even stronger. Another myth
that was shattered in the process is the famous "knowledge is power" one. Instead of encountering
the expected restrain of knowledgeable participants from telling what they know to those in need of
that information, we found people eager to tell their story to an audience which was sometimes
reticent to admit it needed the knowledge. The new myth therefore created is –"sharing knowledge
is power".
The third year of implementation was chosen to be the year in which we would start
concentrating on the fostering of our technological competence centers, as was described in the
previous chapter. During this year we worked with some 40 centers, and we plan to triple this
number on the fourth year.
41
Business Impacts on IAI
The world is changing, and organizations can’t stop the world from changing. The best they can do
is to adapt - the smart ones change before they have to; the lucky ones manage to scramble and
adjust when pushed; the rest are losers - they become history. In the final analysis, the customer
determines the winners from the losers. IAI is learning to become a learning organization, one that
is better prepared for a changing world and Knowledge Management is leading this effort. This is
true for process knowledge as well as for design already invested in and retrieved for reuse.
A major part of this knowledge refers to defined and documented processes in the various
competence centres of the company (some of them technological while others are procedural). The
rest remains in the generalized term of tacit knowledge which refers to the experience of the
company's people, in their head and memory or at the tip of their fingers. The Capability Maturity
Model Integrated (CMMISM) deals with the ways an organization has to follow, in order to maintain
well mapped processes, having well defined stages, because of the assumption that in mature
organizations, it is possible to measure and relate between the quality of the process and the quality
of the product. IAI is trying nowadays to qualify for the CMMISM highest level possible (at least
one of our divisions is already qualified for level 3 while quite a few are already at level 2).
CMMISM enables you to assess your organizational maturity and process area
capability. It identifies priorities for improvement, and provides guidance on the
implementation of these improvements. On the other hand, applying KM can bring
enormous tangible and intangible benefits. These two area studies have different scope
but similar methodologies such as maturity models and the evolvement through the
processes. Interestingly, in the recent studies, one has taken effects on the other. We can
see CMMISM levels and models applied to some KM models, and KM techniques applied
to CMMISM activities. No matter how they affect each other, it is believed that the
debates and learning from each other should improve them both. Further, to learn from
both of the two studies can obtain the knowledge and clear concept of the operation of
the organization as well as problem solving capability. When Knowledge Management
is used with the Capability Maturity Model Integrated, the organization becomes more
efficient and effective in the development of the projects they are used on (Dayan, 2006).
Divisions which are choosing business result measures, are actually taking upon themselves to
adapt their Knowledge Management activity to the level needed in order to reach the goal
prescribed in their PDM since the beginning of the planning year. This is only starting to happen
and is a sign, people at IAI are only now beginning to consider KM as an enabler for extracting the
maximum of the potential they, their departments, and their divisions have.
Impact on Other Organisations
The authors believe that this case is a valuable source of ideas especially for large and complex
organisations such as automotive or aeronautical industries. Such companies can learn from four
aspects of the case:
1. How a large corporation can deploy the values, policy and practices of Knowledge
Management throughout the organisation. This is a non trivial task and some of the methods
used by IAI can be useful in other cases.
2. How to link a KM program to large company wide change program such as lean
manufacturing. In this respect, how to integrate the measurement of KM related challenges
and performance with a general performance measurement system.
3. How to define and support competence centers in a systematic way.
42
Conclusions
The IAI knowledge management project can be declared as a successful project. It is still continuing
to evolve and has created an organizational effect in the past 4 years.
Factors that contributed to the success:
•
Most importantly the top management’s support of knowledge management issues, and it’s
willingness to allocate resources for this purpose.
•
Building the knowledge management program in a way that enables it to fit with the main
macro change plan that the company is assimilating, so that the two programs support and
strengthen one another instead of competing with each other.
•
Planning and establishing the knowledge management program so that it corresponds with
the company's existing culture. Being an organization run and dominated mostly by
engineers, the IAI culture appreciated a plan that would be very structured, very defined,
and included precise objectives, and comprehensive measurements.
•
Using a multi-disciplinary team of consultants, brought different perspectives and
capabilities into the teamwork, while collaborating among themselves and with the client.
•
Having a "fanatically dedicated", full time knowledge manager, who is totally committed to
the mission was a strong in-house force that set the program in motion. Moreover, the fact
that the knowledge manager appointed, was from the R&D unit, a strong influential unit,
also contributing to the success of the project.
•
Establishing an organizational structure for the knowledge management plan that included
part time knowledge managers in every unit of the company, creating a strong
organizational distribution of commitment and responsibility for the issue.
•
Developing the "knowledge management user guide" that communicates the values, tools
and measures of the plan to the knowledge managers of the various units, and is kept
updated.
•
Creating a strategic plan which covers almost every critical aspect of the company. But at
the same time, introducing the plan in a stepwise mode, each year adding two or three
additional “modules”.
•
Focusing on "competence centers" and "communities of practice" while each one
established gets a "kickoff" workshop that enables it to succeed.
43
•
Integrate the KM program with other change programs, which are
introduced to the company rather then managing it as a stand alone
initiative.
•
Continuously stay linked with the operational and business measures of
the division.
•
Start with the core business processes of the company.
•
Create a comprehensive plan, but then implement it gradually,
introducing every year a few new components of the big plan. Don’
t try
to address all challenges in one go.
•
Balance cleverly between corporate central efforts and local initiatives in
order to maximize impact at the local level but at the same time enable
knowledge sharing across the company.
•
Balance cleverly between pre-defined procedures and initiatives, which
evolve without preplanning.
•
Appoint a KM manager (or “
ambassador”
, or “
CKO”or “
champion”etc) in
each important unit – to ensure good deployment of policies and
practices. Support these people and help them create an internal
network.
Acknowledgements
We would like to acknowledge the courageous person in each division, plant and department who lead the
local KM initiative – their commitment, efforts and creativity in implementing the KM program are critical to
the company's success and are well appreciated. They face successfully considerable challenges. Thank
you!
References
Dayan, R. (2003). KM and culture change at Israel Aircraft Industries. The Knowledge
Management Review, 6,(2), 12-15.
Dayan, R. & Evans, S. (2006). KM your way to CMMI. Journal of Knowledge Management, 10
(1), 69-80.
Ulrich, F.(2001) Knowledge Management Systems: Essential Requirements and Generic Design
Patterns. Proceedings of the International Symposium on Information Systems and
Engineering, ISE Las Vegas, 114-121.
Yoshio, K.(1998). Hoshin kanri - a participative way of quality management in Japan. The TQM
Magazine, 10,(6), 425-431.
Palmer, A. (2004). The concept of right first time design. PhD dissertation, Cranfield.
44
Rony Dayan is a retired Lt. Colonel of the Israeli Air Force, with industrial
experience as deputy GM of MBT, one of the Israel Aircraft Industries successful
divisions (http://www.iai.co.il). Before that, he was the corporate marketing
representative in South East Asia where he received the IAI President Marketing
Award for outstanding performance. Rony has been leading the effort to
incorporate Knowledge Management in the company’
s set of actions for the
implementation of a comprehensive change program (for further information on the
IAI KM program please click here). The program is being implemented now across
the five groups and twenty divisions of this 2B$ Aerospace & Defense company.
(http://www.iai.co.il/site/en/iai.asp?pi=15166&doc_id=21797). Rony has also given
courses in Business & High-Tech Marketing and is preparing a course in Industrial
application of Knowledge Management for the School of Business Administration at
the Israeli College of Management. He holds an Engineering degree from the
Technion in Haifa, Israel, and a Masters Degree, both in Electronics, from the US
Air Force Institute of Technology at Wright Patterson AFB, in Dayton, Ohio, USA.
Rony has been presenting the subject of knowledge management in international
conferences along the last three years, and has published two articles on relevant
issues. Rony Dayan’
s research interests are in the field of Knowledge Management
and of measuring its performance and impact in a large corporation; he is also now
studying for a PhD at Cranfield University in the UK on this subject.
Dr. Edna Pasher founded an international strategic management consulting firm in
1978. The firm provides customized consulting services to organizations both in the
private and the public sectors. Edna Pasher Ph.D & Associates specialize in
assisting their client organizations to speed up strategic renewal in a fast changing
environment. In 1994 Edna identified knowledge management as the critical
success factor for organizational renewal and has become the pioneer and leader
of the knowledge management movement in Israel and an active participant in the
international community of the KM Pioneers. Edna's company is also a leader in
the area of Intellectual Capital management, and created one of the first national IC
reports. Edna earned her Ph.D at New York University in Communication Arts and
Sciences and has served as faculty member at Adelphi University, the City
University of New York, the Hebrew University in Jerusalem and the Tel-Aviv
University.
Dr. Ron Dvir obtained a PhD. in Intellectual Capital from Cranfield University,
studied Industrial Engineering at the Technion - Israel Institute of Technology and
received Ms.C. In Computer Integrated Manufacturing at Cranfield University, UK.
In 1987 He joined Programma, a logistics consultancy firm, to design advanced
material handling systems. In the early 90's, Ron worked for ECI Telecom in the
area of Total Quality Management. In 1996 he became the first Chief Knowledge
officer (CKO) in Israel. In 1999 Ron has founded Innovation Ecology which focuses
on the creation of innovation enabling environments such as Future Centers and
Innovation engines. He is the author of several professional and academic papers,
and co-editor of the book "From Knowledge to Value- Unfolding the Innovation
Cube" and the book “Are you Ready to Disrupt It- an illustrated guide to Disruptive
Innovation”.
45
3
Communities of Practice
–A Case Study from the
Automotive Industry
Patricia Wolf
Communities of Practice – A Case
Study from the Automotive
Industry
Dr. Patricia Wolf, Hochschule für Wirtschaft (HSW) Luzern, Switzerland ([email protected])
Dr. Abdul Samad (Sami) Kazi, VTT – Technical Research Centre of Finland ([email protected])
Abstract
Traditional vertical organisational structures can be an impediment to
the sharing and co-creation of knowledge across functional units. This
was realised by a large automotive firm when upon its merger with
another automotive firm, it observed the power of cross-functional
linkages across different car platform divisions. A project was set-in
motion to harness over one hundred communities of practice (CoPs)
across the functional units of the merged organisation. This case
study provides an overview of the conception, implementation, and
optimisation of the project. It provides reflections and key lessons
learned on how cross-functional CoPs can be developed, nurtured,
sustained, and as to what radical changes they may bring into existing
organisational decision making processes.
Keywords: Communities of practice, knowledge management, automotive industry, crossfunctional CoPs.
Background
Industrial Context
During the late 1980s, a large number of firms in the automotive sector made a paradigm shift in
terms of their organisational and operational structures. They essentially moved from being
organised around a traditional divisional structure to being organised around car platforms. This
allowed for dramatic cuts in the product development cycle time, and significant reductions in R&D
expenditures (per platform level). However, while being organised around car platforms, despite
achieving product focus, their where no mechanisms to facilitate proper learning and sharing of
experiences across the platforms. In short, many firms lost their ability to learn from their own
experiences. Many firms later recognised that platform level focus made sense; there was a need to
introduce mechanisms that would allow for experience sharing across these platforms as well.
When in the late 1990s, Knowledge Management started to hit the front page of different
management journals, these firms realised that unless there where means for tap-in-to and learn
from their own respective experiences, they stood to lose market share and competitive advantage.
While still being organised around car platforms, communities of practice (CoPs) seemed a viable
solution and instrument to stimulate cross functional communication and learning across the car
platforms.
49
Communities of Practice –A Case Study from the Automotive Industry
Problem
After a merger of two automotive firms, it was time to learn from each others’experience in an
attempt to streamline processes, build-upon each others’competencies, etc. During this exploration
process, the German counterpart in the merger was impressed with the knowledge management
structure of the American counterpart. The American counterpart made use of a vertical functional
hierarchy (organised on the basis of product model ranges / car platforms) and made use of
horizontal communities of practice to support knowledge sharing and capture across the model
ranges. Furthermore, a searchable know-how database existed to support documentation of key
experiences and lessons learned (see Figure 1). The German counterpart observed that such crossfunctional communication could potentially help resolve its problem of lack of knowledge and
experience sharing across its product platforms. Furthermore, such an approach could also be used
to allow for cross-functional learning across the “merged”organisation.
Know-How
Database
Figure 1: Cross-functional CoPs and know-how database
A project team was created and assigned the task to implement a know-how database and crossfunctional CoPs that would reach across all model ranges for the entire ”merged” organisation
within two years. The project team was made-up of five employees from the organisation and two
doctoral students. It is supported by a so-called coordinating team which is composed of fifteen
employees of the concerned model ranges. This structure aims to provide for an optimal
communication between the project team and the model ranges and also ensures that everyone stays
updated: as the members of the coordinating team are themselves integrated in the work of the
model ranges, they are to promptly report to the project which problems and topics bother the
members of the model ranges.
After a business segment analysis of its product range, management decided to implement
approximately one hundred CoPs in twelve product and process relevant subject areas (see Figure
2).
50
1
2
3
4
5
Technology Field 1: Passenger Car Elements
Functional units
Technology fields
Technology Field 10: Processes
Figure 2: CoP structure
Early on, a serious problem emerges: there is no provision for cost and time compensation to the
functional units whose employees take part in the CoPs. A key question emerges in that, how can
functional management be convinced that it would be beneficial to release employees for their work
in the CoP(s)? The situation is further aggravated by the absence of a formal collaboration
mechanism between the CoPs (seen basically as a service provider) and the divisions / functional
units. In short, the CoPs project team has fundamentally no control authority (see Figure 3).
Management
Reporting
Reporting
Decision
Decision
No formal
Divisions/
Functional units
collaboration
Service Provider
(CoP Project Team)
Figure 3: Service triangle
The service triangle (see Figure 3) implicates the following decision-making situation: the service
provider is instructed by the management to implement a process innovation without necessarily
having obtained permission from the respective divisions/functional units. The service provider may
operate on behalf of management, but whenever problems with the implementation of a measure
51
occur (e.g. in the form of conflicts with the divisions/functional units), it has to apply to the
management for support. The implementation of measures can be achieved either informally, which
means that the members of the project team convince the divisional unit directors, or through
management’s top-down order.
The problem persists until the end of the project. Initially, it is mitigated by top management which
instructs reluctant functional managers to release the required employees for the work of the CoP.
Moreover, a working structure for the CoPs is developed that copies the existing hierarchical
structure: so-called executive CoPs, to which members of the functional management are assigned,
to guide and monitor the work of the so-called engineering CoPs. This allows at least for a partial
integration of the higher decision-makers. In actual fact, the problem is solved by the members of
the CoPs. Despite some continuing dissention with their immediate functional supervisors they
participated in the CoPs and thus enabled product model range-spanning cooperation.
Through this case study, it is expected that the reader will:
•
Learn the dynamics of and need for communities of practice that span across
organisational divisions and functional units.
•
Explore key steps in the design of cross-functional CoPs.
•
Learn how numerous CoPs can be concurrently implemented and then
optimised.
•
Dealing with conflict within CoPs.
•
Different managerial and organisational implications of implementing crossfunctional CoPs
Approach
The focus of this case study is on the introduction of cross-functional CoPs within a large (merged)
automotive firm. It is presented from both a practical and scientific perspective. From a practical
perspective, the focus is on the conception and introduction of different performance measurement
instruments. From a scientific perspective, the focus was on evaluation and determination of the
opportunities and limitations of performance measurement of knowledge management within
complex social systems.
Participatory observation was used as a main approach for studying and analysing the problem at
hand (refer to earlier sections on industrial context and problem). A broad range of both qualitative
and quantitative evaluation instruments were used over a two year evaluation process during the
CoP implementation project. The evaluation phases cover, conception, implementation,
improvement, and reflection (see Figure 4).
52
Phases of the Evaluation process
Conc- Implemeneption
tation
Reflection
Observation and Self-Observation
q
u
a
l
i
t
a
t
i
v
e
Evaluation type
Improvement
RWS* RWS*
RWS*
RWS*
RWS*
Group
discussion
with NonKM-Actors
Review in the
CoPs
Reflection
Documentation of success stories
q
u
a
n
i
t
a
t
i
v
e
Employee
Survey
CL**
CL**
CL**
CL**
Evaluation of the database: Hits, number of articles, feedback
* Reflection- Workshops
** Checklist
Figure 4: Evaluation framework and phases
The quantitative measures used were those that were designed by the organisation itself, whereas
the qualitative ones were designed during the project. A major objective was to better understand
the overall working structure, decision making norms and communications patterns in the
organisation and designed CoPs. This is reflected through the long learning and reflection phase of
the evaluation framework (see Figure 4).
Case Analysis
Conceptual design
The project and coordination team starts its work with commitment. The CoP structure and the next
steps are planned in-depth and approximately four months after the placing of the project
assignment the project schedule as well as the implementation measures are established.
The implementation of communities of practice starts with an attempt to convince functional
managers as well as potential CoP coordinators of the concept. The CoP coordinator plays a
decisive role as he ought to align and moderate the work of the CoP. Thus, he acts as a moderator
and gives structure to the work of the CoP, but he does not act as a supervisor because he has no
formal authority towards each individual group member. The coordinators of the engineering CoPs
are chosen by the members of the coordinating team in cooperation with the heads of the model
ranges. The persons chosen as CoP members have to meet the following conditions:
•
They are considered as experts in the respective topic.
53
•
•
They have experience in the leadership of a team or division.
They are considered to be open-minded towards changes and have been involved with
former change projects.
Furthermore, it is the responsibility of the CoP coordinators to find more CoP members and to
excite them for collaboration in the CoPs. Supported by the project team, the build-up of the CoPs
is managed efficiently and quickly.
Implementation
A first version of the database gets established and tested. Kick off events and training courses for
CoP members concerning their roles in the new work structure and the handling of the database take
place. These activities prove to be of fundamental importance for the project’s success. On the one
hand, the project team gets the opportunity to present and explain the concept to future CoP
members and to answer their questions. On the other hand, the CoP members become acquainted
with each other and thus get the chance to exchange views, ideas and experiences. The trainings for
the handling of the database have an additional effect: the future authors loose possible
compunctions concerning the handling of the new technology or the writing of documents that will
be accessible for other organisational members through the database.
By now there exist about 90 CoPs that first sift through the guidelines of the several model ranges.
Glossaries are written in order to ensure a consistent comprehension of essential terms in the subject
area of the CoPs. The CoP members give presentations about the work in their divisions; problems
are listed and prioritised for discussions. The database fills up.
Unfortunately, the existence of CoPs alone is not a sufficient indicator of success for management.
The management therefore approaches the project team and asks for a performance measurement by
means of a checklist. Quarterly, statistic data such as the frequency of meetings, the attendance of
the CoP members and the number of placed and planned contributions for the database are to be
collected through this checklist. These performance indicators are simultaneously used to make
requirements: one year after project start, each Cop is expected to have placed three sections in the
database and to have conducted three meetings at which at least 80% of the members participated.
The CoP coordinators are very indignant about this concept. On the one hand, they argue, the
indicators allow no conclusion concerning the quality of the achieved work. On the other hand, the
form of control is counter-productive as it steals time that the CoP members would rather prefer to
spend on textual topics. A form for the documentation of success stories, which complemented the
project checklist, gets ignored.
Without success, the project team reports the objections of the CoP members to the management.
The checklist is completed and analysed. The unsurprising results are: Three meetings each quarter
for each CoP with an average attendance of 80% of the members and three contributions for the
database. If necessary, one large chapter was divided into three smaller ones in order to meet the
requirements. The management is satisfied and requires a double number of chapters for the next
six months.
After one year, the implementation of CoPs can be regarded as completed. To celebrate this success
adequately, a big event for interested members of the division is planned and carried out. The
division manager appears at this event and praises the performances of all parties. In the reflection
workshops that were conducted by the CoP coordinators of the project every six months, the spirit
is ambivalent. On the one hand it is acknowledged that the information and knowledge platforms
that were created where crucial questions that reached across model ranges and were discussed
openly and constructively. Here lies most of the benefit of the new work structure according to the
coordinators. On the other hand the coordinators complain anew about the problem that evolves
from double membership in the CoPs and the functional organisation as well as the missing
54
budgeting of the CoP’s work. At the same time they demand from the project a clarification of the
questions: which competences the communities of practice have within the overall organisation; are
they authorised to make decisions for topics that reach across model ranges? Without decisionmaking competence-based trans-sectoral work does not make very much sense because the
developed solutions are not considered as binding for all model ranges. Therewith, the synergy
effects whose creation is one of the CoPs’special strengths would not apply. Even more so, the
motivation for cooperation gets lost.
Optimisation
After the introductory phase of the project, the optimisation of the new working structure begins. In
order to get insights into the current state of the CoP work, its problems, chances and risks, the
project and coordinating team plan a complex audit: on the basis of a structured questionnaire that
focuses on nine core areas of the CoP work, two hour interviews are conducted in the now nearly
100 CoPs. Previous to the audit at least one meeting is initiated to get an idea of the activities of the
respective CoP. Thus, the monitoring part alone comprises 100 times four hours.
A giant project that lasts four months including the analysis, it reveals important results: it becomes
clear that the CoP work “works”, i.e. the CoP members are dedicated and enthusiastic. Moreover, a
new relationship between project and CoP members develops. The latter perceive the audit as a
sensible and supporting counsel because it does not only aim at the recording of problems and risks
of the new working structure but explicitly allows for the presentation of strengths of the CoP work.
Therewith the project expresses its appreciation of the achieved work and the CoP members feel
motivated. The project advances from a bogeyman that is held responsible for unpopular orders of
the management to a counselling unit that attempts to understand the challenges of the CoP work
and represents the CoP members’interests towards the management.
In the audit dialogues a common intention of the project team and the CoP members becomes
apparent: both want to make sure that a supporting team for the functional tasks is available for the
CoPs even after the project’s end in about six months time and that management’s attention
remains. Therefore it is time to collect arguments that make the benefit of the CoP clear for
management. In the aftermath, the so far ignored form for the collection of success stories starts to
enjoy great popularity. The database gets diligently filled as well, so that the goals that were given
by the management can be adhered to. The CoPs conduct self-marketing and the project team
becomes a mediator who forwards the success messages. These are as follows:
•
The CoP work promotes the exchange between the divisions and allows for substantive cost
and time savings through ensuring a fast access of existing knowledge sources and through
avoiding the duplication of work. Structured and effective coordinating processes between
the model ranges become possible.
•
The standardisation of problem solving across model ranges, the usage of the knowledge
database as the central storage medium and the development of a consistent language use all
contribute to shorter development processes.
•
The business area is able to optimise its supplier and customer relations due to a consistent
appearance.
•
The personal interconnectedness of CoP members in their field of activity improves: a
measurable saving of time exists due to the generated transparency of contact persons, the
CoP work fosters the establishment of personal contacts among one another, the CoP
members get an insight into the activities and goals of other divisions and areas, and a
feeling of group membership emerges.
55
For one year, the quantitative value that the CoP members declare on the form for success stories
amounts to a two-digit million Euro sum and more than 100 man-days of capacity savings.
End of the project
The active marketing of the effects of the CoP work yields fruits: the management is now convinced
of their value and does not want to loose the CoPs as working panels. Therefore they obtain the
official authorisation to issue guidelines that reach across the model ranges.
This improvement changes the communication and decision processes within the organisation.
Suddenly a horizontal cross-functional structure (see Figure 2) exists that is in a position to make
decisions which concern the model ranges and to assert them against the vertical hierarchy, i.e. the
functional management. Nevertheless, upper management reserves the right to veto any decision.
Anyway, the change is radical: everyone who is not a member of the respective community of
practice will not be able to realise his or her views in future guidelines. This does not solve the
problem of the missing budgeting of the CoP work, but it leads to a new and very important
motivation for the CoP members. Moreover the management can be convinced of the CoPs need for
support even after the project ends. The project team will be transformed into a department that is to
perform this supportive work and to enhance further knowledge management topics.
At the end of the project the optimistic conclusion is follows: the company succeeded in
establishing the cooperation across model ranges, which was introduced one year ago, as a routine
process that helps to generalize and standardise knowledge. The stability of the new working
structure is ensured by the authority of the CoPs which is limited to areas of expert knowledge that
reach across model ranges. From now on it is defined in the CoPs what knowledge applies for all
the model ranges together and which solutions are to be applied for which problems. The database
is the instrument that makes the solutions which are defined in the CoPs visible and binding in the
business unit. Thus the company succeeded to realise what is the main goal of knowledge
management in theory: to gain influence on existing communication and decision making structures
across the functional units of an organisation.
Key Findings
The description of the case study has made it clear: CoPs do not only create value, their
implementation can lead to a massive change of the organisational decision structure due to new
communication and decision process structures being co-created. In design and function, crossfunctional CoPs correspond to horizontal expert networks in an organisation with a vertical
hierarchy. With this “the community approach is directed to the creation of those social contexts
and processes, that are to enable the actual content experts of a certain topic to solve problems
collectively”(Hilse, 2000).
CoPs are uncommon interfaces in a sense that they were not envisaged as communication or
decision positions in old vertical organisational structures. Now, groups whose work is difficult to
control form at these interfaces. Of course an effort is always made to manage the work of a CoP
rationally, but more than one example exists of organisational CoPs whose members left the formal
CoP scenery due to the management’s massive control and unreasonable work and turned to the
organisational ‘underground’.
CoPs are partially able to solve the problem of knowledge property in organisations. They allow
certain knowledge to be managed and enhanced by those employees who possess and use it. The
56
employees need no longer divulge their knowledge to an unpredictable organisation. Instead it
remains the property of the CoP and is managed through the common context of its members. This
way, the employees themselves become responsible for the solution of their problems (Snyder and
Wenger, 1999). The knowledge owner regains autonomy: knowledge is shared with peers outside
typical organisational control mechanisms.
The same atmosphere that is created though the CoPs simultaneously increases opportunities for
innovations: in groups of like-minded people where comments need not occur according to
organisational political specifications, employees are more willing to ask sensitive questions and to
discuss ideas.
With respect to the CoP members themselves, databases allow for an internal evaluation of the
commitment of its members. Somebody who only consumes receives no reputation within the CoP
and will soon be deprived of his or her membership. Moreover, the introduction of the database
supports the CoPs because it provides a common workspace where CoP members can store,
organise and download documents that contain the developed problem solving techniques (also
called ‘best practice’). Meta data files allow to identify the author of the document, i.e. the context
of the creation of the document can be retrieved. From the perspective of the organisation, the
described databases represent a location where the employees’knowledge can be ‘stored’.
On the one hand, the build up of a common knowledge database helps to make the standard of
knowledge explicit that was commonly acquired by the CoP. Its visualisation reveals which
knowledge the group of experts considers appropriate for building the basis for decision-making in
the respective subject area. Thus, the experience database provides a knowledge pool to which a
decision maker ought to revert for a certain topic. However, he or she is not denied the possibility to
decline this knowledge or to access other knowledge sources at the same time and thus generate
new knowledge. This way, the possibility that solutions which were developed in the CoP are
refused is alleviated.
On the other hand, the credibility and value of information on the level of the organisation
increases: if a person shares his or her answer to a question with everybody, one expects him to
know the answer for sure (Lessner and Stork, 2001). Therefore, entries also serve to build up
reputation within the organisation: databases make the name of employees known and improve their
profile.
Business Impacts
The results that the organisation described achieved from the implementation of their crossfunctional CoPs are the following (Wolf, 2003):
•
Coordination and dissolving of horizontal or cross-departmental problems,
•
Conservation and systematic dissemination of knowledge between ‘generations of
employees’,
•
Avoidance of duplication (e.g. through the combination of tasks),
•
Shorter development cycles (e.g. through synergy effects),
•
Saving of time (e.g. due to more transparency concerning contacts),
•
Continuous build up of knowledge in areas that are crucial for competitiveness,
•
Transformation of new ideas and technological developments into innovations,
•
Consistent appearance and action towards suppliers and customers,
•
Promotion of further training for employees (e.g. through insights into other divisions).
57
Conclusions
It seems to be obvious that CoPs are a KM instrument that can be used for solving problems related
to a lack of communication between different car platforms within a big automotive company. But
the implementation of CoPs is not an easy task. What makes CoPs such a special kind of group? As
far as it is by now observed, the processes of group forming in CoPs do not differ from prevailing
phase models of grouping. According to Tuckman (1965), a formation, conflict, standardisation and
working phase exist and they are all integrated in an iterative process of self-renewal.
Generally, four areas exist where CoPs differ from other organisational groups such as divisions
and teams (Lessner & Storck, 2001) – to simplify matters, these other groups are jointly
summarised under the term working groups:
58
•
Relationships within a working group develop around the tasks that are dealt with for the
organisation. This means that the members of a working group can neither determine who
belongs to their group, nor with whom they wish to interact for a respective topic. Formal
processes of task performance restrain free interaction. Moreover, the members of a working
group may generally have different competences and areas of interest.
•
In CoPs, the relationships develop around a common practice or area of interest. People
become members of an organisational CoP due to their interest in a certain knowledge area.
The rules for interaction as well as the topics that are to be worked on are defined by the
CoP itself and thus determine the formation of relations amongst the members. Generally an
unconfined build-up of ties within the group is possible, unless the CoP itself defines
restrictions for interaction through the appointment of roles.
•
Authoritative relationships in working groups are set by the organisation through defining
one member as authorised to issue directives. In CoPs however, authority develops during
the process of interaction and is determined through expert knowledge. Working groups and
CoPs do however have one point in common: the leadership function relieves the group. But
while this function leads to the sole power of decision in working groups, CoPs tolerate
leadership only in terms of moderation. In the portrayed case study, the CoP coordinators
were responsible for the organisation of the CoP meetings and the coordination of the topics.
However, these topics were defined and prioritised by the whole CoP. Thus, the demands
towards managers in CoPs change dramatically compared to those in working groups: a CoP
coordinator has to possess a comprehensive professional competence in the area of
knowledge that the CoP works on and he/she also needs to be well experienced in the
coordination and organisation of group activities. At the same time the coordinator needs to
be modest enough not to make demands regarding leadership or decision making, despite
the extra work that he/she performs for the CoP.
•
The goals of working groups are typically defined by persons (such as management) that are
not members of the group. CoP goals, in contrast, are goals of the CoP members. The goals
of working groups are very specific and their fulfilment is relatively easy to measure –one
example would be the production of a number of products within a certain time and quality.
In most cases, CoP goals can only be specified in such a way as ‘finding of a common
solution for the problem X’. The described case study should make it clear that there is no
use in setting specific goals for CoPs: the portrayed CoPs have achieved the goal of a certain
number of articles in the database; however the aim was “complied with”by partly dividing
existing chapters.
•
While working and report processes for working groups are provided by the organisation,
CoPs can define them themselves. CoPs are only accountable towards their own members.
In the described case study, the management insisted on receiving performance results of the
CoP work. The management got these results not because the CoP members regarded this
request as legitimate, but as a trade-off for the given support towards sustaining the CoPs.
Insofar one could rather speak of self-marketing than of providing accountability. So in fact,
the CoP members accepted evaluation instruments from which they profited themselves. For
example, the audits provided a mechanism for self evaluation and reorientation, and
collecting success stories that enabled the CoP members to get an insight in the value of
their work. In this respect, the instruments of performance measurement contributed to the
CoPs’cultivation of identity and advancement.
What appears altogether as a trait for CoPs is a growing individualisation of team working. Even
though the common interest for a certain area of knowledge is the identity forming element of
CoPs, it remains nevertheless the choice of its members to decide in which subtopics they want to
play an active part. The forming of subgroups was often observed in the case study. Furthermore,
CoP members can have this status in several CoPs.
So the definition of a group as a strictly closed community with steadily defined roles, tasks and
topics changes as soon as interest comes into play. This becomes especially evident for virtual
CoPs: in this case it is no more necessary to be collectively present in order to be member of a
group.
59
Here are the seven most important practical tips and lessons learned from this case
study:
•
Be aware that CoPs are able to change the structure of your organisation.
Managers who start a CoP implementation project need to be ready to accept
and support this effect even if this means that they have to give away a part of
their power and lose control mechanisms.
•
CoPs need continuous management attention and sponsors within the
organisation. The work of the CoP members needs to be valued.
•
There are a number of simple structural mechanisms that make the build up of
CoPs within an organisation easier (for example, the CoP structure should
rebuild the organisational structure as in the case study which was done by a
differentiation between Executive CoPs and Engineering CoPs. This allowed the
integration of middle and upper management into the CoP work). In addition,
roles and competencies have to be defined clearly from the very beginning in
order to avoid conflicts.
•
Communication is the key for sense making: Continuous and transparent
communication of the objectives and achievements of the CoPs creates a
common understanding within the organisation about the value of the change.
Use all communication channels available, e.g. regular newsletters, trainings,
information events.
•
The implementation of CoPs is a process that can lead to conflicts. Thus it is
favourable to have a mediator who is responsible for the progress of the
implementation process (this mediator can be a project team as described in
the case study). The mediator (team) needs to get incentives for taking over
this often difficult role –for example by highlighting its responsibility for the
implementation process and its successes.
•
The implementation process of CoPs needs to be monitored and evaluated.
Starting this process is an important task for management as evaluation results
facilitate reflection and enable the actors of the process innovation to
understand what has been achieved and to develop a common language for the
communication about the project results. But: Managers should be careful with
the selection of the indicators they would like to use within the evaluation.
Measurement criterions have to be agreed with the CoP members in advance to
avoid organisational paradoxes.
•
The work in CoPs requires a budget, be it a defined time frame, meeting rooms
for free or real financing. This investment helps to minimise the risks of
conflicts between the ‘
old hierarchy’and the new CoP working structure.
References
Hilse, Heiko. (1999). Kognitive Wende in Management und Beratung. Wiesbaden: Deutscher
Universitäts-Verlag.
Lesser, Eric & Storck, John. (2001). Communities of Practice and organisational performance. IBM
Systems Journal, Vol. 40, No. 4, 831-840.
Palass, Brigitte. (1997). Der Schatz in den Köpfen. manager magazin, 12, pp. 112- 121.
60
Probst, Gilbert, Raub, Stefan, Romhart, Kai. (1998). Wissen managen. Wie Unternehmen ihre
wertvollste Ressource optimal nutzen. 2. Auflage. Wiesbaden: Betriebswirtschaftlicher Verlag
Dr. Th. Gabler GmbH.
Snyder, William & Wenger, Etienne. (1999). Communities of Practice. Lessons learned from
Auburn Hills. Stuttgart: DaimlerChrysler Corporate University.
Tuckmann, Bruce. (1965). Development sequence in small groups. Psychological Bulletin, 63, 384399.
Wolf,
Patricia.
(2004).
Eine
Geschichte
Organisationsentwicklung, 2, pp. 10-19.
über
Communities.
Zeitschrift
für
Wolf, Patricia. (2003). Erfolgsmessung der Einführung von Wissensmanagement. Münster:
Verlagshaus Monsenstein und Vannerdat.
Dr. Patricia Wolf obtained her PhD in Business Administration in 2003 at the
University of Witten-Herdecke (Germany). Whilst writing her doctor thesis on
‘
Success Measurement of Communities of Practice’
, she worked from 2000
till 2002 on a Knowledge Management Implementation Project at the
DaimlerChrysler Passenger Car Development. Currently, Patricia is working
as lecturer, researcher and consultant at the Hochschule für Wirtschaft
(HSW) Luzern in Switzerland. In the meantime, she writes her habilitation on
a European- Russian collaboration project at the ETH Zürich, Switzerland.
Her research interests include Communities of Practice and the
transferability of management concepts between different cultures. Prior to
this, Patricia worked for three years as Researcher and Consultant at the
Institute of Human Factors and Technology Management at the University of
Stuttgart (Germany) where she worked on numerous consulting and
research projects in the areas of Knowledge Management and Innovation
Management. Patricia is leading the Zone ‘
Knowledge Management in
Central Eastern Europe’at KnowledgeBoard.
Dr. Abdul Samad (Sami) Kazi, senior research scientist at VTT, the
Technical Research Centre of Finland is a Civil Engineer by profession, and
Construction Manager by area of specialisation. Sami has conducted
research in several areas including: project planning and control, computer
based training, construction management, knowledge management, system
dynamics, construction IT, information modelling, product and process
modelling, cost planning and control, Internet based technologies,
groupware tools and technologies, mobile communications, information
modelling, mind mapping, concurrent engineering, virtual enterprises,
organisational learning and assessment. Sami has edited several books in
the areas of ICT for construction projects and processes, knowledge
management, systemic innovation in construction projects, etc. He is listed
in the Marquis Who’
s Who in Science and Engineering, is a fellow of the
Entovation E100 Global Knowledge Leadership network, and is content
steward at KnowledgeBoard.
61
4
Sharing through Social
Interaction: The Case of
YIT Construction Ltd
Anssi Koivuniemi
Sharing through Social Interaction:
The Case of YIT Construction Ltd.
Dr. Abdul Samad (Sami) Kazi, VTT-Technical Research Centre of Finland ([email protected])
Anssi Koivuniemi, YIT Construction Ltd. ([email protected])
Abstract
The construction industry is a project based one with each project
being the production or provision of a one-of-a-kind product or service.
Each day on a typical construction site brings forth a new problem for
which where possible, a new solution is devised on the site. This is
rarely, if at all documented. In simple terms, valuable knowledge and
experience gained is lost. What remains is stored in the minds of
those who were a part of the problem and the solution.
Knowledge is created, captured, and shared within a particular
context. This is as agreed by most authors more of a social than
technical process. This case study presents the findings from the
PromoBat project that was initiated at YIT Construction Ltd. in Finland
to provide a framework and basis for the capture of new good and bad
practices through a set of social processes. Some of the social
processes developed, site visits, cross-auditing, and horizontal quality
and process teams are presented. The findings indicate that such
social process serve as an effective lever for the capture, sharing, and
dissemination of valuable experience and knowledge. A set of critical
“intervention points” were identified where such interaction (social
processes) would be triggered. Recent experience reveals that social
processes form an effective means for the capture of experience and
knowledge in project based businesses such as construction.
Keywords: Social processes, construction industry, knowledge sharing, intervention points
Background
Domain Context: Knowledge Management
Under the subject of the subject of knowledge management, there is near mutual consensus that two
forms exist, i.e. tacit knowledge and explicit knowledge (Nonaka and Takeuchi, 1995, Awad and
Ghaziri, 2003, Kazi et al. 2001, Davenport and Prusak, 2000). Here, tacit knowledge is seen as that
knowledge which is “personal, context-specific, and therefore hard to formalize and communicate”
(Nonaka and Takeuchi, 1995). Whereas, “explicit knowledge is codified and digitised for example
in books, documents, reports, white papers, spreadsheets, memos and databases” (Awad and
Ghaziri, 2003).
Having the two different forms of knowledge; tacit and explicit, a main challenge has been to
identify as to how for example one form of knowledge could be converted to the other and vice
versa. The underpinning assumption here is that further knowledge is created through the
interaction between tacit and explicit knowledge (Nonaka and Takeuchi, 1995). These conversion
mechanisms were encapsulated within the SECI (socialisation, externalisation, combination, and
internalisation) model by Nonaka and Takeuchi (1995) that is well cited in most literature on
65
Sharing through Social Interaction: The Case of YIT Construction Ltd.
knowledge management. Socialisation is for tacit to tacit knowledge conversion, externalisation for
tacit to explicit knowledge conversion, combination for explicit to explicit knowledge conversion,
and internalisation for explicit to tacit knowledge conversion. Furthermore, knowledge conversion
is a “social”process between individuals and not necessarily confined within a particular individual.
Therefore, interactions between individuals or groups of individuals are required to capture,
convert, and create new knowledge from existing knowledge (Nonaka and Takeuchi, 1995).
From the perspective of the construction industry, the most valuable form of knowledge is that
which is tacit, based on the experience of individuals, and articulated through social interaction. In
fact, in the construction industry tacit knowledge has been passed from skilled workers to
apprentices over the years. At the same time however, since this knowledge is not readily
accessible, and at times more often than not is in the form of rules of the thumb, or intuitions, there
is a need to formalise it in a form and means that makes it accessible to many. Therefore, the initial
focus would be on socialisation (tacit-to-tacit capture) and then externalisation (tacit-to-explicit
conversion) to package the knowledge in a formal form that is accessible by many. In simple terms,
how can experience be captured and made available in the form of good and/or bad practices that
are accessible by all relevant stakeholders through for example a best practices database. Note that
this is but a first and yet very essential step. It is naturally to be followed by combination (explicitto-explicit) and internalisation (explicit-to-tacit) processes in a continuous loop (socialisationexternalisation-combination-internalisation) for the formation of new knowledge.
Industrial Context: Project Based Business
A project may be defined as, “a temporary endeavour undertaken to create a unique product,
service, or result” (Project Management Institute, 2000). When more than one entity (or interorganisational entity) is engaged, then a virtual enterprise (VE) is formed. The operational
dynamics of virtual enterprises have been a subject of research by many (e.g. Browne et al, 1994,
Charbuck and Young, 1992, Afsarmanesh et al., 1997). “The Virtual Enterprise consists of a series
of co-operating 'nodes' of core competence which form into a supply chain in order to address a
specific opportunity in the market place” (Walton and Whicker, 1996). In the case of the
construction industry, this “specific opportunity”would as an example be the delivery of a building.
At the same time, such projects are performed by people, constrained by limited resources, planned,
executed, and controlled (Project Management Institute, 2000).
Capture of and access to the good (and bad) practices within project environments are essential to
ensure better/improved project or service delivery of other projects. Since projects are one-of-akind, there is no standard template or set of practices that is applicable to all. Though a generic set
of processes and associated good practices may be considered, it is the unique processes and
practices that are of value from a knowledge management perspective.
Organisational Context: YIT Construction Ltd.
YIT Construction Ltd., a part of the YIT Group, is the market leader in Finland and largest
developer of market-financed residences. Its operations span the Nordic and Baltic States along
with a growing presence in Russia. At the forefront of construction organisations in the
development and adoption of key emerging technologies for the sector, YIT is developing model
and knowledge based IT applications and related operational practices for internal use as well as for
enhanced collaboration with other companies in the construction value chain. YIT has early on
recognised the value of the experience and knowledge of its workers and is identifying the means to
capture and share them effectively for improved onsite productivity, assurance of better quality
products, and the provision of better value to its customers.
66
Problem
The construction industry is a project-based one with the delivered product changing from project to
project under the influence of many different factors, one of them being context. This context could
as an example be simply the physical conditions surrounding the landscape where the product (the
building) is to be delivered. Consequently each day on the construction site brings forth a new
problem and a new solution. These problems and solutions are rarely well documented and valuable
lessons learned confined to the minds of only those who experienced them.
Construction, being a project based one-of-a-kind production industry, knowledge and experience
from past projects can be detrimental to the non-repeatance of past mistakes and the re-use of good
solutions. This however is not very easy as project partners typically change from project to project
and new partners are little if at all aware of past experiences, good or bad. As such, a means to
capture good/best practices is essential by the main contractor to then be provided to other partners
for use in the joint delivery of a building.
In the coming years, the construction industry will lose a large portion of its skilled force (Kazi,
2005). It is not only the loss of this skilled force that is a matter of concern for construction
organisations, but also the experiences and knowledge that this skilled force will take with them.
Construction organisations such as YIT have experimented with different tools and instruments to
capture some of the good and bad practices that are salient characteristics of a typical construction
site and project. A good practise database has been implemented as a first means for the capture and
appropriate dissemination of the captured good practices.
Work towards the capture and reuse at YIT of best practices started in 1997. The initial focus was
towards the standardisation of different structural components that they regularly used in buildings.
Relying on this as a foundation, YIT’s best practices database was launched in 1998. This tool (used
in parallel with several other ones) provides employees with access to best practices in the form of
cards. The current system operates under lotus notes and is available only through the Intranet. All
updates are done manually by a knowledge manager.
An analysis of the current best practice database has revealed some in-efficiencies. A few are
reported below:
•
Best practices are not properly systematised in accordance to YIT’s ontology
•
Database updates are only a few times a month and are not done automatically
•
Knowledge workers are not able to update items or comment on them
•
When no knowledge is available, clear indication as to whom could be a potential
knowledge provider is missing
•
Customisable user interfaces both at user group and individual level are missing
•
All information is accessible (only in Intranet)
•
There is no possibility to access (in full or partiality) the best practices from outside YIT
offices
•
People replicate the database on their computers, hence they may not have access to the
latest correct information
It was observed however, that more than IT based tools, social interaction between persons is the
most effective means for capturing good practices. This was to be the first step in the knowledge
capture and sharing process rather than reliance on an improved best practices database. A new
development project (PromoBat) was launched at YIT to explore different mechanisms and social
processes that would facilitate both the capture and sharing of best practices and experiences
67
between project personnel. This would then be used as an input to the best practices database, which
was to be further developed based on recommendations of project personnel.
•
Use of social processes for knowledge capture and sharing within project-based
industries such as construction
•
The role of cross-functional audits as a basis for sharing knowledge on
processes and furthermore as a basis for adaptation and creation of new
knowledge
•
The use of different instruments (social processes) in providing a context
through which knowledge and experience can be shared
•
Solving complex problems through cross-functional teams
•
Using “
intervention points”as triggers for instantiating social-interactions when
and where needed
•
New forms of knowledge visualisation through a combination of building product
models and knowledge sources
Approach
The approach used relied on a thorough analysis (supported by interactive user feedback) of the
current state of knowledge sharing and transfer within YIT. A main objective here was the
identification of interaction modalities between individuals and groups of individuals within YIT.
The contexts within which these interactions took place were noted and the potential for knowledge
capture explored. A set of comprehensive interviews with various stakeholders ranging from site
managers to quality experts revealed social interaction (at times within an informal context) was the
most effective means for articulation of tacit knowledge and experience (Koivuniemi, 2004).
Focussing on the socialisation aspect of knowledge capture and conversion (tacit-to-tacit), various
“social processes”were identified. These were in many cases context specific and purpose driven.
Their relevance and potential were discussed through interviews and social process prototyping.
Social processes were then classified and defined in detail. This was followed by an investigation of
the common business practices with typical residential construction projects. Social processes were
then mapped on to the business processes in the form of a set of “intervention points”. These
intervention points constituted those phases within a business process where a social process could
be initiated and on its basis valuable knowledge and experience captured and shared.
To translate the captured tacit knowledge to be of a more explicit form and shareable across a wide
spectrum of relevant stakeholders, externalisation (tacit-to-explicit) was used. In its simplest form, a
set of templates were designed that were then to be filled by a knowledge officer based on the
conversations that took place during a given social process.
A final set of interviews and discussions were held with the stakeholders to solicit feedback on their
experience with and impressions of the social processes, intervention points, and knowledge
captured. Where required, some social processes and templates were redesigned and more
intervention points identified.
68
Case Analysis
Knowledge is created and shared within a context and space (physical or virtual). This space or
setting has been described through the Japanese concept of “ba”(Nonaka and Konno, 1998). Within
the PromoBat project, context specific social processes were seen as a basis for the establishment of
this “ba”.
A set of twelve social processes were initially identified and grouped under three categories: action
evaluations (social processes that aim to improve activities, measure and control site quality and
estimate individual or team competence), internal activities of a construction site (social processes
covering interactions and activities between site managers, employees and subcontractors), and
activities within internal interest groups (social processes aimed at facilitating knowledge sharing
and interaction between different functional units such as project management, sales and marketing,
design, quality, etc.). Of these twelve, five were initially considered for further investigation and
prototyping (these are shown in italics and without highlighting in Table 1). Based on the findings
of these five selected social processes the remainder and other identified social processes were to be
developed.
Of these five social processes which were developed, three have been deployed site visits, cross
audits, and horizontal quality and process teams), while others are to be considered for deployment
in the near future. A detailed investigation and analysis is available in Koivuniemi (2004).
Table 1: Identified Social Processes
Action Evaluations
Site visits
Internal quality auditing
External quality auditing
Cross-auditing
Performance and development discussions
Internal Activities of a Construction Site
Weekly meeting
Site meeting
Contractor meeting
Morning meeting, task assignment/delivery
Feedback
Activities within Internal Interest Groups
Horizontal quality and process teams
Site management meetings
Social Process: Site Visits
Site visits were a new concept introduced within YIT under the PromoBat project. Their main
underlying purpose was to act as an informal setting (“ba”) for the exchange of knowledge and
experience between site personnel and those within the head office or located on other construction
sites (Koivuniemi, 2004). This may in fact be seen as a short termed VE in which various
participants from different units within YIT come together for a particular purpose: sharing
knowledge.
An overview of the site visits social process is shown in Figure 1. The key aspects of this social
process lie in each participant identifying at least one unique good and bad practice from what they
have observed during the site visit. So as an example, if there are ten participants during the visit, a
minimum of ten unique good and ten unique bad practices are captured. Past experiences are shared
69
with the site manager, and important lessons learned documented for further sharing and future
reference. The feedback received within PromoBat has been very positive. Site visits have led to the
establishment of relationships and contact points between the participants. Best of all, practices that
may have otherwise gone un-noticed or un-documented, are now captured and made available for
sharing.
Project (site) Presentation
1
•Participants meet at the construction site
•Brief introduction of participants
•Site manager presents the project (site)
•Achievements
•Problems faced
Site Tour & Observations
2
•Participants take a tour of the site
•Observations are made:
•What is good
•What is bad
•Questions are asked from site manager and
other site personnel
Discussion & Follow-up
3
•Discussion of identified:
•Good practices
•Bad practices
•Key lessons learned
•Identification of follow-up actions & meeting
•People
•Info. Sources
•Best practice template filled
Figure 1: Social Process: Site Visits
Social Process: Cross-Auditing
Cross auditing is used within YIT to allow different functional business units to benchmark
themselves against one another. This allows for effective knowledge sharing across the business
units. They key here is that the purpose of the audit is to capture and share good practices across the
business units as opposed to simply seeing whether a business unit is functioning properly
(Koivuniemi, 2004).
An overview of the cross-auditing social process concept is shown in Figure 2. As may be seen,
both auditor and auditee contribute existing knowledge and through the cross-audit acquire new
knowledge. In this process, the main provider of the existing knowledge is the auditee whereas both
the auditor and auditee are recipients of new knowledge in the form of ideas and good or bad
practices. Both the auditor (one business unit) and the auditee (another business unit) know that the
aim for the cross-audit is to harness new ideas and capture good and bad practices. A cross-audit is
typically repeated at a later time with the roles of the business units changing, i.e. the auditor
70
becomes the auditee, and the auditee the auditor. Current experience with cross-audits has been very
positive. The concept is currently being heavily promoted across various business units within YIT.
Existing Knowledge
•Experience with existing processes
•Business unit specific good & bad practices
•Documents & templates
Auditor
Auditee
•Target of the audit
•Shares processes
•Gets new ideas
•Learns from auditor
Cross
Audit
•Audits the auditee
•Evaluates processes
•Provides new ideas
•Learns from auditee
•Identified new processes, operational models
•Identified new good & bad practices
•New contacts & information sources
New Knowledge
Figure 2: Social Process: Cross-Auditing
Social Process: Horizontal Quality and Process Teams
The concept of the horizontal quality and process teams social process is unique to the Tampere unit
of YIT. It is an informal conversation between individuals from different business units to discuss
and/or solve a particular problem. The team itself is configured to constitute various members of the
value chain constituting representatives from: designing, project management, sales and marketing,
quality management, procurement, site management, etc. A key point of discussion is the
information flow across a given process or set of processes.
An overview of the horizontal quality and process teams social process is shown in Figure 3. A
team member typically presents a problem that is then discussed and deliberated upon. Team
members discuss the problem at first from the perspective of “why”it occurs. This is followed by a
presentation and discussion of the possible solutions to the problem. These solutions are analysed
from various perspectives (e.g. procurement, design, quality, management, etc.). The findings are
reported and captured in the form of a set of good (solutions) and bad practices (aggravating the
problem). These may be generic, or function specific. Positive experiences have been reported
through the use of the horizontal quality and process teams social process. Experience has shown
that to ensure a successful outcome, it is necessary to select problems of importance and relevance
to all participants in addition to ensuring follow-up for further discussions if necessary.
71
Horizontal Quality & Process Team
Sales
Site Management
Marketing
Design
Procurement
Quality Management
Project Management
Tacit Knowledge
(experience)
Problem
Discussion of
Problem
Discussion of
Solutions
New Best
Practices
Explicit
Knowledge
Figure 3: Social Process: Horizontal Quality and Process Teams
Key Findings
Through the case analysis it was observed that different types of social processes (e.g. site visits,
cross-audits, and horizontal quality and process teams) provided an effective means for
communicating across different functional units within YIT. In some cases, this was an opportunity
for people located in offices to experience the dynamics of work on a typical construction site.
Furthermore, through the feedback that was received, all participants found the concept of social
process as a basis through which they could learn from each other, and therefore make better
informed decisions. New contacts were regularly made, and each contact in essence meant a new
knowledge node that could be tapped into in times of need.
The social processes were purposely designed to create a specific context for knowledge capture
and sharing. Experience in the use of social processes indicated that there were times when a social
process enabled the capture and creation of knowledge within one setting, but was limiting in
another. Therefore, care had to be taken as to when what type of social process was to be used. This
was relatively easy to determine after a few rounds of social processes. In fact, once a participant
had participated in a particular social process for a few times, he/she was then able to lead such
social processes (i.e. initiate it, act as the moderator, document main findings, and coordinate
follow-up actions). This was then used as a basis for the co-creation of a “people-2-people”
community within YIT (see Figure 4). Here the main team responsible for the PromoBat project
acted as the “YIT Knowledge Centre”and provided social process instantiation, moderation, and
other support services when needed.
72
Local
(project/SIG/network)
coordinator
Member
Local CoP
Regional
CoP
YIT
Knowledge
Centre
Knowledge
hub
YIT
Community
Figure 4: People-2-People Community
There did however emerge one problem with the social processes. Based on the workloads and
demands of the participants, there was a limit to how many social processes each could be engaged
in despite their desires for active participation. Furthermore, it was unclear as to when a social
process should be instantiated. To solve this problem, a new concept of “intervention points”was
introduced.
The concept of intervention points used in the PromoBat project was an extension of that introduced
by Kazi (2004) and further extended by Koivuniemi (2004). In both cases, intervention points were
seen as those phases within a process or a set of processes where a major decision or change was to
happen. Within PromoBat, YIT’s processes for building construction were examined and
intervention points identified. This concept allows for clear pinpointing of and identification of
critical phases where the potential of knowledge sharing and capture is high.
While social processes provide the context and means for the capture and sharing of knowledge,
intervention points identify the time (process phase) where they are most beneficial. Project
managers and site managers were offered the opportunity to identify as to when they saw a need for
a social process and as to what type of social process they believed would be most beneficial at that
particular moment to assist them in their decision making / learning needs. Each was given the
option of identifying a set of required, preferred, and nice to have intervention points.
The concept of intervention points and accompany social processes is best understood through a
simple example of sketch planning as shown in Figure 5. Here the YIT personnel include the
project manager and sales manger. Within this process flow, two key intervention points were
identified for plan checking and material selection. It was at these points that some major decisions
are made and issues finalised. For each intervention point, as shown in Figure 5, a relevant set of
social process and the participants of the social processes are identified. Note that in the case of plan
checking, the cross audit social process type is one of the selections. In this particular case, this
cross-audit could be between two projects with each sharing its knowledge and experience.
73
- Site visit
- Cross audit
Project Plan
Project
Manager
Architect
Key Participants:
- Project manager
- Architect
- Sales manager
- Other partners
Financial
Calc. s
Instructions
Sketch
Planning
Plan
Checking
Plan
Approval
Plan
Checking
Sales
Manager
Material
Selection
Project
Partners
Approved Plans
Social Proc. type:
Plan
Checking
Social Proc. Type:
- Horizontal quality &
process teams
Key Participants:
- Procurement personnel
- Sales manager
- Project manager
- Quality manager
Figure 5: Intervention Points - Sketch Planning Example
Business Impacts
In the coming years, the construction industry will lose a large portion of its skilled force (Kazi,
2005). It is not only the loss of this skilled force that is a matter of concern for construction
organisations, but also the experiences and knowledge that this skilled force will take with them.
According to Albert Einstein, “The only source of knowledge is experience”1. This has been one of
the earlier guiding principles of the construction industry where, skills were passed on from master
to student, from one generation to the next; the construction industry of yester years was a pool of
skilled craftsmen. Industrialisation, large scale projects, and a large mind shift to profitability over
the years have changed the very nature of the way industry operates today. In the past, focus was on
the transfer of skills to maintain a set of skilled craftsmen, whereas today the focus is on shortening
construction times, minimising costs, and effectively utilising resources. Organisations today are
waking up to the reality that their most valuable asset is the intellectual capital (knowledge and
experience) of their employees. Significant efforts are being put in place to try to manage this
knowledge through social interaction, as was demonstrated through the use of social processes and
intervention points in this case study.
Within leading construction organisations such as YIT, where efforts are made to make efficient use
of information and communications technologies (ICT) to both enable seamless information
interoperability across heterogeneous applications and at the same time to provide a user friendly
1
http://uk.encarta.msn.com/text_761562147___4/Einstein_Albert.html
74
ICT environment. Based on the findings of this case study, considerations were made on as to how
to leverage the knowledge created and captured through the social processes and to make this
available to relevant practitioners for use in their daily work as and when required. Of course, there
exists the possibility of having a well-maintained and updated best practices database, but according
to YIT’s past experience, this on its own was not enough. There was a need for embedding valuable
experiences and work practices within the (virtual) product itself. This was seen to be through an
inter-connection between product models and a best practices database (Figure 6).
Product
Model
Server
Registry
Product
Model
Ontology
Server
BP Card
Manu
YIT
BP
Database
Figure 6: Connecting Product Models and Best Practices
Product models are at the heart of building design and are packaged with volumes of parametric
information. This information on its own however is not structured in a meaningful way for a
simple user to make use of it. Intelligent means and applications need to be developed to make the
product model more “smart”. As an example, the automatic generation of a project status report for
an executive manager, or the generation of a detailed cost estimate report for the finance
department.
At the current state of implementation in YIT, product models and the best practice database,
though sharing a common data structure (as close as feasible) are distinct entities with no physical
links. As such, it becomes necessary for one to perform specific searches to find for example
construction techniques for a beam.
A means to make the use of product models in the daily lives of construction personnel (e.g. at the
site) is currently under investigation at YIT. This involves the linking of the best practices database
with a product model server. Linking (relationships between product model objects and best
practice cards) would be done though an intermediate application. The main ambition being that
once a product model is loaded into the product model server, it would be pre-populated with
relevant best practice links from the best practices databases. It would for example automatically
attach construction techniques; safety directives etc. for all walls that have a thickness of 20 cm or
less.
75
It is worthy to mention here “visual”experience that an end-user (e.g. at the site) would experience.
The user would see a 3-Dimensional representation of the product model. This would be navigable.
Once a user clicks on a certain object (e.g. door), a menu would pop-up containing different best
practice cards, contactable experts, quality manuals, etc. From here, one could select a particular
report (e.g procurement details), or a best practice card (e.g. methods of work). These would be
printable if needed. The ambition is to ensure that a user need not use any more than a mouse to
navigate, select an object of interest, and to then retrieve relevant best practices, production
manuals, quality instructions, links to experts, etc.
In summary, social processes provide the basis through which knowledge is captured and shared;
intervention points are those process phases where social processes may be of most value (the most
opportune time for capturing and sharing knowledge that can influence a decision); results captured
from social processes (using a best practice capture template) act as the main ingredients for the
best practices database; the best practices database acts as “knowledge base”that is connected to the
product model (semantically rich parametric model of building components, materials, geometry,
etc.) of a building.
Conclusions
Within project based industries such as construction, valuable experience and knowledge is lost
primarily because of a lack of proper capturing and sharing mechanisms. This paper advocated the
capture and sharing of knowledge through a collection of “social processes”that were set within a
particular context, i.e. they were defined for a particular purpose within a given setting. Three such
social processes covering site visits, cross audits, and horizontal quality and process teams were
presented. All three covered the identification, capturing and sharing of knowledge between
individuals and groups. The output of these social processes is a set of good and bad practices that
are recorded in a pre-defined template for further sharing, re-use, and the development of new
knowledge. Social processes may be triggered at key intervention points within an organisation’s
business processes. These business processes or sets of processes are where a major decision or
change was to take place. They form the most relevant time to capture new knowledge, discuss and
solve problems with different stakeholders, and identify and document a new set of good and bad
practices. Organisations wishing to take advantage of their valuable knowledge assets are
encouraged to identify critical intervention points within their processes and to position relevant
social processes within them to facilitate the capture of new good and bad practices. These may
then be propagated (shared) further through a database pointing to the practices, the source of the
practice, and the key experts and information sources. The feedback received within the PromoBat
project has been positive with regard to social processes and intervention points. Efforts are
currently underway to create frameworks for more social processes and identify intervention points
in more detail. These will be used as a basis for populating a new best practices database that will
then be connected to YIT’s building product models for ease of knowledge resource access on a
product component level.
76
Practical Tips and Lessons Learned:
•
Within project-based industries such as construction, valuable knowledge is
created within projects, but is often lost as it is not efficiently shared. There is a
need to identify mechanisms (even rewards) that allow for the capture and
sharing of this knowledge (experience).
•
Knowledge is best captured and shared within a context (such as a common
goal). Such a context could be created within a particular event, workshop or
even meeting. If there is no context, limited capture and/or sharing of
knowledge will take place.
•
Social processes are very good instruments to enable knowledge capture and
sharing through social interaction. They could as an example (cross-audits)
offer different functional units to learn from each others’experiences and work
practices.
•
The type of social process required to capture and/or share knowledge will vary
significantly based on what is to be captured and/or shared within a given
context. Special social processes may need to be designed for a particular
context and the setting within which it will take place (e.g. the world café
format).
•
Social processes on their own are not enough. There need to be clear directions
as to when, where, and in which context they occur. One good option is to
embed these social processes within an organisation’
s business processes at
those locations where the opportunity to capture and/or share knowledge is
maximum and where the social processes may influence a key decision.
•
The skilled workforce is fast retiring, and with them the valuable experience
that they have built over the years. Mechanisms need to be introduced to learn
and share these experiences. Social processes may be used as one such
mechanism for sharing of experiences.
•
Knowledge that is created, captured and/or shared needs to be widely
disseminated. This could be through self forming communities (of practice), or
through the embedding of links to these knowledge resources within other
business applications (as an example, best practices, information on experts,
knowledge resources, etc. can be attached to production models or even
process models).
Acknowledgements
The PromoBat project was launched by YIT Construction Ltd. with some financial support from TEKES
(National Technology Agency of Finland) and consultancy services from VTT (Technical Research Centre of
Finland) to serve as a means for the identification and capture of good practices. Gratitude and appreciation
is expressed for their support. The authors wish to acknowledge the support and efforts of all stakeholders
involved during the project, in particular the site personnel who shared their time, experiences, and created a
context for both the capture and sharing of good and bad practices.
77
References
Afsarmanesh, H, Garita, C, Hertzberger, L.O. and Santos-Silva, V. (1997) Management of
Distributed Information in Virtual Enterprises –the Prodnet Approach. Proceedings of the 4th
International Conference on Concurrent Engineering, Nottingham, UK.
Awad, E.M. and Ghaziri, H.M. (2003) Knowledge Management, Prentice Hall.
Browne, J., Sackett, P.J. and Wortmann, J.C. (1994) The System of Manufacturing: a Perspective
Study, Report to the DG XII of the CEC, European Commission.
Charbuck, D. and Young, J.S. (1992) The Virtual Workplace, Forbes, 150(12) 184-190.
Davenport, T.H. and Prusak, L. (2000) Working Knowledge: How Organisations Manage What
They Know, Harvard Business School Press.
Kazi, A.S. (2005) Knowledge Management in the Construction Industry: a Socio-technical
Perspective, Idea Group Publishing.
Kazi, A.S. (2004) Parametric Analysis of Different Project Planning and Control Functions, AIT
Dissertation, CM-04-01, Asian Institute of Technology, Bangkok, Thailand.
Kazi, A.S., Puttonen, J., Sulkusalmi, M., Välikangas, P. and Hannus, M. (2001) Knowledge
Creation and Management: the Case of Fortum Engineering Ltd., Knowledge Management in
the Sociotechnial World: the Graffiti Continues, Springer-Verlag, pp.153-169.
Koivuniemi, A. (2004) Capturing Best Practices in Project-based Construction Business –the Role
of Social Processes. Master’s thesis, The Institute of Business Information Management,
Tampere University of Technology, Finland.
Nonaka, I. and Takeuchi, H. (1995) The Knowledge Creating Company: How Japanese Companies
Create the Dynamics of Innovation, Oxford University Press.
Nonaka, I., and Konno, N. (1998) The Concept of “Ba”: Building a Foundation for Knowledge
Creation, California Management Review, Vol. 40, No.3, pp. 40-54.
Project Management Institute (2000) A Guide to the Project Management Body of Knowledge,
Project Management Institute.
Walton, J. and Whicker, L. (1996) Virtual Enterprise: Myth and Reality, Journal of Control, 22(8),
22-25.
78
Dr. Abdul Samad (Sami) Kazi, senior research scientist at VTT, the Technical
Research Centre of Finland is a Civil Engineer by profession, and Construction
Manager by area of specialisation. Sami has conducted research in several
areas including: project planning and control, computer based training,
construction management, knowledge management, system dynamics,
construction IT, information modelling, product and process modelling, cost
planning and control, Internet based technologies, groupware tools and
technologies, mobile communications, information modelling, mind mapping,
concurrent engineering, virtual enterprises, organisational learning and
assessment. Sami has edited several books in the areas of ICT for construction
projects and processes, knowledge management, systemic innovation in
construction projects, etc. He is listed in the Marquis Who’
s Who in Science and
Engineering, is a fellow of the Entovation E100 Global Knowledge Leadership
network, and is content steward at KnowledgeBoard.
Anssi Koivuniemi, 27, works for the Finnish construction and service company
YIT. He earned his M.Sc. (eng.) from Tampere University of Technology,
Finland, with a major in Information and Knowledge Management. He has
worked internationally both in YIT company’
s internal and external projects on
developing and implementing knowledge management activities such as
knowledge capturing and sharing. It has been Anssi’
s responsibility to develop
and organize various personnel trainings and knowledge transfers between
business segments and business units both domestically and internationally.
79
Addressing
Communication Silo’
s
using Complexity
Techniques and Social
Network Analysis
Sonja Blignaut
5
s
using Complexity Techniques and
Social Network Analysis
Sonja Blignaut, Senior Consultant, IBM Business Consulting Services South Africa
([email protected])
Abstract
The Information Management (IM) division of a large financial
institution in South Africa was struggling with functional silos that
prohibited their ability to leverage the strengths of their different areas
of specialisation. As part of the establishment of a Knowledge
Management Center of Excellence (KM CoE), a project was
implemented to use Cynefin methodologies (specifically Community
Social Network Analysis (SNA) and Cynefin Categorisation) to
investigate the root causes for this problem, and to attempt to address
it. One year later, a follow up SNA was completed to determine the
impact of the initial SNA and the initiatives that followed.
The results of these two initiatives were compared by an independent
team of consultants.
The 2004 SNA confirmed the existence of silos in the environment,
and further more provided a view of where the specific issues were.
The Cynefin categorisation indicated that perspective bias differences
were part of the underlying reason for the perpetuation of the silo
behaviour. The two result sets provided information that allowed the
team to design interventions targeted at specific problems.
The 2005 SNA results showed definite improvements in the network,
with the silos still in evidence, but with much more cross-silo
collaboration taking place.
Keywords: Social Network Analysis, Financial Services, Perspective Bias, Cynefin Background,
Cross-silo, Community
Background
Industrial Context
In 2001 Knowledge Management (KM) was selected as one of four strategic drivers of Intellectual
Capital in one of the ‘big four’banks in South Africa. It was later identified as one of twelve core
capabilities that the organisation would be focusing on as part of their strategy, and was also chosen
as one of 6 core capabilities that needed to go through a process of accelerated development to
ensure sustainable success.
83
s using Complexity Techniques and Social Network Analysis
A strategic decision was made to initially focus on the creation of a central KM capability in the
group, not the implementation of KM throughout the group. A KM Center of Excellence (COE)
was established within the Information Management (IM) division, focusing on two aspects of KM:
•
Technology enablement and;
•
Mobilisation and consulting, which is focused on Knowledge strategy formulation and the
‘human’aspects of KM.
The focus of the Technology Enablement area has been on the centralization of the content
management capabilities in the group, as well as establishing and entrenching content management
governance throughout the group.
The mobilisation and consulting team of the newly created COE, assisted by IBM, initiated the
development of their service offering, governance framework and roadmap going forward. An
overarching Knowledge Strategy was created, which is currently being implemented through
various initiatives that were identified as part of the strategy formulation process.
The KM methodology of choice for the organisation is Dave Snowden’s ‘Just-in-Time KM’
(Snowden 2002a and 2002b), which exploits the dualistic nature of knowledge (knowledge is
simultaneously a ‘thing’and a ‘flow’). This methodology is primarily based on three heuristics:
•
Knowledge can only be volunteered, never conscripted
•
I only know what I need to know when I need to know it
•
We always know more than we can say, and we always say more than we can write down
These heuristics led to a focus on the following KM elements:
•
Context management: Social network analysis/stimulation, expertise location, communities,
workplace design, knowledge culture mapping
•
Narrative management: Narrative enquiry, narrative database, Patterning, Story, Knowledge
transfer processes
•
Content management: Document management, portals and search/taxonomies
Proof of concepts for most of the CoE’s service offering has now been completed, including two
community SNA’s within the bank’s IM division. These initiatives have had a positive impact on
the environment, and they led to more and more requests for assistance flowing into the KM CoE
from the rest of the group.
This case study will focus on the projects that were run in the IM division of the bank, as part of the
process of establishing the KM CoE.
Problem Statement
The IM division in this bank comprises several largely independent sub-divisions. The main focus
areas of the division include maintaining and mining an international award winning data
warehouse and providing customer analytics data to the rest of the group. Because of the highly
specialised nature of the different sub-divisions, silo behavior was very prevalent in the
environment. This led to many lost co-operation and innovation opportunities.
A need existed to better understand the nature of the network ties between these sub-divisions to
enable the KM team to design interventions to start breaking down these silos and to increase
collaboration in the division.
The project also served as a proof of concept for Social Network Analysis and the Cynefin
Perspective Bias methodologies, which are components of the KM CoE’s service offering to the
84
rest of the group. To limit scope, it was decided to focus only on network connections within the
formal communities (internal sub-divisions) within the IM division.
•
Explore the advantages and pitfalls related to the application of Cynefin
techniques in large corporations
•
Evaluate Social Network Analysis and its applications
•
Determine the value of Cynefin perspective bias as used in conjunction with an
SNA
Approach
Our approach involved the comparison of two sets of SNA results data that were collected during
two separate SNA initiatives run exactly 12 months apart. The SNA graphs, as well as the
statistical results data were compared and analysed by the KM project team.
The Cynefin approach to SNA was selected because of its ability to minimise the risk of political
gaming when completing SNA questionnaires. The more traditional approach to SNA involves
investigating the connections between individuals, the Cynefin approach is to focus instead on the
relationships between identities, which typically include formal and informal communities as well
as prominent roles or functions. One of the key benefits of this approach is that questionnaires are
completed collectively by groups of people. The results are therefore more reliable, as much of the
risk for data contamination by personal agendas and political positioning is removed.
Another benefit is that the impact of context is minimised. When conducting an individual SNA one
is often faced with the problem that there is confusion between an individual’s business and
personal relationships with their colleagues. Often this confusion leads to skewed results. As the
Cynefin approach is centred on the collective perspective, this issue is largely minimised.
That being said, one of the main criticisms we had on the Cynefin SNA approach was that the
results aren’t granular enough, i.e. where the results were not what people expected i.e. connections
existed where they weren’t expected to be, decision makers would have like to be able to drill down
to an individual level to gain more insight into that specific interaction. There may therefore be a
case to be made for a combined approach, possibly applying the Cynefin SNA approach during the
initial phases of a project, followed by an individual SNA focussed on gaining a deeper
understanding of identified problem areas.
The team decided to run a perspective bias initiative concurrently to the second SNA on order to
gain a deeper level of insight into possible reasons for the existence of silo’s in the environment.
The standard Cynefin categorisation process was followed to determine the bias differences
between the communities in the division, the KM team analysed the results from the Cynefin
perspective bias initiative in consultation with Dave Snowden, as this is a very new area and the
work that was done were often ground breaking.
Case Analysis
The methods used in this case study are based on the Cynefin methodology, provided by Dave
Snowden and the Cynefin Centre.
In short, the Cynefin methodology is based on social complexity principles, and focuses on
Narrative, Networks and Sense making. It is built around a sense making framework, which
distinguishes between ordered and un-ordered systems.
85
The project team used the Cynefin Community SNA method, where the SNA is conducted between
‘identities’(in this case, formal communities) and not individuals. According to Snowden (2005)
this approach minimizes the risk of individuals deliberately influencing the results according to
their own agendas.
An initial SNA was conducted in May 2004 with a follow up SNA planned for May 2005 – this
would enable the team to determine the impact the SNA and other subsequent interventions had on
the environment.
As a result of the initial SNA, a Cynefin Perspective Bias exercise was run in the environment to
investigate the communication disconnect between communities, which were more technically
orientated vs. communities with a strategic focus. The SNA also showed that many of the
communities didn’t know what expertise existed in other communities, which led to a pilot project,
using an individual SNA (Cross, Parker 2004) to identify experts in one of the communities in the
IM division.
Comparative Results – 2004/2005 Social Network Analysis
It was decided early on in the process to focus the SNA on formal communities only and to conduct
it across 4 dimensions – learning, meaning, influence, and warmth. Questionnaires were designed
and sent out to the formal community leaders and the communities were given the choice of
electing a proxy to complete the questionnaire, or to complete it as a group. Many communities
chose to elect a proxy, which in the end proved detrimental to the process, as the results were
invariably skewed according the proxy’s perspective. During the follow-up SNA, communities
only had the option to complete the questionnaires as a group during a facilitated session – this
eliminated the problem experienced during the first SNA.
The project team collected and collated the SNA results; the process included capturing the
questionnaire data and generating network statistics (cohesion distance, density and degree
centrality) and diagrams using UCINET software.
It was decided to focus only on meaningful interactions; the data was therefore dichotomised on
values greater than 4, to ensure that only high ranking interactions were shown in the graphs.
Once the data was collated and all the diagrams drawn, sense-making workshops were held with
three groups, being the:
•
Project team –objective analysis of the diagrams and statistics based on theory and patterns
only;
•
IM Executive team – subjective analysis (EXCO perspective) of diagrams, statistics and
patterns identified by the project team;
•
IM staff (proxies from each of the communities) subjective analysis (staff perspective) of
diagrams, statistics and patterns identified by the project team;
We discovered early on the importance of obtaining inputs from all three of the above mentioned
groupings on the SNA outputs, as it provided us with valuable perspectives about the reasons for
the existence of many of the patterns in the graphs. Often a pattern that may at first seem negative
can be justified when the context is known.
In the following section, we will look at the comparative results of each of the four dimensions in
detail.
The following statistical measures were used in this study:
1. Network density – indicates the robustness of a network. The density figure reflects the
number of connections that exist out of 100% possible connections in a network. More
86
connections means quicker and more accurate information flow, however, too many
connections may have a negative impact on efficiency in the network
2. Degree centrality –may indicate influential nodes in the network
3. In-degree –number of arrowheads coming into a node
4. Out-degree –number of arrowheads coming from a node
5. Network cohesion or distance – indicates the ease with which a network can connect.
Distance is shortest path between two nodes. Aggregate measure at network level reflects
the average distance between all nodes in the network.
The names and acronyms for several of the communities changed during the period between the
two SNA’s. The nodes on the network diagrams are color coded to indicate related communities.
Table 1: An indication of community name changes from 2004 to 2005
2004
Communities
Related 2005
Communities
IMOSAD
SAD
IMOIB
IB
IMODWM
ETL
KMMCS
KMCS
SIE
BDM
BCE
PO
BPM
MIS
BI
BSC
Dimension 1 - Learning
Question: Most of us obtain information and / or learn from various groups of people within our
organisation. How much does your community depend on each of the following groups or
communities for information and / or learning?
Results focused on the top two values on the response scale (‘Dependent & Highly dependent’) –
i.e. meaningful ties
87
2004
2004 - Network Measures
Density = 12.86%
Distance based cohesion = 0.180
Average distance = 1.833
2005
Density = 28%
Figure 1: Comparative results: Learning Dimension
The 2004 SNA results for the learning dimension was particularly concerning, as several of the
communities were completely isolated. This meant that some of these communities were
completely cut off from the rest of the network, implying that the rest of the network was unaware
of their value, and therefore they were being marginalized.
Overall the 2005 SNA results show a marked improvement: there are no more isolates, although
one or two of the communities (e.g. the personal assistants (PA)) are only connected to the network
through one other community (EXCO – the community they report to). Network density has
increased significantly, from 12.86% in 2004 to 28% in 2005. This indicates that information flows
more effectively through the network. Cohesion has also increased from 1.83 to 2.53, which seems
to indicate increased robustness in the network.
In both the 2004 and 2005 results a strong triad relationship is clear between the IM Operations
communities - IMODWM (ETL in 2005), IMOIB (IB in 2005) and IMOSAD (SAD in 2005).
These communities together form the operational hub of the IM division, responsible for the
development and maintenance of the data warehouse, as well as the provision of critical reports to
the rest of the organisation. The strong collaborative relationship between these three communities
is clear across all the SNA dimensions, and together they form a highly influential hub in the
network. What was concerning in 2004 is how far the EXCO community was removed from this
operational hub. The 2005 results show improvement, with a link to the hub via the IMA (architect)
community. This is a very positive development, as it allows for the IMA community to translate
technical information into a language that EXCO can relate to.
These improvements are largely due to initiatives implemented by the managers of the relevant
communities and the KM team based on the 2004 results. Brown-bag lunch sessions (termed IMTalk) are now held on a monthly basis with the main purpose to provide an informal knowledge
sharing platform.
88
Another reason for the improvement in this dimension is concerted effort from the team leaders to
encourage knowledge sharing between their respective communities. Having been confronted by
undeniable proof of where their teams are in the network, they had a better idea of where to focus
their energies. A good example is the IM Architecture (IMA) community (circled on the diagram).
In 2004 the IMA community was a learning ‘sink’i.e. they many outward directed connections
(they went to other communities for learning), but no incoming connections (none of the
communities came to them for learning). In spite of being very centrally placed in the network,
they were like sponges – learning from many other communities, but never passing on the
knowledge.
The manager of the area put specific initiatives in place to address this issue, including adapting the
performance measures for the team to include specific measures to ensure that all deliverables were
distributed to other communities, and that these deliverables were understood by relevant parties.
As can be seen by the 2005 results, these initiatives had the desired effect, with IMA now an even
more central community in the network, and a significant number of incoming arrows indicate that
many of the other communities are now learning from them.
A negative pattern in the 2005 results that needs to be addressed is the high dependence on
contractors and consultants (CC) for learning. The KM team is currently developing knowledge
transfer processes to ensure that critical knowledge is transferred to permanent staff members
before the consultants leave the environment.
Dimension 2 - Meaning
Question: When you come across a work-related problem which you cannot make sense of, how
likely are members of your community likely to turn to each of following communities or their
members for assistance?
Results focused on the two top values in the response scale (‘Likely & Very likely’) – i.e.
meaningful ties.
The meaning dimension is an important dimension as it shows the network ties for a deeper level of
interaction than simply information exchange. This dimension indicates collaboration and joint
sense-making between communities. The 2004 results showed an isolated sub-network; what was
concerning about this was that the Executive (EXCO) community, who are the main decision
makers in IM, was part of this sub-network. Once again, this community was completely cut off
from the operational hub. These results were corroborated by a very real issue in the environment
around strategy formulation and communication. Many of the employees felt that they weren’t
included in strategy formulation, and therefore the Executive found it very difficult to obtain
effective buy-in into their strategy.
Based on the results of the 2004 SNA, the Executive team focussed on adopting more of a bottomup approach to strategy formulation. Subsequent feedback from other communities confirmed that
their experience of the strategy formulation process was a great deal more positive in 2005. The
2005 SNA results showed a definite improvement in the network on the meaning dimension.
Although the network density is still quite low, all communities are now connected, and many of
these communities now feel comfortable to engage with the EXCO community when they need to
make sense of work issues.
89
2004
Density = 16.19%
2005
Density = 22.86%
Figure 2: Comparative results: Meaning Dimension
Another community that shows marked improvement is the SIE community (the name of this
community changed to BDM in 2005, circled with a dashed line on the diagram). This community
was established in 2004 to promote and sell the IM division’s service offering to the rest of the
group. The 2004 result indicated that they were very much on the perimeter of the network, only
connected to the rest of the network through one other community. This was concerning, as they
needed to have a clear view of the roles and services of all the other communities in order to fulfil
the role they were created for. In 2005, we can see a dramatic improvement in their positioning in
the network. This is once again due to targeted actions initiated by the relevant community leads.
As in the learning dimension, high dependence on Contractors and Consultants (CC) remains a
problem in the meaning dimension.
90
Dimension 3 - Influence
Question: Our actions and decisions are often influenced by others. How much impact does each
the following communities have on the work and decisions of your community?
Results focused on the top two values in the response scale (‘Great & Very great’) –i.e. meaningful
ties
2005
2004
Density = 16%
Density = 22.86%
Figure 3: Comparative results: Influence Dimension
As in the previous two dimensions, most concerning pattern in the results for the influence
dimension was the positioning of the EXCO community in the network. As their main role is to
provide strategic direction, their positioning seemed to indicate that 1) they weren’t being
influenced by many other communities in the network, i.e. they didn’t consult their experts enough,
and 2) they had very little influence over the rest of the network.
The most influential communities in the network were the three operational communities. CC –
contractors and consultants was also an influential community, which was concerning from a risk
perspective as this community is made up out of non permanent staff members. Their influence is
slightly less in 2005, which is a positive trend. The main reason for this is a successful recruitment
drive which brought many new permanent staff members into the environment.
In 2005, the EXCO community became more central to the network, and seems to have built
stronger relationships with other critical communities.
The IMA community also became much more influential in the network, a necessary change, as
they are the architects who design the blueprints for new systems and products that other
communities need to implement.
91
Dimension 4 - Warmth
Question: As we interact with others, our comfort level varies.
community when it interacts with each of the following communities?
How comfortable is your
Results focused on the top two values in the response scale (‘Comfortable & Very comfortable’) –
i.e. meaningful ties.
2005
2004
Density = 46%
Density = 62%
Figure 4: Comparative results: Warmth Dimension
Warmth is the dimension that is most open to political game playing. Very few people are prepared
to indicate discomfort with someone else, however conducting the SNA on community level
mitigated part of the problem. It is interesting to note that the density in this network is
significantly higher than in the other three dimensions. This seems to indicate that people in the
network are very comfortable interacting on a superficial or social level, but less so when having
meaningful, collaborative interactions.
This is the only dimension where the density decreased from 2004 to 2005. A possible reason for
this is the high number of new employees that entered the environment during this period. It stands
to reason that comfort levels will decrease with the introduction of many unfamiliar people.
Currently the project team is investigating ways of leveraging off this highly connected network to
address some of the problems that were found in the other dimensions. ,Is there a way to
incorporate learning activities in informal sessions, such as the Friday afternoon open bar, for
example?
92
Using Cynefin Methods to Expose Perspective Bias Differences
The results of the initial SNA indicated disconnected communication flow between many of the
communities in IM, but most notably between the EXCO and IMA (architecture) communities and
the rest of the network. As these are the primary strategically focused communities in the
environment, the project team felt that this issue was worth further investigation, as it probably had
a big impact on strategy development and implementation in the division. The team decided to
investigate how big the role was that perspective bias differences between these communities had in
this communication problem.
A case study of where Cynefin methods were used to expose perspective biases between
government entities in Singapore, led to the selection of the Cynefin Categorisation method for use
in this initiative.
What is Perspective Bias?
“We don’t see things as they are. We see them as we are.”, Anais Nin
We all make sense of the world differently, and therefore we react and communicate accordingly.
This often leads to difficulties or conflict when people (or communities) who view the world
differently try to communicate with each other.
There are two aspects to perspective bias:
•
Difference - this is not necessarily negative, as people in different positions often need to
approach the world differently, e.g. someone working with data and statistics needs to be
more ordered than someone managing people.
•
Dissonance – this is often negative as it shows that the same concepts are understood and
interpreted differently by different communities. This could lead to communication
disconnects and therefore a very a slow response rate in times of crisis in an organisation as
different communities attempt to reach consensus.
The Cynefin framework (see Figure 5) is a sense making framework based primarily on order vs.
un-order. It has 5 domains - visible order, hidden order, complex un-order, chaotic un-order and
disorder.
When a single dataset is categorised into this framework by several different communities, the
results indicate differences in the way that the communities view and understand the concepts
represented by the data points. The different distribution patterns reveal the bias differences
between the communities. Looking at where specific data points were placed by each community
reveals bias dissonance.
93
Un-order
Order
Complex
Complicated
CàE
CE
(Discoverable)
(coherent in retrospect)
Issues are not transparent but
become clear through analysis
Issues have multiple
root causes
Emergent Solutions
Disorder
Chaotic
C¦ E
(No perceivable relationship)
Several good solutions
Simple
C=E
(Known/Obvious)
Issues are familiar and
obvious
Issues catch you completely
off guard
One best solution
No stable solution
Figure 5: The Cynefin sense making framework
Approach
During a workshop (attended by proxies from each of the formal communities), the Cynefin
framework was introduced to the participants. Once we were sure they understood the framework,
they were asked to brainstorm around concepts and issues etc. which were present in the IM
division that they felt fit into each of the Cynefin domains. The output from this session was a fully
categorised dataset, which represented the collective perspective of all the communities in the
division.
The dataset was reproduced on stickers, and workshops were arranged with each of the
communities where they were asked to categorise the concepts into the Cynefin framework from
their own perspective. This resulted in a set of categorised frameworks from the perspective of
each of the individual communities.
The project team collated the information, and compared the placement of the data points by each
of the communities to discover the primary bias of each of the communities. The Cynefin domain
into which the majority of the data points were categorised indicates the primary bias for the
community (Figure 6).
An interesting result (although by no means unexpected when one considers the environment) is
that the majority of the communities showed a strong bias towards the ordered domain. There were
some communities (e.g. EXCO) with a strong bias towards un-order, specifically the domain of
Complex Un-order, which once again is to be expected of a strategically focused community.
These results shed some light on another reason for the problems that the Executive had when it
came to strategy formulation and communication. It was quite probable that the language they used
when communicating to the rest of the communities was too abstract, not specific enough for the
ordered communities to buy into. Often communities with an ordered bias are very task-orientated,
they get frustrated when communications are vague and abstract; they want to know the specifics
94
and get on with it. Communities with a bias towards complexity are usually more comfortable
communicating on higher levels of abstraction; they tend to see the bigger picture, and think
strategically. It is often difficult for them to translate their ideas into clear and concise messages
that would resonate with the ordered communities.
Un-Order
Complex
Order
High Abstraction
Complicated
•IMA (19%)
•EXCO (26%)
•KMMCS (26%)
•KMTE (30%)
RCM (28%)
•BI/BCE (35%)
•BI/BCE CC (34%)
•IMODWM (27%)
•IMOIB (30%)
•IMOSAD (28%)
•KDD (33%)
•PA (23%)
•BDM (34%)
Chaotic
Low Abstraction
Teaching Cultures
Learning Cultures
•BPM (30%)
Simple
Figure 6: Bias difference results based on primary bias percentage per community
The secondary bias percentage for each community was also taken into account, as it allowed the
team to identify bridging communities that could be leveraged as ‘mediators’between communities
with incompatible primary biases. The KMCS (Knowledge Management Consulting Services)
community is a particularly important bridging community, as they straddle three of the four
domains (simple, complicated and complex). This indicates that they are perfectly suited to their
role as KM consultants, but in addition, the community could be leveraged by other communities to
assist with communication difficulties.
95
% Bias1 Bias2
Legend:
Simple (Visible order) = V
Complicated (Hidden Order) = H
Complex = CX
Boundary between Visible & Hidden Order = B
KMMCS
KMTE,EXCO, IMA, BPM
BDM, RCM, PA, BI/BCE,
BI/BCE CC, IMOSAD, IMODWM
21 (H)
24 (H)
18 (H)
12 (H)
19 (CX)
19 (B)
18 (CX)
24 (B)
16 (CX)
21/ 21 (H/V)
28 (CX)
16 (H)
20 (H)
24 (H)
D
KD
35 (V)
34 (V)
30 (CX)
26 (CX)
19 (H)
27 (V)
30 (V)
28 (V)
33 (V)
26 (CX)
30 (H)
23 (V)
28 (B)
34 (V)
B,
OI
IM
BI/BCE
BI/BCE CC
BPM
EXCO
IMA
IMODWM
IMOIB
IMOSAD
KDD
KMMCS
KMTE
PA
RCM
BDM
Placement of communities on the framework based on
primary and secondary bias percentages
Figure 7: Bias difference results based on primary and secondary bias percentages per
community
To determine bias dissonance between the communities, the placement of specific data points was
compared.
When considering bias dissonance it is important to note that four of the Cynefin domains (simple,
complicated, complex and chaotic) may each be divided into four sub-domains.
•
Stable - the central area in each domain is the area where concepts are viewed as being
stable in that domain, i.e. the concept has always been in that domain and can be reasonably
expected to remain stable. Understanding aeronautical mechanics is and will always be
complicated, for example.
•
Extreme - Towards the furthest outer corner of each domain, is where the concept is
categorised if it is viewed as being extreme, almost to the point of becoming unstable in that
domain. E.g. if a process in the ‘simple or visible order’domain is over constrained, it may
collapse into chaos.
•
Boundaries - There are two boundary conditions for each of the domains, which indicate
that a concept may be viewed as transitory between the two domains on either side of the
boundary.
Figure 7 shows an example of the dissonance results in the environment on the concept of ‘decision
making’. From the categorisation results it seems that when the different communities have to
make joint decisions, it may be problematic. Many of the communities view decision making as an
ordered task, i.e. if you’ve done your research and you’ve got the relevant information, making the
decision should be simple. Other communities view it as an un-ordered task, and may seem to be
procrastinating while they weigh all the options.
These different perceptions and decision making styles manifest in the environment as a perceived
lack of decision making capability by the workers on the ground. It is interesting to note that none
of the communities classified this data point as being in the Extreme part of any of the domains.
Many communities saw it as being transitory between two domains. Other data points with
significant dissonance include: Collaboration, teamwork, leadership, management, customer,
customer service, delivery and measurement.
96
To address this problem, the KM team is currently implementing an initiative where the
communities will collectively participate in defining a ‘common language’for the division. To this
end, several workshops will be held where the participants will collectively agree on the meaning of
each of these concepts. These definitions will then be diffused into the rest of the division to ensure
that, when a discussion occurs about a customer (for example), everyone is on the same page and
talking about the same thing.
BI/BCE CC
Complex
BPM
IMOIB, KMTE,
KDD
BDM, IMA, IMODWM
Complicated
EXCO, BI/BCE,
KMMCS
RCM, IMOSAD
PA
Simple
Chaotic
Figure 8: Bias dissonance results on concept of Decision Making
Key Findings
Since every diagnostic is an intervention in itself, simply participating in these initiatives changed
the environment significantly. While this is recognised by most of the participants, the biggest
challenge the project team faced during this process was related to the measurement of the specific
impact and derived value of each of these initiatives.
The initiatives discussed in this document weren’t conducted in a vacuum. Many other changes
occurred in the environment at the same time, e.g. new people joined, others left, and other
initiatives were run during this same time period. It was therefore extremely difficult to determine
how much the SNA and the resulting initiatives contributed to the positive changes in the
environment.
Results of the SNA confirmed the existence of silos in the environment, and although management
was always aware of these silos, they had no clear view of where the biggest problems were, or
similarly where pockets of cross-silo collaboration existed. The SNA results provided them with
critical information to design targeted interventions in order to address specific problems. It is
important to note that it is never the objective to connect all communities in the social network to all
97
other communities, as this would significantly hamper the effective functioning of an organisation.
A more effective approach is to determine which communities should be connected, and to focus on
creating only those connections which will benefit the business.
The Perspective Bias exercise provided further insight into the communication disconnects that play
a role in the perpetuation of these functional silos. Awareness of the fact that the different
communities view the world differently had an immediate effect on communication in the
environment.
One of the main issues that were highlighted by the 2004 SNA was that the Executive community
(EXCO) was largely cut off from the rest of the network, across the key dimensions of learning,
meaning and influence. This was further substantiated by the results of an independent quantitative
study conducted in the division prior to the SNA. These results showed that the strategic direction
of the division (set by the executive) wasn’t well understood by the rest of the staff. Specific
interventions were put in place to address this issue, with the executive focussing on being more
accessible to staff members, and on communicating in ways that resonated with the environment.
The positive impact of these interventions was clear in the 2005 SNA results, which revealed that
across all dimensions the density of the networks have increased dramatically. The EXCO
community has increased their network ties in terms of learning and meaning. Subsequent feedback
from the environment confirmed that the overall perceptions of the EXCO community and the
division’s strategy have become more positive.
Not all of the projects resulting from these initiatives were focussed on intangibles. The SNA
results made it clear that the different communities did not understand the value of each of the other
communities. Further investigation showed that there was general confusion in the division
regarding who the experts were that they needed to consult with in each of the communities. An
initiative was completed in August 2005, where an individual SNA was combined with a peer
review process to identify experts on relevant topics in one of the communities. The results of this
initiative was used to determine succession plans for critical areas and was so valuable that the
methodology is now being formalised to be re-used in other areas of the bank.
Another consistent theme across all of the SNA dimensions is the high dependence on consultants
and contractors. This is an ongoing problem, which is largely caused by a shortage of specialised
skills in South Africa. There is no real solution to this problem, but the business impact can be
minimised by ensuring that proper knowledge transfer processes are in place to make sure that
critical knowledge is retained even when the consultants leave the environment. Establishing these
processes is another project that resulted from the SNA.
The general feeling within the division is that there has been a definite improvement in
communication flow, and while silos prevail, they are less prominent than a year ago. Crossdivisional communication has increased, and this has led to better collaboration and the ability to
leverage the full potential of the IM division when responding to client needs.
Business Impacts
Combining information about the flow of knowledge and information through the existing
networks, and perspective differences within those networks, provides a level of understanding that
only one technique does not provide.
Within highly specialised environments (in all industries), silo cultures are a pervasive problem that
stifles innovation and leads to loss of sales and growth opportunities. This project proves that silo’s
can be broken down by utilising techniques that have relatively little time impact in the
environment. The techniques also expose the underlying causes for these silos, and allow
organisations to not only address them superficially (the symptoms) but to address the underlying
causes.
98
If perspective bias dissonance is exposed and addressed, it enables an organisation to become more
agile in most any situation. When faced with a crisis, it should be able to respond much quicker
than its peers and competitors, as all the key decision makers should have a common understanding
and shared context around key concepts. It is also useful to know which of communities are
strongly biased towards specific domains, and therefore are comfortable in those domains. For
example, communities with a bias towards the complex or chaotic domains may be very effective in
a crisis, as they are more comfortable than other communities when faced with a chaotic situation.
Similarly, it will be more efficient to assign administrative responsibilities to communities with an
ordered bias.
Conclusions
Some areas of the environment saw significant positive changes after the initiatives were
completed. The silos, although they are still there, are much less in evidence. Internal
communication has become much more effective, and many of the communities are collaborating
with each other and creating new product offerings together.
Articulating the real value of these initiatives is an ongoing challenge. The derived value typically
tends to be intangible, and often it is not what revenue-focussed executives expect. In a taskorientated environment such as the IM division, where most of the discussions are centered on data
and systems, conducting a Cynefin intervention based on exposing and addressing softer issues, was
a definite challenge. A big frustration for the project team was that, despite the fact that the
majority of the participants could see the value that the initiatives brought to the division; very few
of them could clearly articulate that value. Simply creating a forum where decision makers and
staff could discuss these issues made a big difference, but how does one place a value on such a
discussion?
Subsequent to the initiatives mentioned in this case study, the project team conducted interviews
with key stakeholders to get their perspectives of the value of these initiatives, and to gain a view of
how they’ve experienced participating in a Cynefin project. The overall feeling was that a lot of
value was derived simply from participating in the process.
Key points that were raised include:
•
Participants generally indicated that most of their expectations were met. They found that
the SNA provided them with a view of the organisation that they’ve never considered before
and they felt that the method had the capability to expose the root causes for many of the
existing problems in the environment. (Reads difficult, especially the first fragment after the
bullet.)
•
They were concerned that in some cases the results of the community SNA weren’t granular
enough and expressed an interest in investigating the implementation of an individual SNA
in the environment.
•
A general feeling was that the biggest value of the SNA and the Cynefin Categorisation was
the creation of a general awareness of the underlying issues, and that this awareness led to
sub-conscious behavioral changes that are difficult to articulate.
•
The discussions that occur as part of the sense making process already lead to changes in
behaviour and therefore changed the system.
•
SNA highlights key areas to focus on to change behaviours in the system.
•
Cynefin interventions provide a different perspective – focusing on softer issues and
providing a mechanism to deal with those issues. In task orientated environments this is
especially valuable as it assists leaders to influence the culture in their environment.
99
The Cynefin methodology is very different to other traditional consulting methods. One of the main
differences (and a key strength) of the methodology is that it isn’t based on attempts to prove a predetermined hypothesis. Often when we conduct an initiative to prove such a pre-existing
hypothesis, we end up not seeing the real problem, because we try to interpret the findings in such a
way as to prove our own thinking. Cynefin methods allow the real issues to emerge from the
environment, and as such provide a platform to air issues that may otherwise have remained hidden.
Because the approach is unfamiliar and different, participants usually receive it very positively and
it definitely assists in getting to the root cause of complex issues. Because the process is often
ambiguous, the risk of people ‘gaming’the process to ensure desired results is minimised.
Cynefin remains a difficult concept to sell, especially because of the difficulty in measuring
tangible benefits. In subsequent projects, we’ve ensured that stakeholders are aware that most of
the benefits will be intangible right up front. We’ve recently started a project where we hope to use
SNA and other Cynefin methods to address high levels of staff attrition in another area of the bank.
In this instance we hope to be able to show real tangible benefit.
•
Key measurements should be agreed up front to make sure that value derived
can be quantified. If the derived benefits will be mostly intangible, make sure
that stakeholders agree to this up front.
•
A high-level sponsor is needed to ensure 100% staff participation, SNA results
may be skewed if there aren’
t enough participants.
•
An SNA should never be done simply for the sake of doing it. A real business
problem must exist, and the SNA questions must be tailored to elicit
information specific to the problem.
•
Never forget the context –if an SNA was run to determine the knowledge
network, don’
t extrapolate the graphs to give cultural insights
•
Be prepared for contentious results –often the communities don’
t like what
they see in the graphs
•
Be careful who the results are communicated to, they can easily be used to
further someone’
s own agenda –this is especially true when doing an individual
SNA.
•
Similarly, care must be taken when communicating perspective bias results.
Some people may try to assign a value to a specific domain bias, e.g.
communities with a complex bias are better than communities with a simple
bias. This is completely untrue, as no Cynefin domain is more desirable than
any of the others and for an environment to operate efficiently it needs
resources that are comfortable in each of the domains.
•
Participants were frustrated by a lack of accountability to take forward the
suggested initiatives that resulted from the SNA. To address this issue, agree
upfront that business will own the results of the initiatives and specify who will
be responsible for the implementation of initiatives that fall into certain areas.
E.g. People Management or HR will take forward all Culture related issues, KM
will focus on Knowledge initiatives, etc
Acknowledgements
The author would like to acknowledge the following people for there contributions to this case study: Dave
Snowden, Zachda Prinsloo, Roy Lottering
100
References
Snowden, D. (2005) “From Atomism to Networks in Social Systems”, The Learning Organization,
Special Issue ‘Knowledge Sharing’, Vol. 12, No. 6, 2005, Peter A.C. Smith(Ed.)
Snowden, D. (2002a), “Complex Act of Knowing: Paradox and Descriptive Self Awareness”,
Journal of Knowledge Management, Vol. 6, No. 2, pp. 100 –111.
Snowden, D. (2002b), “Just in Time Knowledge Management”, KM Review, Vol. 5, No. 5, pp 1418; and Vol. 5, No. 6, pp 24027
Cross R., Parker A. (2004), The Hidden Power of Social Networks, Harvard Business School Press,
Boston, Mass.
The Cynefin Centre, http://www.cynefin.net
Author Biography
Sonja Blignaut is a Certified Cynefin Practitioner with application experience
in areas such as Social Networking, Corporate Culture Transformation, Knowledge
Management and Customer Experience.
She is an experienced facilitator and is skilled at guiding groups through complex
sense-making processes. Sonja has a degree in Meteorology and earlier
experience includes consulting around Enterprise Content Management Solutions
and Portals.
101
6
Beneath the Surface of
Conscious Patterns: Using
Narrative to Characterise
the Culture of Innovation
at a Leading R&D
Organisation
Gillian A. Maree
Dirk J. Roux
Mario A. Marais
Beneath the Surface of Conscious
Patterns: Using Narrative to
Characterise the Culture of
Innovation at a Leading R&D
Organisation
Gillian A. Maree, South African Council for Scientific and Industrial Research (CSIR), ([email protected])
Dirk J. Roux, South African Council for Scientific and Industrial Research (CSIR) ([email protected])
Mario A. Marais, South African Council for Scientific and Industrial Research (CSIR) ([email protected])
Abstract
Knowledge is a core asset for the CSIR and a direct determinant of
the organisation’
s relevance. A change process currently underway
aims to reemphasise the strong R&D base in line with the
organisation’
s mandate.
Strategic management of innovation is both a social and key business
process for converting knowledge into value. Projects typically entail
a high degree of new knowledge generation, as opposed to
repackaging existing knowledge. Human skills and knowledge remain
the primary limiting resource. Most work is multi-disciplinary and
project teams are assembled by drawing experts from a number of
organisations. The challenge is to effectively leverage knowledge
towards innovation excellence, and to achieve this, it is necessary to
understand the prevailing culture associated with innovation.
This chapter provides insights into the use of narrative techniques to
uncover behaviours, themes and archetypes beyond everyday
conscious patterns of recognition. The objectives were to explore the
usefulness of narrative techniques for making sense of complex social
processes, and to gain insights into cultural issues surrounding
innovation. Aspects investigated include understanding the current
underlying organisational values and how narrative techniques can
help understand these complexities. Outcomes from this exercise
support the CSIR change process.
Main findings include ten archetypes for innovation namely the:
Clumsy Puppy, Couch Referee, Courageous Captain, Bright-EyedBushy-Tailed Researcher, Inspiring Coach, Intellectual Maverick,
Intrepid Explorer, Meticulous Bureaucrat, Narrow-minded Nitpicker
and Willing Victim. Key issues that emerged are the client-scientist
relationship, the value of mentoring, and a culture that may value
beaurocratic behaviour at the expense of innovation.
Keywords: Narrative techniques, culture, innovation, archetypes, values and themes, change
process
105
Beneath the Surface of Conscious Patterns
Background
Narratives connect ways of knowing with ways of organising (from Patriotta, 2003, pp 353).
Industrial Context
The South African Council for Scientific and Industrial Research (CSIR) was founded in 1945, and
has developed into the largest research-based science and technology organisation in Africa. The
CSIR aims to provide science and technology services and/or solutions to a variety of stakeholders
in both the public and private sector through a portfolio that includes:
•
Research, development and implementation
•
Technology transfer and assessment
•
Scientific and technical education and training
•
Policy and strategic decision-support
•
Global science and technology links
•
Specialised technical and information consulting
•
Prototyping and pilot-scale manufacturing
•
Commercialisation of intellectual property, including venture establishment
The NRE Unit (Natural Resources and the Environment) is one of the core business units within the
CSIR. It focuses on environmental research to support sustainable development within the context
of national priorities and global challenges. The vision of the CSIR’s NRE unit is to conduct worldclass, directed interdisciplinary research and technological innovation, with partners and
stakeholders, in the fields of natural resources and the environment to contribute to the social,
economic and environmental improvement of South Africa and Africa. This is achieved by
undertaking research in areas of mining, forestry, pollution and waste, water resources, resourcebased sustainable development and ecosystems. Currently there are approximately 258 researchers
employed within the NRE unit.
Knowledge has always been a core asset of the organisation, and is a direct determinant of the
organisation’s relevance. Strategic management of innovation is seen as both a social and a key
business process for converting knowledge into value. The CSIR defines knowledge simply as the
capacity for informed action (Roux, 2004), which is closely related to definitions provided by KarlErik Sveiby, namely a capacity to act (Sveiby, 1997); and Ross Dawson, namely the capacity to act
effectively (Dawson, 2000). For the purpose of this chapter, innovation is defined as “an idea,
practice, or object that is perceived as new by an individual or other unit of adoption” (Rogers,
1995). CSIR strives to facilitate effective transfer of innovation, where the adopter has both the
absorptive capacity (understanding) as well as the emotional and financial commitments to allow
sustained use of the acquired knowledge.
Note that while this case study subsequently refers to the CSIR, the research was confined to the
NRE unit.
Problem
Current reality within CSIR NRE Unit is that typical projects entail a high degree of new
knowledge generation as opposed to repackaging or reusing of existing knowledge. Projects often
draw on a variety of specialist skills, necessitating project teams that draw experts from both within
and outside the organisation. Availability of human skills or knowledge is seen as the primary
106
constraint to achieving business objectives. Flowing from these realities, a key business imperative
is to attract and retain key talent, but also to effectively utilize internally (within the CSIR) and
externally (outside of the CSIR) available knowledge. An overall challenge facing the organisation
is to effectively leverage human skills or knowledge towards innovation excellence.
It is increasingly acknowledged that an organisation cannot manage knowledge per se (Snowden,
2003; Ungerer and Roux, 2005; Davenport and Prusak, 1997 and Denning, 2002) rather it is the
processes and tacit aspects that require management attention. Furthermore, Snowden (2003, pp 24)
believes that knowledge is both a thing and a flow that requires organisations to focus more on
context and narrative elements of knowledge (caters for tacit features) than content; ‘… human
knowledge is deeply contextual –it is triggered by circumstance’.
It should be noted that the larger the organisation, the more difficult it may be to clearly distinguish
patterns in a complex, social space (for example related to innovation). While leadership skills are
not necessarily control orientated, they are about trying to create order, and leaders need to sense
and respond to the emerging organisational patterns from observation and experience. Leadership
responses may include the encouragement of patterns perceived to be favourable and
discouragement of patterns perceived to be unfavourable.
Snowden (2003) sees knowledge management as being informed by complexity and not constrained
by it, recognising the differences between content, context and narrative management. The
implications for organisations of this are that certain types of knowledge (particularly context and
narrative) can not be elicited or managed by traditional techniques. The assumption behind using
narrative techniques is that innovation within the organisation has happened in complex space, that
is; patterns of knowing can only be recognised in retrospect.
To avoid ending up with “the same old answers”, narrative techniques and storytelling can be used
to uncover values, issues and characters or archetypes beyond the everyday conscious patterns of
recognition. For example, anecdotes represent a source of unguarded information, which may not
surface through question and answer techniques. Typically, patterns that emerge from the anecdotes
differ from the answers that people involved in innovation may give if asked to fill in a
questionnaire or during an interview. This rationale has led to a narrative enquiry into the patterns
that characterise innovation at CSIR, the outcome of which provides a baseline from which to
influence behaviour, lead change, share knowledge and guide decisions regarding CSIR’s future
direction.
•
To investigate or explore the usefulness of narrative techniques for complex
social processes
•
To uncover the archetypes present in innovation at an R&D organisation and
understand where they add value or hinder innovation
•
To uncover the dominant themes associated with innovation and creativity
within a research environment
•
To understand the cultural complexities surrounding innovation at CSIR NRE
Approach
James and Minnis (2004) believe that the more intricate organisational knowledge is, the less
effectively it can be codified. Stories provide examples of nuances that otherwise may be difficult to
write down and may even remove some ambiguity and make meaning more clear. Further benefits
of organisational storytelling include: allowing people to track or benchmark their own behaviour
107
and theories; generating new knowledge and ideas; obtaining a deeper understanding of underlying
cultural issues; providing a means of communicating possibly difficult issues in a non-threatening
manner; allowing a far greater depth and texture of knowledge to be conveyed; and using anecdotes
to provide repositories of accumulated wisdom (Mitchell, 2005, Patriotta, 2003, Snowden, 2001b).
Within an organisation work occurs in a complex social environment where anecdotes (for example
told around the photocopier or during a coffee break) allow people a means of distilling information
on behaviours and values that influences attitudes and shape the prevailing culture. Narrative
enquiry can be used as a mechanism to organise human cognition and make sense of multiple
signals present in organisational anecdotes (Snowden, 2001a).
This study made use of two techniques developed by the Cynefin Centre for Organisational Change,
namely:
•
Discovery technique: Anecdote circles were used to facilitate the informal sharing of
experiences by participants within a peer group. An anecdote is a real life account of an
individual's or a community's experience. It is usually unstructured, simple, conveys various
values, messages and rules, and is told from a particular perspective. Anecdotes are
associated with a historical event or experience and as such have a time imprint. Anecdotes
differ from stories in that they are based on facts and real experiences, where stories reflect
altered realities and carry a specific message (Snowden, 2003).
•
Sense-making technique: Two-stage emergence was used to extract emergent constructs
(Values, Themes, and Archetypes) from the outputs (anecdotes) of the anecdote circles. The
first stage uncovers ordered properties from the outputs (characters, issues and behaviours),
and the second stage reveals unordered properties culminating in the archetypes, themes and
values.
Case Analysis
“The real voyage of discovery consists not of seeing new lands but in seeing with new eyes”Marcel
Proust (quoted in Snowden, 2004, pp5)
During November 2004 two anecdote circles were held, the aim of which was to capture anecdotes
and experiences of participants relating to innovation. A total of 20 senior researchers were selected
to participate in this exercise because of their levels of experience and represented three different
geographic locations (Pretoria, Stellenbosch and Durban). Divided into two groups the researchers
sat in informal circles and told anecdotes of their experiences of innovation from various projects
and contexts. Using an informal setting (an open plan lounge area with bean bags and snacks)
created a comfortable environment conducive to sharing of stories. Each session was facilitated by a
member of the facilitation team familiar with the Cynefin techniques and was supported by a young
researcher.
It was emphasised that an anecdote should focus on an experience and not on opinion or personal
knowledge. During the anecdote circle, a number of pre-determined open and prompting questions
were asked by the facilitator upon which the participants self-selected the order in which the
anecdotes were shared. Anecdotes were recorded with a digital recorder and manual notes. The
collection of anecdotes was captured in a narrative database, following a protocol that allows future
extraction according to themes, issues, characters, emotions, behaviours, and etcetera. One of
several uses of this narrative data is to, through a further facilitated session, extract organisational
archetypes, emerging themes and emerging values, which in turn may be used to inform
organisational strategies and change management.
This discovery phase resulted in 144 anecdotes being captured which became the basis for a twostage emergence workshop held in March 2005. A new set of 14 participants worked together to
extract the character attributes, behaviours and issues from the anecdotes. A new set of participants
108
were selected to provide further diversity in the participants and further interpretation of the stories.
The participants followed a facilitated predetermined process allowing the attributes to become
archetypes, the behaviours - values and the issues - themes. A cartoonist was present during the
later stages who, with no prior briefing, sketched the attributes of the archetypes according the
discussions.
The final step in the process was to populate the narrative database that examined each of the
original anecdotes for the extent to which the archetypes, values and themes were present. The data
were captured by using participants from the two-stage emergence workshop working in pairs. The
data were then analysed in order to determine any emerging trends. The sense-making stage yielded
ten archetypes, twelve themes and fourteen values associated with either supporting or hindering
innovation at the CSIR.
“Narrative is the way human beings actually manage to understand and interact with other human
beings in the world … if you want to understand the world of human beings –and organizations are
fully of human beings –you’ve really got no choice. You have to understand narrative” [own
emphasis] (taken from Denning, 2002, pp18).
Results
The Archetypes
Ten archetypes that characterise innovation resulted from the narrative process namely the: Clumsy
Puppy, Couch Referee, Courageous Captain, Bright-Eyed-Bushy-Tailed Researcher, Inspiring
Coach, Intellectual Maverick, Intrepid Explorer, Meticulous Bureaucrat, Narrow-minded Nitpicker
and Willing Victim.
The cartoon drawings of the archetypes are provided below with explanatory text of their typical
attributes. The archetypes were developed during the two-stage emergence workshop, and the
method allowed the final archetypes to be mapped back to the characters that were contained in the
original anecdotes.
Each archetypes attributes was mapped back to the characters captured within the anecdotes. The
text box shows the original characters the archetype is composed of. As each archetype is composed
of a number of characters, only the dominant characters (those that constitute at least 60% of the
archetype) are shown. As an example, 29% of the students and/or learners in the anecdotes are
associated with the Clumsy Puppy Archetype.
109
The Clumsy Puppies are new recruits with little working
experience and always raring to try new things. They are
most in need of support, mentoring and are dependant
on more experienced staff for nurturing and skills
development.
Once they overcome their insecurities and given the space
and opportunity to learn, the Clumsy Puppy can be a
valuable contributor. However, in their enthusiasm a
Clumsy Puppy may not listen well and will need
constructive feedback from others to build their
confidence.
Clumsy Puppy
character
composition:
The Couch Referee is a ‘know-it-all’ who
believes things would have worked out if only
s/he was listened to in the first place. This is the
type of person who can be heard saying; “I told
you so!”
Couch Referees are more often than not
negatively perceived by colleagues who see
them as demanding, insensitive and arrogant,
with unrealistic expectations. The Couch
Referee on the other hand, feels that no one
listens to them or values their inputs enough.
Couch Referee
character
composition:
Clients - 45%
Scientists - 22%
The Courageous Captain is enthusiastic, inspirational and
forward looking. His or her energy, dedication and strength
of character afford many followers and is likely to be heard
saying; “Right everyone, follow me I know where we need to
go… ”
Self motivated, energetic and appreciated for being reliable,
the Courageous Captain is generally optimistic and
unperturbed by what lies ahead on the horizon. Within the
CSIR s/he is seen to have access to funding and great research
projects.
The Courageous Captain is the subject of many an
organizational story and these stories are likely to take on
mythical status.
Courageous
Captain character
composition:
110
Achievers - 39%
Followers - 23%
Students and Learners - 29%
Followers - 14%
Scientists - 14%
The Bright-eyed Bushy-tailed Researchers are young,
enthusiastic, goal orientated people. They can be
relied upon to achieve results, organise and get things
done. This does tend to pull them into managerial and
operational tasks (as opposed to research).
The Bright-eyed Bushy-tailed Researchers tend to be
highly productive over-achievers who continually
move their own goal posts. However, as a result of
their management tasks they are often office bound and
don’t get to see the bigger research picture.
Bright-eyed
Researcher
character
composition:
Managers - 46%
Leaders - 18%
Achievers - 18%
The Inspiring Coach is a true team player and the
organizational cheerleader! This is a people-person who is
supportive and committed to making a team successful.
An Inspiring Coach adds value by providing support
where it is needed, focusing on a common goal and having
the ability to accept criticism.
Sharing knowledge and communicating is important to
an Inspiring Coach. While acknowledged as an achiever,
the Inspiring Coach likes to have a hands-on approach to
getting things done. An Inspiring Coach tends to multitask, but always remains committed to the team.
Inspiring Coach
character
composition:
Team member - 19%
Achiever - 19%
Follower - 19%
111
The Intellectual Maverick is a visionary who has
insights into the bigger picture and a broad spectrum of
knowledge to draw upon. S/he is a lateral thinker who
enjoys the challenge of solving complex problems.
Intellectual Mavericks love a challenge and are not
afraid to take risks. New, original ideas often come
from Intellectual Mavericks and they will focus on an
idea until it is conceptualised to a manner that they feel
can be passed on to someone else to take further. An
Intellectual Maverick is seen to have strength of
character and key insights to the market.
Intellectual Maverick
character
composition:
Achievers - 15%
Analyst - 15%
Scientist - 15%
Leader - 15%
Intrepid Explorers love to explore new directions or ideas and
are creative, lateral thinkers. They have an adventurous
spirit and love a challenge. Where the Intellectual Maverick is
not afraid to take risks, the Intrepid Explorer is not afraid to
take the initiative and make the first move.
The Intrepid Explorer gets bored easily and constantly needs
new stimuli in his/her environment. Colleagues see the
Intrepid Explorer as arrogant, individualistic, opinionated
and complex. In top form an Intrepid Explorer could be heard
saying; “Guess what I just discovered … I bet you couldn’t!”
Intrepid Explorer
character
composition:
Innovator - 38%
Scientist - 38%
The Meticulous Bureaucrats are perfectionists who
like to focus on details and are painstaking
particular with their work. Unlike many of the other
innovation archetypes they are driven by targets and
deadlines and are analytically minded. Come what
may, a Meticulous Bureaucrat will deliver their
work on brief, on budget and on time.
A Meticulous Bureaucrat will clash with others,
get involved in organisational politics and is not
easily swayed from their values. S/he is easily
stressed by errors or change. Colleagues see the
Meticulous Bureaucrat as dictatorial and inclined
to sabotage.
Meticulous
Beaurocrat
character
composition:
112
Analyst - 19%
Family - 19%
The Narrow-minded Nitpicker is detail focused
with a very narrow perspective on what is
acceptable or not. They prefer to operate alone
or in a silo, do not communicate well and
generally have a low morale.
A Narrow-minded Nitpicker will work hard on
micro-task details, which is seen as selfish or
non-value adding work by bigger picture
thinkers. Colleagues perceive the Nitpicker as
lacking commitment to the organization or
team and selfish.
The Narrow-Minded Nitpicker is the archetype
that displays the most negative qualities.
Narrow-minded
Nitpicker
character
composition:
Victim - 33%
Follower - 17%
Analyst - 17%
Willing Victim is the gloomy person who feels
sorry for him- or herself. S/he believes that his
or her contributions are never adequately
acknowledged or properly appreciated. A Willing
Victim is pessimistic and quick to pass blame.
Colleagues see the Willing Victim as egotistical,
overly critical and generally naïve to larger
issues. The Willing Victim will not take the
initiative or volunteer an idea as they believe
that their inputs will not be appreciated anyway.
Willing Victim
character
composition:
Victim - 46%
Innovator - 16%
The Intrepid Explorer was by far the most common and strongest featured archetype in innovation
anecdotes, followed by the Intellectual Maverick. While the impact of the Bright-eyed Bushy-tailed
Researcher is not as great, the archetype is featured consistently throughout the anecdotes. The least
frequent occurring archetypes are the Narrow-minded Nitpicker and the Willing Victim; however
the Willing Victim is the only archetype to score high on project management, an important part of
leveraging innovation.
Many of the archetypes share similar weaknesses and strengths, for example, effective
transformation is consistently weak. It is interesting to note that while transformation is a core
organisational theme results this exercise indicate that innovation is probably not the most effective
‘home’to achieve transformational goals. However, mentoring has been shown as a key aspect of
successful innovation. The highest ranking archetypes (Intrepid Explorer and Intellectual Maverick)
are role models for innovation and the type of characters around which coffee table talk and
mythical stories form. For this reason showing support for the Intrepid Explorer and Intellectual
Maverick types could assist in initiating and gaining support for implementing transformation.
113
Themes and Values
Anecdotes extract shared views and meanings for a community (Patriotta, 2003). From an
organisational culture perspective these shared worldviews are important to understand the common
features of an organisation’s culture. For learning to be most effective a certain degree of overlap in
understanding is needed (James and Minnis, 2004; Roux and Claassen., 2005,) and sensemaking of
the anecdotes helps understand where these overlaps may be.
The tables below show the outcomes from the themes and values emergence process.
Themes
Scientific integrity
Effective communication
Leadership
4
Stimulating creativity and innovation
Top es
m
Manage staff turnover
the
Effective transformation
Appropriate reward and recognition
Branding and corporate image
Effective project management
Effective Systems and structures
Value your staff
Defining and implementing a common objective
Values
Systems that encourage the right behaviour
Endorsing the right values
A knowledge enabling environment
5
A positive learning culture
Top es
u
Val
Open and effective communication
Accepting or tolerating under performance
Lack of a project management imperative
Management focus on power and bottom line to the expense of innovation
Personal ownership of career development
Good performance management
Organisational commitment to creative space
Organisational commitment to understanding and responding to motivational needs
Roles and responsibilities are clearly aligned and communicated
Promoting positive team dynamics
The most notable feature from the themes and behaviours is that they are consistently very positive.
It became clear during the sensemaking phase that the results, particularly for the values, show
114
some manipulation to reflect what should happen for successful innovation and not necessarily what
did happen.
It is interesting to note that during the emergence workshop people were the least comfortable
grouping the values for likeness and continually tried to rather group according to what they
thought the outputs should reflect. The facilitation team noted that the level of discomfort among
the participants to group values for likeness resulted in a tendency to try and pre-empt the
outcomes. For this reason the archetypes will remain the focus of this chapter, and not the values
and themes. However, in this context it is worth noting the results and the observations.
A dominant theme in the anecdotes is an obvious one: ensuring the ability to stimulate creativity
and innovation. Other dominant themes include effective communication, leadership and scientific
integrity. Interestingly, the ability to manage staff turnover, effective transformation and appropriate
reward and recognition do not feature strongly, yet these top the list of organisational level goals.
The results showed that the strongest values for innovation are open and effective communication,
having a positive learning culture and creating a knowledge-enabling environment. The least
frequent values are the negative values of lacking a project management imperative and accepting
or tolerating underperformance.
Effective communication is shown to be very important for innovation, ranking high as both a value
and a theme. Put differently; the results indicate that if you want successful innovation you need a
culture that values and encourages the sharing ideas through effective communication mechanisms.
Communication is important for sharing knowledge and leveraging ideas into something ‘do-able’.
There are very few anecdotes that tell of an experience where innovation happened because
someone did not want to share an idea for fear of lack of recognition or losing control. Sharing and
communicating across teams was a clear theme throughout the anecdotes captured.
The text boxes below show excerpts from the anecdotes.
Quote on the value of experiential learning…
“When you get a heart bypass you don’t pay for the hours that the doctor works on
you; you pay for the hours that he spent learning and working on other people. “
Anecdote on reward and recognition…
“After we completed a huge EIA [Environmental Impact Assessment], one of the vice
presidents indicated that they wanted to reward the team appropriately. As project
manager I got up and said: ‘I think you can't do better than just give us the space to
go on a game drive. That will reward our families as well for having put up with
us.’ We were then allowed to spend a week at the Victoria Falls. Our families
joined us for two days and stayed for free. The core team was there for a couple of
days afterwards. It was the best reward they could have given us.”
Key Findings
In general, outputs from the sensemaking process reveal a number of interesting issues which are
characterising innovation at the CSIR NRE unit. This section provides a brief overview of the key
findings and concerns.
115
It is important to note that the values, themes and archetypes identified arise directly from the
anecdotes and thus the actual experiences of staff. While there may appear to be certain gaps in the
list, this is because these issues did not arise strongly from the anecdotes and is an indicator of those
values that pattern entrainment tells us should be a part of innovation, but may be rarely practiced in
the organisation.
Snowden (2003) believes that an individual builds his or her own patterns through learning
processes and experience which enables him or her to understand complex situations and thus make
decisions when surrounded by uncertainty. The anecdotes collected for this exercise point to
uncertainty and risk being present in an innovation environment, where previously existing patterns
are broken in order to form new ones.
Pattern entrainment in communities can result in ‘group-think’ and the unwillingness for a
particular community to recognize or accept patterns outside of their own paradigm (Kurtz and
Snowden, 2003). Retrospect and internalization of experience, is a good collector of patterns.
Individuals that participate in a project contribute their own personal experiences and patterns to
processes of innovation. The following discusses how the archetypes for innovation each contribute
to an innovation environment.
A Space for Innovation
The CSIR case study revealed a number of archetypes that raise concerns for innovation practices
namely: the Coach Referee, the Willing Victim, the Bright-eyed Bushy-tailed Researcher and the
Intellectual Maverick.
A key issue emanating from the outputs of this exercise is the client-scientist relationship. To
demonstrate, the Couch Referee archetype comprises 45% of the client character. From a Couch
Referee you are most likely to hear ‘if you had just listened to me in the first place we would not be
in this situation.’Understanding exactly what the client wants and can use effectively is critical to
the success of any project and requires good listening and interpretation skills; which are lacking in
some of the archetypes (notably the Couch Referee, Intellectual Maverick and the Intrepid
Explorer).
Extract from an anecdote:
“Having finished the project the client rejected all the solutions in our big report
because he said it would be too complicated for the people to understand.”
The results point to issues where clients may have trouble relating to scientists and getting them to
listen and understand to their needs. This may mean that new ideas or solutions may either not be
tailored to the client’s actual needs, or they can be overlooked. Furthermore, if the automatic
reaction to a client is a negative one, and the client is seen as interfering, the likelihood of
successful innovation is limited. Active engagement and conversation between the client and
scientist is necessary for successful innovation.
116
An anecdote on success…
“We were having problems relating to our client. During a discussion about why the
project was not running smoothly we were surprised at one of the key things that the
client insisted on, namely that we work from their offices. He asked how we can
understand our clients when we don’t experience their working environment firsthand –and the suggestion was to work for a year in their environment to learn how
they do things and get a better sense of how we can deliver value into that
environment.”
At first glance the Bright-eyed Bushy-tailed Researcher represents the young scientists eager to
contribute ideas and work hard at tasks; someone you most likely would want in your team.
However, closer examination of this archetype revealed that two thirds of the characters that make
up this researcher archetype are the manager and leader. The concern this raises is that a situation
may exist where staff that are young, inexperienced and in need of mentoring and growth from
senior scientists, are taking on managerial tasks. This reflects a situation where enthusiastic new
researchers in the organisation are pulled into project management and administrative tasks and not
primarily into R&D activities.
These findings are in line with developments within the CSIR where concern for the development
and growth of young researchers at the CSIR has led to the establishment of a Young Researchers
Forum and funding is set aside specifically for skills development of young researchers. A recent
survey of young researchers to determine what the necessary skills were that they needed to develop
revealed a strong demand for practical experience on research projects to grow their basic science
skills and opportunities for networking and linking to other leading scientists.
Anecdote…
“It is a disheartening thing; you can so easily get caught up in politics. You then try
to keep to the rules so much that you forget what you are [employed] there for.”
While a Clumsy Puppy is likely to volunteer for a task they need to be paired with a strong mentor
to guide them. Due to their presence and standing within the organisation the Intellectual Maverick
and Intrepid Explorer archetypes are likely to include a number of the mentors. However, if a
particularly sensitive mentee is paired with the less sensitive archetypes (like the Intellectual
Maverick or Intrepid Explorer below) this may develop into a destructive relationship where the
Clumsy Puppy may become despondent and lose their enthusiasm. The short anecdote below
demonstrates how this can occur.
117
Anecdote told by a Clumsy Puppy about his/her mentor the Intellectual
Maverick…
After a lot of hard work and I had finished writing up the report I asked him
[mentor]: “What do you think?”
He stared into the distance and said; “I am really worried about this project.”
And then I panicked and thought there must be some fundamental design fault.
“Where did I go wrong? Did I misconceptualise [sic] the whole thing? Was there
something wrong with my science procedures? “
All of this went through my mind, as he remained silent.
The Willing Victim archetype shows that within the CSIR there might be a lack of recognition and
respect of those people with new ideas (16% of the Willing Victim comprises the Innovator
character). There is a danger that the organisation may be rewarding bureaucratic behaviour and not
the true lateral thinkers.
The Intellectual Maverick, while showing a strong tendency towards innovative behaviour, does,
however seems to be very scheming and is not perceived favourably. The Maverick likes to set him
or herself apart from everyone else and is not worried about being unpopular or disruptive to a
process that they might not agree with. This may lead to suppressing or holding back good ideas by
less confident and outspoken staff.
A further trend seen in the archetypes is that the scientists are seen to be working towards achieving
the greater good in contrast to the management types who are seen as bottom-line driven and
hindering the innovation process.
The very strong positive bias in all the outputs paints a very rosy and probably skewed picture, of
innovation. This positive approach may be masking a reality where people might be conforming too
easily and even suppressing new ideas to maintain the current status quo. Many of the answers or
results from the values and themes show standard answers and there may be fear to tell the truth or
unconscious adherence to the ideal picture.
Myth and Fable
Anecdotes play a large role in building institutional memory and there are certain stories that most
people know or have told themselves. These are typically the stories that new people joining the
organisation are exposed to early on in their employment. Some of these stories have been around
in different forms for many years and have developed mythical status. These stories carry messages
about the values and behaviours that the organisation strives for.
Mythical stories are important to identify as they point to important cultural systems that are subtly
embedded in the organisation (Snowden, 2001b; Denning, 2002; and Kurtz and Snowden, 2003).
As mythical stories grow and develop they become rich in metaphor and emotion and though they
will be told by many story-tellers, certain phrases will remain and become rooted in the lingo of an
organisation.
While all the anecdotes captured were kept anonymous, many people involved in the process could
easily identify the people from some of the stories as they had often heard the stories many times
before. An emerging thread from these stories is that they are emotive, rich in metaphor, often
scornful and self-reflecting. What binds the story together is the message which is mostly about
bucking the system, learning through doing and taking chances. It is worth noting that in many
cases these specific stories are about unease or dissatisfaction in the way things are done, rather
than supporting organisational processes.
118
A key lesson learnt is to keep to the facts of the anecdotes as much as possible as people will easily
pick up subtle changes from the version they are familiar with. An example of a mythical story in
the CSIR is provided in the box below. This story is told to demonstrate a number of values in the
CSIR. First, the story highlights the value of experiential learning of a scientist; the problem could
not have been solved without having had so many years of experience. Second, that innovation and
value can not be measured by filling in timesheets to account for time. This is a story told often
within the organisation and supports the message from the archetypes where emphasis and pressure
is placed on scientists to make money; a message that can be in conflict with the time and space
needed for good quality research.
“A few years ago one a senior scientist who has been working at the CSIR for many
years doing research into water resources was urgently called out to a mine to have
a look at a water quality problem they were experiencing. This scientist was able to
understand the problem and suggest a solution within a few minutes and without too
much trouble. The client was very happy as the solution could be easily and cheaply
implemented. When the time came for payment the scientist suggested that instead of
paying the usual hourly charge out rate, the client should pay a small fraction of the
savings they had achieved by implementing the solution. The client was only to
happy to agree to this arrangement, resulting in a good day for all.”
Working Relationships
An aspect that came out of the anecdotes told during the anecdote circles were the values around
‘how we treat each other’; particularly relating to relationships between co-workers, mentors and
mentees, as well as between managers and subordinates.
Many of the anecdotes were not related to innovation per se, but rather to creating the space for a
new recruit (e.g. Clumsy Puppy) to pair up with a more experienced scientist (typically the
Courageous Captain or Intrepid Explorer). These anecdotes less frequently related to the Intellectual
Maverick who prefers to work on ideas independently and who may intimidate or discourage
juniors. Many of the anecdotes where the Intellectual Maverick was present relate to conflicts with
management or administrative systems and procedures. The Intellectual Maverick is the archetype
most likely to clash or rebel against organisational systems.
Anecdote about the Intellectual Maverick:
“A key part of my research plan indicated the need for mathematical modelling and
physical limnology skills that were not available in our group. No one else saw the
scientific value of doing the work on this system, so I decided that I would do the
work myself. I worked out a suitable sampling and measurement program and
borrowed the equipment I needed from local universities and consulting firms. I
ended up doing the work in my own time over weekends as the research managers
saw no technical value in the work. By the end of the study I had published 5 papers
in international journals, made 3 international conference presentations, and in a
four year period I had received requests from over 100 overseas guests visiting our
institute to be taken on a guided tour of the study site”.
119
Anecdote about the Clumsy Puppy:
“I remember right at the beginning of my career going on a project field trip with a
Marine Biologist and a Coastal Engineer (both PhDs) to a bridge at the mouth of the
Kromme Estuary. Then, while standing on the bridge, this engineer and biologist
started arguing for about an hour and a half about how the mouth dynamics of the
whole system worked and should they dredge the estuary or not. Each had his own
perspective and could not get the other to agree with their understanding of the
problem. They were making such a scene yelling at each other that a bunch of people
stopped to watch what was happening. I just stood staring at them; my eyes were big
because they were at the point of being quite abusive of each other arguing their
points. Eventually they resolved their problems, shook hands and went off for a beer!
Looking back, at that point in my career it was a wonderful experience for me to
learn about how these estuarine systems work from the two very different
perspectives of an engineer and biologist.”
A further point that the anecdote about the scientist working on his own demonstrates, concerns the
conflicts that arise between scientists driven by ideas and the managers responsible for managing
the organisation. This shows the culture of “us versus them”that has developed and which can, and
has, lead to conflict.
The archetypes highlight the value and necessity of mentoring and peer support in innovation
processes. It is interesting to note that this is not followed though in the values or themes. Stories
about taking personal ownership of one’s career are most closely associated with the young
researcher taking initiative on his/her own and not pushed or driven by a mentor or leader. Stories
of leadership speak mostly to someone who was willing to interface with management structures or
clients.
Business Impacts
Impact can be achieved through the use of effective communication techniques to disseminate the
results. The cartoons of the archetypes have impact as they provide a non-threatening mechanism
for staff to see a small part of themselves in each of the archetypes. Archetypes are important KM
tools as they allow complex adaptive systems (which organisations are) to be understood in a
manner that does not involve expert interpretation (Snowden, 2001a). Anecdotes and the archetypes
create role models and build a sense of identity. Furthermore the archetypes are useful as:
•
a representation of culture,
•
a means to understand customers, and
•
a means to bring two differing cultures together (situations of ‘us versus them’). (after
Snowden, 2001b)
This is supported by Patriotta (2003) who sees stories as an important component in an organisation
discourse because they:
•
show how knowledge can be mobilised to deal with situations (i.e. ‘how we deal with things
here’),
•
are connected to actual experiences and reveal common sense wisdom, and
•
are able to highlight shared worldviews.
120
An important area of impact for the results is the change process currently underway in the CSIR.
The aim of the process is to re-emphasise the R&D base in line with the organisation’s national
mandate. Outcomes from this narrative case study largely support the change process at the CSIR,
shifting focus toward strong innovation archetypes like the Intrepid Explorer. Aspects include
processes to understand the current underlying organisation values and culture and narrative can
add value to these complexities.
Emerging Patterns
Diversity in its multitude of facets has long been acknowledged as a contributor to organisational
knowledge and innovation (Hill, 2004). Correctly managed diverse work groups can add value
where homogenous groups never could. Having a diverse workforce means having to communicate
key messages and values through a variety of media. The value of archetypes (as opposed to
stereotypes) is that they resonate with a more diverse audience.
Capturing anecdotes from people will allow organisational patterns to begin to emerge (Mitchell,
2005). These patterns can add value by allowing a deeper understanding of the cultural patterns of
an organisation, how problems tend to be resolved, and can help describe the working atmosphere
between colleagues. The archetype cartoons are intended to allow people to identify a small part of
themselves in each cartoon, whether it is in retrospect or current experiences. For example, a senior
researcher may have seen him or herself in the Clumsy Puppy when starting their career and now be
able to identify more with another archetype like the Intellectual Maverick or Courageous Captain.
While everyone should be able to identify with the archetypes most people will identify more
strongly with one or two archetypes (Snowden, 2001a). The organisation is composed of a
complement of all the archetypes that interact and work together on a daily basis. This section has
tried to highlight what the possible impacts or benefits are for innovation of having these archetypes
in the organisation.
Some of the key patterns emerging from this exercise include:
•
The client-scientist relationship. Problems with client relationships and the communication
gap that exists between the scientist and the customer. Some of the scientist-type characters
are shown as always knowing better and clients as being angry and frustrated.
•
The very positive spin on the values and themes resulting from the emergence workshops
point to a concern that staff at the CSIR are either unwilling or afraid to share their real
feelings, or more likely, that the organisation has become smug. This is a perception that is
often shared by non-scientists; cartoons depicting scientists often show them as aloof and
arrogant.
•
There is a danger with the underlying bureaucracy that we may be rewarding bureaucratic
behaviour and not the real lateral thinkers, at the expense of innovation. This behaviour
might also be pushing the Bright-eyed-bushy-tailed Researcher into managerial tasks in an
attempt to allow the more experienced scientists to free up some time for research work and
to comply with the administrative requirements. Another interpretation is that the Brighteyed-bushy-tailed Researcher may be pulled into managerial tasks as the organisation may
value and reward management behaviour more than innovation.
From an organisational cultural perspective this points to a culture that values bureaucratic
behaviour and may be suspicious of staff that pushes the boundary of what is considered to
be acceptable behaviour.
There is an encouraging trend of an organisation that values mentoring. This highlights the
value that is placed on effective mentoring to achieve organisational goals.
•
121
Conclusions
Within a South African context, storytelling is an important element of cultural heritage. Oral
histories and the use of stories to communicate messages and values is an important component of
South African society, and audiences are often very receptive and appreciative of a story.
The use of narrative techniques allowed a number of the staff to learn about the value of narrative
sharing in an organisational context. As this was a new technique a number of lessons were learnt.
Key issues that arose include managing representivity and diversity of participants, techniques or
ways of extracting meaning from the data and communicating the outcomes. While knowledge
management is an essential part of the CSIR’s business, getting buy-in to the softer sides of
knowledge sharing can be difficult.
KM in CSIR embraces a human-centred approach where understanding culture and learning
practices is important to grow the knowledge base. An additional KM activity is to entrench KM
principles into business practices by developing strategies and plans for knowledge sharing and
learning. Narrative has provided an additional tool in the KM toolbox.
Through the use of a narrative technique to understand aspects of innovation experiences at the
CSIR a number of issues emerged that require further exploration:
•
•
•
Client relationships and some exploration into the perceptions clients may have of the CSIR.
This could be facilitated by running further anecdote circles with key clients who are willing
to participate. The aim would be to gain different perspectives on how clients see innovation
at the CSIR and whether new ideas are allowed to develop within the current systems.
A further issue that surfaced was that the archetypes reflect perceptions of clients and of the
organisation that differ greatly. There is also interesting differences between how the
archetypes see each other. Value can be added by using techniques that allow people
identifying with aspects of these diverse archetypes to examine how archetypes illuminate
the sources of conflict and thus find creative ways of dealing with conflict.
There are areas of highly successful and innovative environments in the CSIR that
continually push boundaries, but there are also pockets of failure. By examining and
understanding what makes the successful areas successful, learning can be transferred or
implemented in the less successful areas.
A key lesson from the exercise is that innovation can only happen when innovative behaviour is
well-regarded, encouraged and enshrined in the organisational culture. People who contribute to
successful innovations should be recognised for the particular value they add and the role that they
play. As much as an individual may add value, they can equally destroy it if their particular abilities
are incorrectly used.
122
•
Typically, literature on narrative techniques has shown that stereotypes are
believed to exist in archetypes; however this exercise has shown that within an
organisation strong biases may also exist within themes and values. For this
case study, it was particularly evident when patterning values. The stereotype
represents prejudices that exist, or the answers that people think they should
be giving, versus those that they actually experience. Understanding the
behaviours that make up organisational values is far less intuitive than
understanding the characters present.
•
A great deal of value from narrative exercises lies in the sensemaking process.
Lessons learnt during the sensemaking include:
o
Capturing all the data throughout the various stages. A great deal of data
is generated (particularly through the two-stage emergence workshop)
and it is valuable to account for all inputs that are made and to be able
to map the results back to the original anecdotes. This allows one to link
the outputs back to the original anecdotes that contributed to a particular
outcome.
o
It is valuable to involve people outside of the organisation and project
team to lend a fresh perspective and ask difficult questions which
otherwise may not have been considered. An outsider will see different
patterns emerging from the anecdotes.
o
The more diverse the range of participants, the more diverse the inputs
and interpretations of the stories are. Here diversity relates not only to
cultural and gender diversity, but specifically to diversity in terms of
career development (e.g. an intern and principle scientist) and fields of
specialisation (e.g. an engineer or ecologist).
•
For people to truly participate effectively they need to see the value in the
exercise and benefit personally from it. Sharing anecdotes among participants
adds value by allowing them to learn from one another’
s perspectives.
•
There are many stories (with subtle differences depending on the story teller)
that most, if not all staff at CSIR NRE unit are familiar with even if the event
occurred long before their joining the organisation. Many of these lessons on
‘
how we do things around here’will never be found in any organisational or
orientation document, but are rather communicated to new staff through story.
These stories point to networks of ‘
invisible processes’whose patterns and
purposes may be elicited through stories. When these anecdotes are told, they
are often told with high levels of vivid detail.
•
Most stories about successful innovation have shown similar ‘
take-homemessages’
: that great things can happen when you take a risk, learn from a
mentor and challenge the existing systems if they hamper progress.
•
Innovation does happen when the boundaries of current paradigms are pushed.
Many of the stories told relate to encountering problems or issues with current
processes and new ideas were successfully implemented only when a new
direction was taken. This however requires strong, determined archetypes who
are prepared to take calculated risks.
123
Acknowledgements
The authors of this chapter gratefully acknowledge the guidance and assistance received by:
Sonja Blignaut (IBM South Africa) and Dave Snowden (Cynefin Centre) for their valued insights
and assistance with using the narrative tools. Estelle Botma and Marita Kritzinger of the CSIR for
their facilitation of the workshops and hard work to capture the narrative data. Marius Claassen,
Martie van Deventer, Pat Manders and Brian Moloi for reviewing the chapter. Rob Hooper for the
archetype drawings. The many CSIR colleagues for sharing their stories and giving their time to
participate in this process.
References
Davenport, T. H. and Prusak, L. (1997) Working Knowledge: How Organizations Manage What
They Know. Harvard Business School Press, Boston.
Dawson, R. (2000) Developing Knowledge-Based Client Relationships – The Future of
Professional Services. Butterworth Heinemann, Boston.
Denning, S. (2002) The Narrative Angle: The Seven Most Valuable Forms of Organizational
Storytelling. A Managers Handbook on Narrative, Chapter 1: The Importance of
Organizational
Storytelling,
(date
accessed:
23
August
2005)
www.line56.com/images/articles/ attachments/km0302.denning_ch1.pdf
Hill, L. (2004) Diversity as a Function of Knowledge Transfer, Diffusion and Absorption, CSIR
Internal Report No. ENV-P-I 2004-038, Pretoria: Council for Scientific and Industrial
Research
James, C.H. and Minnis, W.C. (2004) Organisational storytelling: It makes sense, Business
Horizons, Vol 47/4 July-August, pp. 23-32
Kurtz, C.F. and Snowden, D.J. (2003) The new dynamics of strategy: Sense-making in a complex
and complicated world, IBM Systems Journal, Vol. 42, 462 No. 3
Mitchell, H.J. (2005) Knowledge Sharing – the value of story telling, International Journal of
Organisational Behaviour, Volume 9(5), pp 632-641, ISSN 1440-5377
Patriotta, G. (2003) Sensemaking on the Shop Floor: Narratives of Knowledge in Organisations,
Journal of Management Studies, Vol 40:2 March, pp. 349-375, Blackwell Publishing
Rogers, E. M. (1995) Diffusion of Innovations. The Free Press, New York.
Roux, D.J. (2004) From 2nd generation knowledge theory to practically growing human capital in a
S&T organization, Paper presented at the 4th I&KM Conference: Knowledge Management in
a World of Diversity. IBM South Africa, Sandton, 25-26 May 2004 (Presented by Centre for
Information and Knowledge Management, Rand Afrikaans University)
Roux, D.J. and Claassen, M. (2005) Human-Centred Knowledge Management: Performance
Indicators to Leverage Tacit Knowledge, CSIR Internal Report No. ENV-P-I 2005-003,
Pretoria: Council for Scientific and Industrial Research
Snowden, D.J. (2001a) Archetypes as an instrument of narrative patterning, ARK: Knowledge
Management, Story Special Edition November 2001.
Snowden, D.J. (2001b) Narrative Patterns: the perils and possibilities of using story in
organisations, ARK: Knowledge Management, Story Special Edition November 2001.
Snowden, D.J. (2003) Complex Acts of Knowing: Paradox and Descriptive Self-Awareness,
Bulletin of the American Society for Information Science and Technology, April/May 2003,
pp 35 –28
124
Snowden, D.J. (2004) Facilitating innovation within the organisation, in Finance and Management,
September, Issue 114, available at www.icaew.co.uk/fmfac
Sveiby, K.E. (1997) The New Organizational Wealth: Managing and Measuring Knowledge-Based
Assets. Berrett-Koehler, San Francisco.
Ungerer, M. and Roux, D. (2005) Paradox as Thinking Frame for Knowledge Development.
Management Today. September 2005: 26-28.
Gillian Maree has worked as an environmental researcher at the CSIR for the last four
years. Her current research work and interests focus in the water resources field and
include spatial planning, systematic conservation planning, biodiversity assessment
and interpreting the work physical environment from a knowledge management
perspective. Gillian has a BSc in Urban and Regional Planning from the University of
the Witwatersrand, Johannesburg and MSc in Biodiversity Management from Saxion
University in The Netherlands.
Dirk Roux has worked in CSIR’
s contract R&D environment for the past ten years.
His scientific experience covers the fields of limnology, aquatic toxicology,
environmental monitoring and reporting, biodiversity assessment, systematic
conservation planning, adaptive resource management, and institutional learning and
knowledge processes. Dirk has considerable experience in planning and leading
technical as well as policy-level projects related to the monitoring, protection and
management of freshwater ecosystems. He holds MSc and PhD degrees in Applied
Limnology from Rand Afrikaans University, Johannesburg.
Mario Marais is currently a researcher in the Centre for Logistics in the Built
Environment Business Unit of the CSIR. His current interests and research include
Knowledge Management strategy development, the impact of Knowledge
Management on economic development and poverty reduction, the role of logistics in
development, problem structuring via soft operations research techniques and the
study of decision-making.
He was trained as a physical chemist (MSc in Crystallography) and has grown
semiconductor crystals for a living and has also developed gas sensors. Before his
appointment in the Centre for Logistics, he was Product Manager of a Radio
Frequency Identification tag, Technology Manager and CIO of a CSIR business unit
and a part-time student (Masters Degree in Theology).
125
Industry Analysis and
Business Modelling
7
Applying Business
Modelling to
Lale Citipitioglu Eidal
Flemming Hjorth
Jørgen Kadal
Applying Business Modelling to
Lale Citipitioglu Eidal, DNV Maritime ([email protected])
Flemming Hjorth, DNV Maritime ([email protected])
Jørgen Kadal, DNV Maritime ([email protected])
Abstract
Today’
s knowledge-based organisations are complex entities. They
are comprised of a wide range of processes, technologies and
competencies. Understanding this complex environment poses a
challenge when trying to establish effective Knowledge Management
(KM) initiatives. This case describes how one organisation, DNV
Maritime, discovered and applied a unique approach to Business
Modelling (BM) for developing and communicating effective KM
solutions. The approach was introduced by the KM team and it
combines Enterprise Modelling/Architecture (from the ICT perspective)
and Intellectual Capital theory in a way which enables a common
understanding of the business environment in which an organisation
operates, providing the KM team and Management the information
they need to make better decisions on how to leverage critical
knowledge in the most efficient way possible. The approach has also
proved an effective way of communicating KM and its value to the rest
of the organisation. In addition to improving decision-making for KM,
the KM team’
s successful efforts in establishing and applying a holistic
framework and meta-model has also resulted in a positive change in
how the organisation approaches organisational and process
development.
Keywords: Knowledge Management, Business Modelling, Enterprise Modelling, Maritime
Industry, Classification Society
Background
Industrial Context
Det Norske Veritas (DNV) was established in 1864 and currently employs over 6000 people in a
network of 300 offices in over 100 countries. As DNV’s oldest business unit, DNV Maritime has
been operating true to its main objective of ’safeguarding life, property and the environment at sea’
by offering classification and certification services for vessels and other mobile offshore units. As
one of the world’s leading classification societies, DNV Maritime has also been harvesting the
experiences of its ship surveyors to develop its knowledge base as a basis for developing its own
standards for ship classification. In other words, Knowledge Management has had a central role in
the company for nearly 142 years.
Historically, classification societies have developed their own standards towards which they have
classified vessels. These standards are developed as requirements to both the new building phase of
vessels as well as for vessels in operation, as a measure to ensure high standards in the construction
131
and daily maintenance of the vessels. It has therefore been critical to DNV’s success that the
classification standards (or rules, as they are formally called) are regularly updated to reflect the
highest levels of safety based on the technical knowledge in the company. At DNV, acquiring,
improving and applying knowledge related to safety at sea has been a top priority. This knowledge
can stem either from a variety of resources, such as academic work in related disciplines or the
hands-on experiences of our ship surveyors in the field. DNV’s surveyors work in an environment
where systematic competence building and knowledge transfer go hand in hand with everyday
work. For example, new surveyors join the on-the-job tutoring scheme to ensure they are exposed to
the tacit knowledge of more experienced surveyors. This is a routine which is valid for all surveyors
in DNV. Another example is the experience-feedback system, in which surveyors report incidents
that have occurred in the field back to headquarters. These reports are processed into lessons learnt,
and in some instances incorporated into new rules and/or internal guidelines for service execution.
These two key knowledge-management initiatives existed long before KM became known as a
discipline in its own right.
Problem
In more recent years, DNV has grown both geographically and in terms of the sheer volume of
services and products it offers to the Maritime industry. This has created new challenge of how to
leverage knowledge and ensure learning throughout the organisation.
In the early 1990s, developments in information and communication technology (ICT) were
harnessed to create a production system that would make service delivery much quicker. The same
technology that improved the efficiency of service delivery was also seen as a means to overcome
the challenges that greater geographical distances pose for knowledge sharing. Large investments in
ICT were made, with the approach at the time tending to focus more on capturing explicit
knowledge in sophisticated systems and less on supporting tacit knowledge needs. The
implementation of a corporate ICT platform and the introduction of a new production system
revealed some unexpected effects of this new technology on knowledge sharing. In the ‘new way of
working’ using ICT, service-related documentation could be transferred at a much faster rate
between the customer, surveyors and our employees at our offices. This led to increased pressure
on employees to work at an accelerated pace, leaving them with less time to share knowledge with
one another. A new corporate intranet was also launched in the late 1990s, but quickly grew out of
control with the countless sites and pages that lacked a clear content management strategy.
Alongside the development in ICT technology, people networks continued to develop organically as
they had in the past. But, these networks had their limitations. They were available only to a limited
number of employees who knew one another as a result of working together for years, leaving the
growing population of less experienced and geographically dispersed employees without the
contacts they needed to learn and share knowledge across the organisation. Consensus was
emerging among managers that DNV would benefit from a more systematic approach to managing
knowledge in these new conditions.
Market demand for new services, greater competition and the continuous pressure on lowering costs
also led management to focus more heavily on reducing service or product development cycle time
and improving quality also increasingly become a key focus area in DNV Maritime. Various
approaches and techniques (including KM) have been applied to address this in more recent years.
DNV has observed that many companies have successfully applied business process modelling as a
means of identifying measures to achieve more effective and efficient operations. However, a
shortcoming of traditional process modelling activities has been that they fail to reveal the
interrelations between the processes and other aspects of the business (such as information flow,
competences, etc). These also fail to reveal activities which can be done in parallel. Knowledge
Management is interested in core work processes because they provide the interface through which
132
knowledge and information can be made seamlessly available to those who require it without
having to spend addition effort or time. More often than not, KM involves ‘meta-processes’which
interact with core work processes without necessarily being recognized as such by employees. All
industries and businesses can benefit from KM, in that successful operations and development
activities require use of information and knowledge, and KM is focused on just that.
Today’s knowledge-based organisations are often complex entities. They are comprised of a wide
range of processes, technologies and competencies. An ultimate success factor for such
organisations is leveraging the knowledge that lies within people’s heads in the most effective way
possible. A solid understanding of the internal and external business environment is essential for
successfully developing the mechanisms through which knowledge is created, synthesized and
applied. Typical questions for KM initiatives are:
•
What knowledge is truly ‘critical’for our business?
•
Who has this knowledge?
•
How can it best be leveraged to boost results?
But how does one go about answering these when faced with a complex business? One growing
trend in the business world is to use methods for designing and maintaining information
architectures in a broader context, to include all aspects of the business environment. DNV
Maritime has applied Business Modelling as a tool for establishing a common understanding of KM
mechanisms in the context of the whole business environment. The objective has been to define and
visualize all elements of concern for our business, and how these different elements inter-relate, in
order to identify and establish effective KM solutions. DNV Maritime’s example of applying
Business Modelling to supporting KM could be applied to other businesses, regardless of which
industry they operate in.
•
Learn about the key business drivers that drove DNV Maritime to establish its
Knowledge Management Program
•
Understand how business modelling can be applied to support KM
•
Learn about the experiences DNV Maritime has had with business modelling in
areas other than KM
Approach
This case study is a story of how one organisation, DNV Maritime, discovered and applied a unique
approach to business modelling for developing and communicating effective KM solutions. The
case covers the following phases of DNV Maritime’s journey:
1. The KM Program in DNV Maritime
2. Discovering of business modelling as a potentially useful tool for KM
3. Developing a modelling framework and meta-model
4. Piloting the modelling framework for KM
133
Case Analysis
DNV Maritime and the KM Program
Det Norske Veritas (DNV) was established in 1864 and currently employs 5,500 people in a
network of 300 offices in over 100 countries. As DNV’s oldest business unit, DNV Maritime has
worked towards its main objective of ’safeguarding life, property and the environment at sea’by
offering classification and certification services for vessels and other mobile offshore units. As one
of the world’s leading classification societies, DNV Maritime has been harvesting the experiences
of its ship surveyors to develop its own standards for ship classification for 139 years.
In more recent years, DNV has expanded globally, which has created the challenge of how to share
knowledge and learn across great distances. Especially in the early 1990s, developments in
information and communication technology (ICT) were harnessed and large investments in ICT
were made. The approach at the time tended to focus on capturing explicit knowledge in
sophisticated systems and less on supporting tacit knowledge needs. The mere complexity of the
business, with its more than 400 services and products, and over 3500 employees spread across the
globe, proved to be a challenge for capturing and sharing knowledge. But this also posed an
opportunity: with so many employees there was also a great wealth of knowledge; knowledge
which could be better leveraged to support business goals. The organisation had just to find out
how.
In DNV Maritime launched a three-year knowledge-management programme in January 2002.
This Program took a holistic approach to managing information and knowledge in alignment with
business objectives and needs. The KM team quickly established the following goals:
1. Conduct a baseline assessment of our current knowledge systems;
2. Recommend measures for DNV Maritime to improve knowledge management;
3. Implement improvement initiatives together with key DNV Maritime stakeholders.
The KM team set out to answer question that would reveal how KM could support existing
operations and service delivery:
•
What knowledge is truly ‘critical’for our business?
•
Who has this knowledge?
•
How can it best be leveraged to boost results?
However, from an innovation perspective this is not sufficient to fully capitalise on the knowledge
an organisation has. One should also look at how existing knowledge could potentially benefit the
business in the future. The KM team was therefore also interested in questions such as:
•
What do we know?
•
Can this further benefit our business and others?
•
Who knows it?
•
How can we make it benefit our business and others?
The main challenge was how to answer these when faced with such a complex business. That was
when the KM team discovered that the IT department’s experience with Enterprise Architecture
(EA) modelling could be combined with their knowledge of Intellectual Capital models to explore
and visualise the business.
134
Discovering Business Modelling as a Potentially Useful Tool for KM
In mid-2002, the KM Team began exploring the potential benefits of Business Modelling as a tool
for establishing a common understanding of KM mechanisms in the context of the whole business
environment. The KM Team, which had competence in Intellectual Capital theory and KM, joined
forces with individuals with competence in Enterprise Architecture modelling, and together they
sought to identify viable solutions.
Several members of the combined team had worked in projects where DNV Maritime had already
been applying Enterprise Architecture (EA) modelling in the context of the Information and
Communication Technology (ICT) environment as early as 1999. Business Process Modelling
(BPM) and IDEF0 methodologies were also being used throughout the organisation. However,
these activities focused primarily on the design and development of specific ICT solutions, and did
not meet the need of understanding the overall business environment. The ICT environment was
mapped out in the Microsoft software, MS Visio. This application later proved unable to provide an
efficient solution for updating and maintaining the rather complex maps over time. There was a
need for more effective software. In 2002, DNV Maritime began experimenting with the Enterprise
Modelling software, METIS, which proved to be a more robust solution for the kind of modelling in
question. In other words, it was able to manage the definitions and visualizations of all elements of
concern for our business, and how these different elements inter-relate.
It soon became clear that one potential solution for exploring the interrelations between various
aspects of the business environment was by building a framework combining Intellectual Capital
models with those deployed by Enterprise Architecture initiatives (for ICT). One could model both
tangible and intangible assets as well as the interrelations between these. The KM Team was
convinced that this would shed light on the complex discussions surrounding KM, and make these
discussions more ‘tangible’for Management to understand and address.
While the team had now established a path for developing the modelling framework, it also
established clear objectives for what DNV Maritime believed it could achieve by applying
modelling in this unique way:
•
define a common description of the business to leverage communication and decisionmaking at all business levels
•
facilitate a general understanding of the business environment
•
explore challenges and improvement opportunities
•
facilitate the analysis of consequences of changes
Developing a Modelling Framework and Meta-model
DNV Maritime’s approach to business modelling combines Enterprise Modelling/Architecture
(from the ICT perspective), Business Process Modelling (BPM), IDEF0 and Intellectual Capital
models and methodologies in a way which enables a common understanding of the impact of the
various factors involved in the business environment, both internal and external.
The team’s focus was on establishing a comprehensive framework which could be used for
modelling, especially with the modelling of KM meta-processes and tools in mind. This included a
generic framework and an underlying meta-model which describes the objects to be used for
modelling as well as how to use these. Figure 1 below illustrates a partial view of the generic
framework. This partial view focuses on External Conditions and Intellectual Capital, and excludes
elements pertaining to Financial Capital, though these are also included in the original framework.
135
Figure 1: A partial view of the DNV Maritime Business Modelling Framework
The generic framework was accompanied by full descriptions of each of the elements included in
the framework. In other words, a thorough ‘DNV Maritime interpretation’of External Conditions
and Intellectual Capital (the latter was based on the works of reputed thinkers within that area such
as Sveiby, Edvinsson, Roos).
Figure 2 illustrates the meta-model, which describes the objects to be used in modelling.
136
Figure 2: The DNV Maritime Meta-model
In addition to developing the framework and the meta-model, the Team focused on establishing
governance for all modelling activities. Some of the main requirements to modelling in DNV
Maritime were:
•
Compliance with Business Modelling Governance Protocols and Procedures
•
Use of METIS software (delivered by Troux Technologies Inc)
•
Compliance with the following standards for process modelling:
o IDEF0 –to be used for information flow modelling
o BPMN (Business process modelling notation) –to be used for work flow modelling
•
Availability of trained, centralized staff (METIS tool & modelling skills)
Piloting the Modelling Framework for KM
Once the framework and the governance were established, it was time to pilot these for KM. In
order to address strategic KM issues, it was necessary to model some of the key elements of DNV
Maritime’s overall business environment. Among these were the objectives, vision and values of
the business were modelled, as well as with key strategic decisions which KM should support. This
137
activity enabled the modelling team to familiarise itself with both the software (METIS) and the
meta-model. It was decided that only very few individuals should be involved in this, as the
software and the meta-model required extensive training.
OBJECTIVE
To Safeguard Life, Property,
and the Environment
VISION
To be our Customers' First
Choice in providing services
for Managing Risk
VALUES
We never compromise on
quality or our integrity
Figure 3: Modelling key elements such as objectives, vision and values of a business
Modelling specific services and the work process for delivery of these was also deemed necessary
in order to lay the basis for modelling the KM meta-processes. DNV Maritime had already
modelled work processes using IDEF0, so this was not done from scratch. However, these process
models had to be developed within the METIS software, something which took additional time.
Figure 4 illustrates a partial view of services and corresponding work processes modelled in the
framework.
138
Figure 4: Modelling services and corresponding work processes
The pilots run by the team focused first on business objectives and goals, and then on the processes
involved in achieving these. In that sense, the pilots followed a process-oriented approach to
identifying potential KM meta-processes and tools (i.e. ICT systems and applications) through first
modelling the information and knowledge needs within those processes. Figure 5 illustrates a
partial view of this process-oriented modelling using the framework.
139
Figure 5: Process-oriented modelling
The pilots included modelling a select few KM meta-processes (depicted in Figure 6) and their
potential and/or current contribution to improving operations. Again, a process-oriented approach
was followed, where the interrelations between the KM meta-process and the core work processes
were modelled. Some of these KM meta-processes or initiatives which were included in the pilots
were:
•
Communities of practice (DNV Maritime was piloting several of these through the KM
Program)
•
Knowledge-enhanced production system (DNV Maritime’s in-house production system,
called ‘Nauticus Production System’or NPS, was being upgraded to deliver more contextspecific information and knowledge to employees in select work processes)
•
Web content management scheme (A new regime for governing the use of DNV Maritime’s
web channels had recently been implemented)
140
Figure 6: Visualising KM meta-processes and the interrelations between the work processes
and critical knowledge, competence and ICT applications in the overall context of the
business
Initially, the main challenge was to identify how (if at all) Business Modelling could support KM
and convince Management to invest in a pilot. Several discussions to convince Management that
this was worth piloting proved to be difficult, as it was nearly impossible to demonstrate how this
could be used without actually running a pilot in the first place. Management did consent to
funding ‘some exploratory activities’, although this was half-hearted by most. However, there were
a few ‘visionary’managers who showed their support early on. One of these had been responsible
for initiating the Enterprise Architecture initiatives for ICT.
Another challenge was the fact that in order to run a pilot, the KM Team had to first establish a
comprehensive framework and a meta-model. This required a good deal of resources before a pilot
could be run successfully. The KM Team used this opportunity to run several ‘mini-pilots’which
also served as a re-iteration process for the framework itself. In other words, demonstrating how to
apply the framework and improving it went in parallel.
During the pilots, modelling a select few KM meta-processes in this framework enabled the KM
team to understand the role KM plays in the business environment. It facilitated a discussion about
whether the current KM initiatives were focusing on the high-impact areas of the business, given
the explicit objectives and goals at the time. It also enabled the Team to communicate what
constitutes KM and how these activities create value for the business.
141
After the pilots, this approach to Business Modelling proved to be a beneficial tool for supporting
better decision-making in other aspects of the business by facilitating discussion around the
following questions:
Will this involve change
in the global DNV
Maritime organization, or
is it primarily a localized
effort?
Will this
improve/enhance
existing internal work
processes and
organization, or will it
create something new?
Will this meet existing
goals or will it create
new ones?
Will this imply a change in
attitudes and behaviours
of employees?
Will this improve/enhance
an existing service, or will
it result in a new one?
Will this require more
human resources than
we have today, or will it
contribute to building
existing resources?
Will this impose changes
to our ICT
infrastructure/tools?
Will this improve how we
capture and learn from
knowledge and
experiences?
Figure 7: Business Modelling enables better decision-making
Exploring the use of Business Modelling for KM has generated several discussions about how to
improve other aspects of the business, especially work process development. Modelling processes
in order to identify knowledge and information needs and potential solutions has resulted in
increased interest among middle management in applying modelling to work process improvement
initiatives beyond KM.
Conclusions
DNV Maritime’s unique approach combines Enterprise Modelling/Architecture (from the ICT
perspective) and Intellectual Capital models in a way which enables a common understanding of the
business environment in which an organisation operates, providing the KM team and Management
with the information they need to make better decisions on how to leverage critical knowledge in
the most efficient way possible. The approach has also proved an effective way of communicating
KM and its value to the rest of the organisation. In addition to improving decision-making for KM,
the KM team’s successful efforts in establishing and applying a holistic framework and meta-model
has also resulted in a positive change in how the organisation approaches organisational and process
development.
DNV Maritime’s approach has begun to support management decision-making within the business
beyond just KM. A similar approach could be applied by other organisations to:
•
define a common description of the business to leverage communication and decisionmaking at all business levels
•
facilitate a general understanding of the business environment
142
•
explore challenges and improvement opportunities
•
facilitate the analysis of consequences of changes
While this unique approach requires commitment and discipline to maintain and be applied
properly, it is generic in nature, and could therefore be applied to any organization in any industry.
•
Securing Management buy-in and stewardship is essential for long-term
success of any KM initiative; applying business modelling to KM is no different
•
Explore and engage individuals with different competences in your organisation
to help you find innovative solutions to KM problems.
•
Understand that many KM initiatives require a long-term commitment by
management
•
KM can not be addressed as a stand-alone process; it needs to be incorporated
into other management activities
•
Business modelling can help describe and visualise the all aspects of a business
environment for use in identifying KM (and other business) needs and solutions.
•
Business modelling requires structure and discipline if it is to be useful over
time; it requires resources for governance and support
Acknowledgements
The authors would like to thank Tore Christensen and Geir Hardt (both from DNV Maritime) for their
extensive contribution to the DNV Maritime business modelling framework. Special thanks goes to Bjørn
Haugland and Eirik Andreassen (both also from DNV Maritime) for their support and foresight in seeing
modelling as a worth-while investment when others were in doubt.
References
Citipitioglu Eidal, L. and Hjorth, F. (2003). 139 Years of Managing Knowledge, web article
published
in
KM
Magazine,
3
November,
2003.
http://www.kmmagazine.com/xq/asp/sid.0/articleid.6605089B-A6C6-42A0-AD9A21A7970A2A8D/qx/display.htm
Bontis, N. (2000). Assessing Knowledge Assets: A Review of the Models Used to Measure
Intellectual Capital. Paper, 11 October 2000.
Edvinsson, L., and Malone,M.S. (1997). Intellectual Capital: Realizing Your Company’s True
Value by Finding its Hidden Brainpower. HarperBusiness, New York.
National Institute of Standards and Technology (1993). Integration Definition for Function
Modeling (IDEF0), Draft Federal Information Processing Standards Publication 183, 21
December 1993.
Roos, J., Roos, G., Dragonetti, N.C., and Edvinsson, L. (1997). Intellectual Capital: Navigating in
the New Business Landscape. MacMillan: London.
Schekkerman, J. (2005). Trends in Enterprise Architecture 200: How are organizations
progressing? Report of the Third Measurement, Institute for Enterprise Architecture
Developments, December 2005.
143
Sveiby, K.E. (1997). The New Organizational Wealth: Managing and Measuring Knowledge-based
Assets. San Francisco: Barrett-Kohler Publishers.
Zachman, J. A. (1987). A framework for information systems architecture. IBM Systems Joutnal,
Vol.26, No.3, 1987.
Zachman, J. A. (1996). The Information Framework. IBM Systems Journal, Vol.35, No.1, 1996.
Lâle Ç
lu Eidal was responsible for the Knowledge Management Programme
at DNV Maritime from 2001-2004. She has since been working with further
development of Knowledge Management in DNV Maritime. Her academic
background includes a Bachelor of Science degree in Public Administration from
Middle East Technical University (Ankara, Turkey) and a Master of Science degree
in Mass Communication from Boston University (Boston, MA, USA). Her
professional experience includes Marketing Communications, Corporate
Communications, process development, Content Management and Organizational
Development. Lale is a proud member of the KnowledgeBoard.
Flemming Hjorth was a project manager in the Knowledge Management
Programme at DNV Maritime from 2001-2004. He has also been working with
further development of Knowledge Management in DNV Maritime. In addition, he
has over the last 3 years built expertise in Business model management and
governance. His academic background includes an Information Specialist degree
from The Royal School of Librarianship in Denmark. His professional experience
includes, Community of Practice development, Business model development,
Process mapping facilitation & modelling, content management, information
research and maritime experience from 1 1/2 years of training onboard merchant
vessels.
Jørgen Kadal was a project manager in the Knowledge Management Programme
at DNV Maritime from 2001-2004. During this time he also played a central role in
the development of the new IT production system for the global operations of DNV
Maritime. He has since been working with Business intelligence and initiatives to
generate knowledge through datawarehousing solutions. His academic
background is a Bachelor of Engineering from University of Glasgow. His
professional experience includes working as a classification surveyor and approval
engineer as well as a technical consultant. He has also played a key part in
establishing the new business modelling activities in DNV Maritime.
144
8
From Knowledge
Management to Business
Intelligence
Joël Van Hoolandt
From Knowledge Management to
Business Intelligence
Joël Van Hoolandt, Magma Consulting ([email protected])
Abstract
This article discusses the changes that occurred in a global
knowledge management program within a big financial services
company. The major aim of the initiative was to implement a
worldwide knowledge management programme (in 1998) that has
been adapted, transformed and has moved to what we have today in
place. This article returns the main findings of this project, reminding
the initial scope to the concrete economic intelligence job performed
nowadays, establishing the link between the market segmentation
(per industry line), the kind of knowledge the firm needed to target and
the concrete realisation we put in place.
The adopted methodology is clearly empirical, based on internal
documents and testimonies, shared experiences and conclusions
acquired in a 100,000+ employees organisation. The main findings
are continuous adaptation on project scope level, concrete down earth
application of knowledge management concepts and employee
flexibility.
Keywords: project management; financial services; knowledge strategy; business intelligence
Background
Industrial Context
According to some well-known IT management 'stages of growth' models (postulated by (a) Nolan
(1979); (b) Earl (1983; 1986 and further amendments); (c) Bhabuta (1988); and (d) Hirscheim et al.
(1988), there is a strong link between management approaches, especially technical and information
systems structures and the growth of organisations. When the international initiative had been taken
within a major financial services company to start and support the implementation of a worldwide
knowledge management strategy, it has been postulated that all involved countries, lines of
business, management department were on several and different development stages, according to
the national IT culture and the business development level reached.
With a strong willingness to maintain a high level service delivery model, plus a strong corporate
image, the company took several measures for achieving its goals and promote what had been
perceived, in those times (1998), as a knowledge management revolution. The context in where the
project had been started was fairly simple: as a major audit, tax and advisory firm, it had to provide,
within a fast changing business environment, strong advices in order to consolidate the client
relationship. One of the top advantages it had was a first class human network, handling millions of
contacts with the business community, managing sometimes complex and difficult cases, but not
taking the time to centralise and share this tremendous amount of information.
The firm engineered knowledge management processes, based on major economic questions, like
establishing a digital nervous system and a knowledge dashboard for client service improvement.
The project team wished to integrate the firm's intellectual capital, with setting up an universal
business management tool and gathering information from all other knowledge based systems. The
147
international management board wanted to get an universal business management tool,
encompassing the policies, procedures and systems associated with the creation, collection,
safeguarding and dissemination of the firm’s intellectual capital. After having identified the major
KM leaders (in those times, it was Microsoft, General Electric, General Motor, the US Navy,
Fannie Mae, Disney, Monsanto, Seidmans and Philips) the management team proposed and branded
a clear project scope to support the delivery of the highest quality client service around the globe;
building channels to the clients that leverage the client’s expertise internationally; creating an
unmatched knowledge sharing environment for superior performance and people retention and
executing an on-going R&D collaboration with key strategic partners (academia, clients and
vendors).
On a research side, the firm had to negotiate external content acquisition in order to reduce
acquisition costs and avoiding double content and costs control. Therefore, an international
initiative was taken, before setting up local national knowledge management teams, for starting
offices of intellectual capital, targeting the improvement of news filtering to the knowledge
worker’s desktop; privileging some sources within context, enabling source specific searches,
developing existing native language filters enabling personalisation and linking to news stories. For
achieving those goals, some technical functionality was to design national country news
homepages, inserting news in context, proposing some free-text searches in context or some freetext searches within a specific publication, sorted by relevance or time. Some filtering options were
already, in those times, integrating competitive intelligence on discipline levels, product rollups or
combinations of both of it.
Project Deployment
After having fixed a general global strategy, a project deployment strategy to the local countries
was proposed and explained. The set up of the intellectual capital office was clearly in charge of the
content that had to reflect the discipline’s core strengths and key markets. The project roadmap
provided major jobs descriptions assigned per roles: content managers; subject matter experts;
global knowledge managers and knowledge manager coordinators. Those profiles have been
assigned in every project countries, invited to prepare a content inventory, following a content plan
and a content management process. These parts of the project will be discussed in the Approach
section of this case study.
Problem
The starting point for pushing up a worldwide knowledge management program was the
competitive pressure that had to be handled, with a correct answer to the market needs.
The main competitors were focusing their strategies on having core knowledge-based solutions
teams to spearhead the KM initiative and to create key proprietary products that can be
complemented by other KM products available on the open market. Those major competitors were
concentrating on organisations that were ‘knowledge-centric’in their missions and were working
with them to help realize their goals and to build internal expertise.
One other major market player was to leverage its own experience with its KM program and offered
KM services to clients, primarily in conjunction with other types of services, such as supply chain
management and customer value management.
Another extreme market segmentation that had been established (early 90’s) was the leveraging of
experiences and alliances in data mining and data warehousing, along with the internal knowledge
management expertise. Some KM services were addressing the issues of creating, harvesting,
148
sharing and leveraging knowledge, while paying attention to some critical enablers, including
people, culture, processes, structure and technology1.
On a client level, the biggest companies had a KM approach for targeting new products and services
to customer segments. Reduction in time to implement test findings was reduced from eight months
to three months, with a 40 percent increase in response rate from targeted prospects. Some large
pharmaceutical company developed a KM system to reduce drug development time and improve
regulatory approval. Regulatory filing time was reduced from 18 months to three months and
regulatory approval time from three years to nine months.
In the energy sector, a KM solution is based on leveraging knowledge from projects around the
world with the goal of reducing the total installed cost to their clients by 30 percent in engineering
and construction of industrial facilities. The initial implementation is considered a significant
contributor to more than $100 million in new projects.
Competitors developed programs to motivate and enable teams to communicate and record
knowledge. The aim was to build a source of information that would enable people to learn from
the experience of others throughout the organization.
•
Strategic KM market segmentation within an audit firm
•
Review of evolving information systems strategies in three major products
•
Extended relationship between strategy and KM objectives
Approach
The approach that has been followed is an internal empirical analysis, participating into the project
and its evolution. We have analysed the project documentation from an internal perspective,
integrating personal experience and colleague’s feedbacks.
This paper reflects an opinion about the project changes and tries to give evidence about the cultural
change that happened in a big financial services firm.
Content Management Process
The project roadmap provided details about steps associated with each phase, individuals (identified
as ‘performers’) who are recognized to execute each step and appropriate guidelines.
According to the project management roadmap 2, the content management process was based on
four major actions:
1) The author/content identification with as recommended action to identify content;
2) Index and classification, with indexing and providing some URL references or required
classifications;
3) Contribute to content manager, within the required classification;
4) Maintenance steps for the edited content [life cycle review, expire/update/deletion
processes]
1
Internal Research services report, research services, Montvale, 1999.
2
Internal knowledge management project documentation –Boston conference –1999.
149
Strategic alignment
Under the pressure of its global market, the strategic move had to be consistent with some other
industry cases, taking into account all major market segments we could identify in the late nineties.
Those selected segments were: Commercial, Industrial and Manufacturing (CIM); Information,
Communications, Entertainment (ICE); Health Care (HC); Financial Services (FS) and Public
Sector (PS). The main strategic objectives of those segments had to be identified in order to propose
to clients a realistic knowledge management approach answering concrete needs and requests, like
in any other studied firm. In fact, we could summarize this project step as a best practice study
before the worldwide deployment began. The reflection was based on four major targets:
1) Customer intimacy [use of knowledge management for answering the customer service,
customer support and customer relations]
2) Operational excellence [cost reduction, process improvement, communication improvement]
3) Product leadership [new product enhancements and innovation to market quicker]
4) Employee expertise [access to best practices and better decision-making]
A summary of the market segmentation and identified knowledge targets is shown in Table 1.
150
Table 1: Market segmentation and knowledge identification
Rank
Knowledge target identified
Segment
1
Operations excellence and improved communications. Employees need to share
complex, tacit knowledge.
CIM
2
Focus is on processes that will yield specific business results; reduce the need to
send experts to remote site to solve problems.
CIM
3
Bring innovations to market quicker, improve operational efficiency across
business lines and serve customer better.
CIM
4
Improve operations through cost reduction and drive for best practice
CIM
5
Avoid duplicate work and ensure re-use; improve employee expertise by sharing
leading practices and lessons learned.
CIM
6
Motivate people to share best practices knowledge; emphasis on re-use of previous
knowledge; improve chip fabrication operations.
ICE
7
Capture and systemize much of the knowledge in projects with customers;
improve customer value and services
ICE
8
Hire customer-centred people who gave excellent customer feedback, thanks to
available technicity
ICE
9
Increase flow of information across the organization through workflow process;
improve customer service
ICE
10
Allow practitioners to access a single database to apply the knowledge of the
organization to an individual's healthcare issues.
HC
11
Use of knowledge management to improve product development and operations
(produce faster and cheaper).
HC
12
Speed up the employee decision-making process, speed up product delivery and
improve customer service.
HC
13
Reduce customer response time from days and weeks to hours and days,
regardless of geographic location; increased promotions of the talented people in
the company; increasing morale.
HC
14
Becoming a clearinghouse for information on economic development; help
employee by reducing the number of reports they have to compile, quality
improvement and reduce publication production time.
FS
15
Align and support business strategy; connect with all stakeholders such as
suppliers, debtors, customers and regulators; improve underwriting operations
FS
16
Transform the physical knowledge sources into a virtual environment; reduce the
amount of information search time.
FS
17
Ability to collect process and disseminate a steady flow of information and deny
the enemy ability to gain and use battle-relevant information.
PS
18
Transform a large bureaucracy in a service organization.
PS
19
Improve operations
PS
According to this knowledge management utility study, the advisory firm had to create internally
similar conditions for answering its client needs as far as it could. This strategy is formulated and
described in the next section.
151
Case Analysis
Executive Summary
The Office of the Global Chief Knowledge Officer manages the single knowledge environment
through which any person –within or outside the firm –can access the firm’s intellectual capital.
The system is comprised of three main products groups:
•
Knowledge sharing supplies all the tools, applications, processes and methodologies to make
information available to employees, including internal document management system and
external news.
•
Collaboration provides a protected environment where teams and clients exchange information
using a virtual workspace tool.
•
Internet products enable the firm to showcase the depth and breadth of its capabilities through
Internet and extranet sites, including international .com domain names.
Each product delivers a constantly growing list of tools, processes and applications, where the
purpose of the tools is to assist practices in planning their knowledge management strategy by
providing descriptions of each of the global knowledge management tools. By using the system and
the local KM teams, the practices can compare tool purposes, features and functionalities and
determine the best tools to use for sharing knowledge within teams, throughout the firm and
ultimately, with clients.
The selection of a tool or particular shared environment (or a combination of) is determined by the
type of content to be shared and the community for whom the content is intended. Documentation
was made available to provide information to assist practices in making some determination.
Local Knowledge Sharing System and Support from Regional Services
The Knowledge Management environment, which is accessed by firm members all over the world,
serves as a corporate gateway to both global and local systems, even if some local systems play an
important role in knowledge sharing.
Practices wishing to develop their own presence within the global knowledge environment had first
to consult their respective regional services group: EMA (Europe, Middle East and Africa), ASPAC
(Asia Pacific) and LAC (Latin America and the Caribbean).
Contacting regional services ensured that practice initiatives were aligned with regional and global
strategies. In addition, regional services could assist in finding and completing necessary
documentation.
When developing new local knowledge sharing initiatives, practices were strongly encouraged to
use tools defined in the project documentation. If there were some concern that the global systems
did not meet business needs, regional services could facilitate consultation with the Global CKO
Office. In some cases, when requirements were of global relevance, the Global CKO Office could
develop enhancements to the global tools to meet practice requirements.
Getting Started with the Knowledge Management System
The Global CKO office provided a variety of information about the tools and products available
(including project management guides, training guides, communication materials and other
documentation). These documents were located in the worldwide repository and gathered in the
Global CKO Web site.
152
Additional assistance in getting started with the tools (e.g. developing a local intranet sub-website
or requesting a public mail folder) was provided by the regional services group, which had to assist
in securing CKO project approval for projects, completing forms and other tasks as appropriate.
Knowledge Sharing
A summary of the knowledge sharing entry points, their purpose and key features/functionalities are
summarised in Table 2.
Table 2: Realising knowledge sharing
Sharing
Entry point
Purpose
Features/Functionality
Major
international
gateway
Access to all knowledge sharing
tools,
methodologies,
contact
Preferred Internet browser start page for
information and intellectual capital.
employees in connected countries.
It’s one portal gathering all tools
and all shared knowledge.
Global Home
page
Start page for all globally relevant The global home page should be used to
news and knowledge tools. It provide access to globally relevant news or
tools.
emphasizes global presence.
Universal
search
This enables easier knowledge
sharing by providing a simple
interface for searching a broad
range of content. With one search
query, users can search multiple
knowledge sharing systems, internal
web sites and certain databases.
Universal searches tool is available for all
users. All practices are encouraged to add
their intranets or other knowledge sharing
repositories to Universal searches.
Global
document
repository
Global document repository which
can be shared within the company.
It allows employees to access any
document submitted by another
employee in any member firm
connected to the system.
All employees are encouraged to enrich the
global
knowledge
repository
by
contributing with documents, tackling the
following content: clients; competitive
intelligence;
contact
information;
engagement
processes;
insights;
methodologies and tools; postings; sales
and marketing material; vendor and alliance
contract.
Local practice
sites
Simple html page that is developed
and maintained by a practice. Their
purpose is to provide users in the
practice one single page where links
to all practice-relevant content are
located.
Those pages should be used when a practice
wishes to provide an inexpensive, simple
and easily maintained interface for links to
site that are relevant to the practice area.
Content
Management
Application
(CMA)
All practice knowledge management
organizations should identify individuals for
This is a third party tool available
training as CMA administrators. They will
through the knowledge management
use the CMA for managing content in the
system for practice knowledge
Global Document repository. The CMA
managers and other administrators.
will also be used to manage other internal
web sites.
153
Sharing
Entry point
Purpose
Discussions
This enables employees to share
information in a public forum.
Discussions
assist
in
the
identification of experts and the
growth of human networks.
Discussions are available to all users.
Practices may wish to include discussions in
their knowledge management strategy. For
instance, knowledge managers may include
specific
discussions
in engagement
methodologies or performance managers
may include discussion participation in
knowledge sharing goals.
Clients and
targets
This application provides a single
source of information on clients and
targets. Its purpose is to provide all
employees access to news and
information about clients, targets
and client teams.
Clients and targets are accessible by all
employees. Practices wishing to share
information about client and targets on a
global basis may wish to make population
of the clients and targets system a part of
their knowledge management strategy.
CV system
The CV system is accessible to all
employees. Practices may choose to
This enables connected employees encourage their community to enter CVs so
to quickly and easily find that they may be easily gathered for
employees with specific skills or proposals. Web site, Clients & targets and
practice page administrators may wish to
experience.
use some function to create links to groups
of CVs.
Internal web
services
Internal web services provide
practices with existing internal Web
sites global server hosting and
support services.
The internal Web Services meet the needs
of practices with the existing internal Web
sites (built with basic technologies and
containing globally relevant content) that
wish to simplify their support processes
while making their site globally accessible.
Sub web sites
This enables any community to
create a small-scale knowledge
sharing system based upon a
template environment. It provides
alternative ‘front-end’navigation to
the content on the Global
documents repository as well as
additional capabilities by using its
Resources & Information area. It
allows practices to draw upon the
global knowledge repository by
inserting links to document loaded
in the Global repository.
This meets the needs of practices wishing to
quickly establish a multi-level Web
presence with minimal resources spent on
design.
Collaboration
The collaboration tool allows teams of employees and their clients to work together in a protected,
virtual workspace regardless of their geographic location. Teams share project documents, ideas,
discussions and proposals, resulting in improved communications, relationships, efficiencies and
revenue.
154
The web space collaboration tool can also be used for internal collaboration by internal groups with
no need for collaboration with external parties. In all cases, the collaboration tool is intended for
collaboration on work in progress. Final work with global relevance is entered in the Global
documents repository for general knowledge sharing.
External facilities are recommended for proposal teams seeking to address client collaboration
requirements. Internal facilities are recommended for teams wishing to share work in progress that
is not appropriate for sharing globally or with other knowledge sharing tools.
The secure web space provides a place for information sharing and idea generation. It provides a
central location for sharing knowledge (calendar, contacts, sharing work in progress, etc.) within
local teams and teams that are geographically dispersed.
Internet Products
A summary of the key internet tools used, their purpose and key features/functionalities are
summarised in Table 3.
Table 3: Internet tools
Sharing entry
point
Dot com web
site
National web
sites
Extranets
Purpose
International web site. It
provides news and information
about the company on an
international level and its
services. It provides also access
to the national web sites of the
countries of the group.
This tool provides the local
firms a template-based tool for
quickly developing a simple
Internet site that is aligned with
general look and feel standards.
This market segmented extranet
approach is designed to provide
worldwide value-added content
to selected clients through the
Internet.
National practices wishing to present
themselves as one of the group. Local practices
with news to report to the worldwide Internet
audience should do so by submitting ideas to
their national Marketing and Communications
teams or, if they are internationally relevant, to
Global Marketing and Communications.
This tool meets the needs of practices wishing
to establish an Internet presence aligned with
the international look and feel and wishing to
spend minimal resources on design and
development, focusing instead on content.
This initiative is coordinated by a worldwide
Innovation centre.
Key Findings
The major findings of this international knowledge management project have been that the firm
found a rapid market leader role in knowledge management initiatives for itself and for its major
clients. The company improved the speed of his information flows from a rather traditional way
(older common business communication channels like phone, fax and internal snail mail systems)
to an innovative tool collection, integrating knowledge sharing possibilities mentioned here above.
But the biggest project realisation has been to maintain international knowledge management
structures in place, even within the bad times, in order to achieve all the long-term project goals, i.e.
reaching a real change management in the cultural professional behaviour of all employees, passing
from a traditional organization to a knowledge-driven company.
155
The major key findings that have been identified after a seven year project life cycle are:
•
Content, improved in accuracy, quality and controls, thanks the designation of some knowledge
managers within the practices (audit, tax and advisory). Those people are in charge for checking
that the published content is useful to their colleagues and that the information can easily be
located;
•
Technology, the supervision of every new hardware or software implementation had to be the
result of coordination between the local and international IT authorities, the professional
department defending his business case and the KM department. Support and implementation
optimization have to be provided from an end-user perspective, taking into account the
specificities of the requesting business unit;
•
Processes, from several natures have been identified for promoting knowledge management
steps in the daily business and in its communication. The content management course and the
related data protection policy is supervised by the Chief Knowledge Officer and its department.
Regular meetings between the information contributors and the KM team are organized, in order
to evaluate the contribution and the publications procedures that have been put in place;
•
People, the communication and the people interactions are evaluated and promoted every year
in their performance evaluations. Complementary job features and aspects like human resources,
training, marketing and communications are involved in the evaluation criteria of employees
appraisals;
•
Management, where traditional managerial tasks still have to be done and where the KM
project team added some more market oriented operations, like the systematic market watching
processes and promoted on-line visibility on corporate web sites and web search engines.
These core centric competencies, answering to the identified key findings, have been assigned to the
local KM teams.
Content
The first responsibility for local KM teams was to ensure the accuracy of the proposed information,
taking into account the following processes:
•
ensure content ownership and responsibilities through the national practices
•
ensure that content is reliable, accurate, timely and consistent
•
assess and ensure the quality management of the data
•
ensure everyone can contribute and find content in the Global document repository
•
ensure, on a risk management level, the client confidentiality, copyrights and data protection
according to international and local requirements
•
manage appropriate publishing and security controls
•
pro actively research and recommend new local external information sources
•
coordinate legacy system migration
•
coordinate parallel information systems developments
•
coordinate content on national and international level
Technology
Taking into account that the IT department is responsible for the delivery and support of the
technical infrastructure and that a specific IT security officer is in charge of the security, some very
specific tasks have been delegated to the KM department. Those were mainly:
156
•
understanding the Global document repository technologies
•
planning the implementation and upgrades and required training for end-users
•
provide end-users support
•
reporting of access control failures to the main system administrators
•
follow global developments
•
coordinate potential parallel database developments
•
centralize major IT applications in one application gateway
•
coordinate new KM application developments
Processes
Key tasks for processes include:
•
coordination of all KM processes and activities for the national practice
•
ensure that international KM processes, policies and guidelines are effectively implemented
•
change management: design and implement national Content Management in accordance with
global policies
•
national and international liaison to ensure the cohesive delivery of KM strategy and processes
•
risk management, with reducing the risk of fragmentation that could result from other business
units
•
embed knowledge management into appropriate daily business processes
•
ensure that collaborative environment is created where people contribute and share knowledge as
part of their working life
•
ensure ownership and sponsorship exists for KM processes, knowledge centres and information
assets
•
develop procedures to measure, recognise and reward knowledge management
•
develop and implement tools and processes to facilitate knowledge management procedures for
local usage
•
ensure processes are in-place to pre-empt legal problems (copyrights and client confidentiality)
•
coordinate all discipline business processes and identify common practices in accordance with
regional and global requirements
•
encourage continuous review and improvements of these processes
People
Key tasks related to people were identified as:
•
communication: provide global and local communication channels between local and
international authorities, regarding user needs and development priorities.
•
marketing and promotions: agree with business on promotional strategy regarding knowledge
management initiatives
•
training: provide courses on content management systems, upgrades, organizing KM conferences
•
change management: accelerate the creation of a knowledge-sharing environment and facilitate
the adoption of KM processes in the national practices
•
management infrastructure
157
•
provide performance management
•
establish a KM competency framework to assist in performance evaluations
•
propose new ideas exchanges and brainstorming sessions
Management
The main management tasks were:
•
define the KM strategy for local country
•
ensure the national and global integration of all knowledge management activities
•
liaison of all national KM activities
•
market watching about the competitors
•
on-line and off-line visibility about knowledge management and business intelligence initiatives
•
ensure the knowledge management department is seen as a key contributor to national and
international KM initiatives
•
coordination and facilitate the implementation of the annual plans for all knowledge managers
nationally
•
provide reports on KM progresses
•
financial management: budget and cost control
Business Impacts
According to us, the described business case revises some well known stages of growth models, like
the Nolan and the Earl models. We have seen in the context section that important financial services
companies had to react on the market pressure, answering a classical client market demand stimulus
and a competitor pressure. If Nolan argues that the information systems management focus is
concerned with technology per se during the earlier stages of growth, with a transformation point
occurring at the completion of stage three (control), after which the focus is on managing the
organization's data resources, utilizing database technology and methods (Galliers: page 35), we
believe that this shift is not spontaneous or not generated ex abrupto, but that companies have to
switch from technological enhancements to data flow analysis for economic reasons, pressed by the
market and clients expectations.
The revised Nolan model (1989) states also that different parts of a single organisation may well be
at different stages of growth with respect to a particular IT (Galliers: pages 35-36), which is clear
for IT strategies but also, according to us, for countries and KM initiatives. We tried to demonstrate
that international private organisations, implementing decentralised knowledge management
projects shifted from knowledge sharing, collaboration and Internet products strategies to local
organised teams, coming back to market priorities, like competitor analysis and competitive
repositioning using collaboration, knowledge sharing and Internet concepts.
Following research on current information systems planning (which is essentially ad hoc in nature)
and an additional factor, concerning the focus of the planning effort. In the latter context, Earl
argues that the focus has tended to change over the years from a predominantly isolated,
information systems function orientation, through an organisational focus to a competitive,
environmental focus. Earl's argument is essentially that organisations begin their planning efforts by
the first attempting to assess the current 'state of play' with respect to information systems coverage
and IT utilisation. Increasingly, the focus shifts to management concern for a stronger linkage with
business objectives (Galliers: pages 36-37). We experienced a similar approach within the follow
up of a worldwide knowledge management strategy, moving from the systems to the people,
158
concentrating management efforts on change and adoption efforts, tackling more and more business
intelligence questions.
Conclusions
In our article, we have tried to describe a project evolution, following similar IT maturity models
(Earl and Nolan principally).
Started in 1998 in a major audit firm, the worldwide knowledge management programme had to
answer two principal questions:
1. A partner strategic alignment, where major clients were asking for KM competencies in five
industry sectors [CIM; ICE; HC; FS; PS];
2. A market competitive situation, where some KM initiatives were already taken.
The answer to those two conclusions was the official launch of worldwide KM projects, proposing
to work on three levels:
1. Knowledge sharing;
2. Collaboration;
3. Internet products.
The segmentation of those three approaches resulted in tasks definitions and enhancements on
content, technology, processes, people and management levels. Of course, and as supposed in the
Nolan and Earl models, the development of those solutions was not a one-speed process and change
management had to remain a continuous attention point. Seven years later, with a constant effort on
maintaining KM structures in place everywhere in the world, a powerful business intelligence tool
is answering market needs and is coping with though competitors.
Through the KM project, the company achieved a complete coordination of processes and tools,
integrated in the daily business people life of more than 100.000 employees. All those internal users
can access (internal) tools and libraries, regional intranets and publications, skills and experience
systems, servicing clients with collaboration tools and secured web spaces. The conclusions are that
none of this is possible without a constant upper level management support, a continuous strategy
revised at least on a yearly base, and with a major implication of communication, training and
business intelligence marketing people.
159
•
Knowledge management strategies are evolving in stages, like for certain IT
maturity models (Earl and Nolan);
•
Management shifts in KM strategies are not accidental. They express a market
and client demand pressure; once the decision taken, the management has to
support the KM and business efforts with a long term perspective;
•
International Knowledge management projects embed changes in itself, going
from a technical stage to a people focus and, finally, a business intelligence
stage. This means that people involved in KM activities are today using tools
that have been designed some years ago, shifting from a strict internal KM
approach to real market watching management activities;
•
Project monitoring must be continuous. It required fully dedicated resources in
charge of the repositioning of the project scope with market and clients
expectations. This remains a continuous research question about the tools
and dashboards that can be developed, adapted or created for specific
business structures and strategies.
Acknowledgements
For the development and production of this article, I feel a deep gratitude to my KM team for its help and
intelligent support. Thanks also to my family and my wonderful son for being like they are.
References
Andriessen, D. (2004), Making sense of Intellectual Capital, Elsevier, ISBN: 0750677740.
Babhuta, L. (1988) Sustaining productivity and competitiviness by marshalling IT. In Proceedings:
Information Technology Management for Productivity and Strategic Advantage, IFIP TC-8
Open Conference, Singapore, March.
Becker, B. and Huselid M. and Ulrich D. (2001), The HR scorecard, Harvard Business School
Press, ISBN: 1578511364.
Bukowitz, W. and Williams, R. (2000), The Knowledge management fieldbook, FT, ISBN:
0273638823.
Cohen, D. and Prusak L. (2001), In good company –how social capital makes organizations work,
Harvard Business School Press, ISBN:087584913X.
Cope, M. (2000), Know your value? Value what you know, FT, ISBN: 0273650327.
Davenport, T and Beck C. (2001), The attention economy, HBS Press, ISBN: 157851441.
Denning, S. (2001), The Springboard – How storytelling ignites action in Knowledge-Era
organizations, Butterworth-Heinemann, ISBN: 0750673559.
Earl M.J. (1983) Emerging trends in managing new information technologies, Oxford Centre for
Management Studies Research Paper 83/4. In The Management Implications of New
Information Technology. (ed. N. Peircy), 1986, Croom Helm; London.
Galliers, R. (2003) Strategic Information management, Butterworth-Heinemann, ISBN:
0750656190.
Guidelines for KM tools, (August 2002), corporate internal document.
Hartman, A. (2000), Net ready, Mc Graw-Hill, ISBN: 0071352422.
160
Harvard Business Review on Knowledge Management (1999), Harvard Business School Press,
ISBN: 2708117351.
Hirscheim, R., Earl, M., Feeny, D. and Locket, M. (1988) An exploration into the management of
the information systems function: key issues and an evolutionary model. Proceedings:
Information Technology Management for Productivity and Strategic Advantage, IFIP TC-8
Open Conference, Singapore, March.
Knowledge management in Belgium, Roles and responsibilities (rev. February 2005), corporate
internal document.
Kotler, P. and Dubois, B. (2004), Marketing Management, Pearson Education, ISBN: 2744070408.
Neubig, P. (2001), Knowledge Management consulting, corporate internal document.
Nolan R. (1979) Managing the crises in data processing, Harvard Business Review, 57(2), MarchApril.
Norman, D. (1993), Things that make us smart, Perseus books, ISBN: 0201581299.
O'Brien, J. (1999), Management Information systems, McGraw-Hill, ISBN: 0072906111.
Overview of the global knowledge management tools, Office of the Global Chief Knowledge
Officer, (June 2002), corporate internal document.
Peppers, D. (1999), One to one fieldbook, Doubleday publishing, ISBN: 038549369X.
Rao, M. (2005), Knowledge Management –Tools and techniques, Butterworth-Heinemann, ISBN:
0750678186.
Renard, J. (2004), Théorie et pratique de l'audit interne, éditions d'organisation, ISBN: 2708130390.
Rylatt, A. (2003), Winning the Knowledge game, Butterworth-Heinemann, ISBN: 0750658096.
Shailer, D. (2001), The project manager's toolkit, Butterworth-Heinemann, ISBN: 0750650354.
Spradley, J. (1980), Participant observation, Harcourt Brace College Publishers, ISBN:
0030445019.
Thiétart, R-A (2003), Méthodes de recherche en management, Dunod, ISBN: 2100066641.
Tiwana, A. (2000), The Knowledge Management toolkit, Prentice-Hall, ISBN: 0130128538.
Wiig, K. (2004), People-focused Knowledge Management, Butterworth-Heinemann, ISBN:
0750677775.
Author Biography
Joël has nearly 10 years experience in knowledge management, change management,
learning and adoption programs and international project review for the European
Commission. Prior to joining KPMG in 1998, he has been working at SAP Belgium as
technical project leader for the end user community and promoting new technology. He
has been responsible for designing end users support strategies and leading the project
management office program within the Belgian KM department. He has been appointed
national expert for the Belgian federal government in 2003 and teaches at the Lerian
Communication School. He is currently Chief Knowledge Officer in Belgium.
161
Enabling Cross-Project
Knowledge Creation
through “Knowledge
Oriented Project
Supervision”
9
Waltraud Grillitsch
Alexandra Müller-Stingl
Robert Neumann
Enabling Cross-Project Knowledge
Creation through “Knowledge
Oriented Project Supervision”
Assistant Prof. Mag. Waltraud Grillitsch, University of Klagenfurt, Austria ([email protected])
Assistant Prof. Mag. Mag. Alexandra Müller-Stingl, University of Klagenfurt, Austria
([email protected])
Prof. Mag. Dr. Robert Neumann, University of Klagenfurt, Austria ([email protected])
Abstract
The modern economy is characterised by knowledge intensive
production processes, increasing specialisation, division of labour and
specific customer needs. Companies have to face a stronger
competition and the velocity of change has been increasing in the past
years as well as the complexity of corporate environments and internal
processes. Considering these circumstances innovations and the right
management of knowledge become more and more important and are
indispensable for future success. The creation of project teams is a
familiar and good approach to deal with these challenges. The precondition is that the team and knowledge transfer processes are
running smoothly and efficiently. Referring to this more and more
companies need to deal with Knowledge Management (KM) in
projects and how to organise and institutionalise knowledge transfer
and learning as well as how to collect and use project experience and
lessons learned. In practice knowledge created in projects often is lost
when the team splits up and the members return to their tasks in the
organisation. This leads to inefficiency as time and money is spent in
inventing things which are already known inside the organisation. A
systematic and structured way of sharing and transferring knowledge
is required to overcome these problems. The case study provides a
concept and process of a successful institutionalisation of “knowledge
oriented project supervision meetings”to actively accumulate, transfer
and utilize knowledge acquired in projects to improve future business,
through the efficient use of resources and the optimization of project
workflows.
Keywords: Knowledge Orientation, Best Practices, Lessons Learned, Process Improvement,
Background
Industrial Context
The department for business technologies (biztec) is dealing with Knowledge Management in
theory and practice; as an industry funded institute we are working in the field of applied research
and practical implications for partners and externals. One of our partner companies, dealing with
specialised “Customer Relationship Management”(CRM) software wants to optimise processes in
165
Enabling Cross-Project Knowledge Creation through “
Knowledge Oriented Project Supervision”
the consulting department. These consultants are responsible for the implementation of the software
in banks worldwide. Over 90 % of the turnaround is realised with international customers in
Europe, Africa and Asia. Marketing and sales responsible for the acquisition of customer orders;
software developers program the individualised software solutions in house. Marketing and sales
passes the projects to the consultants; their main task is the implementation of the software on-site
at the customer. The consultants are working in different project teams, work overextends them and
there is no time to train newcomers. The consultants are specialists in CRM-software but most of
them are not experienced in organisational change processes which are accompanying software
implementations.
The consultants have changing roles in different projects because the project teams are arranged
depending on urgency and disposability. They have to be familiar with every detail of the software
features to be able to deal with special customer requests. In addition they need competence in
consulting, organisational development and change management to ensure a successful
implementation of the software at the customer. The project managers are in charge of the
coordination of different projects. Due to the fact one teams project manager of one team can be a
participant of another team overlaps and bottlenecks might arise.
As a result of these conditions the training of newcomers proves to be a long-term process and
important projects are predominantly processed by the experienced consultants. To provide good
customer service and avoid internal problems the top management wants to improve project
management through Knowledge Management. As the enterprise has few experiences with
Knowledge Management methods, external partners (biztec) are consulted for the implementation
process. The consulting department is selected as Knowledge Managements first field of realisation.
The exchange of experience in this area promises a high profit because coordinating the software
project processes and the implementation are challenging and frequently similar difficulties occur at
different customers.
Experiences gained in projects are only partly available for colleagues because they are not
systematically shared and documented; exchange of experience happens only through informal
talks. If required meetings within the individual project teams are called up; a general and organized
exchange between project teams does not take place. Frequent problems and solutions are not
analysed and productively used for new projects. Consultants are often out of the office; therefore
informal knowledge exchange is difficult. The systematic approach towards Knowledge
Management should help to conquer these problems and to discover and use synergy potential.
Problem
In “knowledge societies” (Schmitz/Zucker 2003), “knowledge economies” or “knowledge era”
(Logan/Stokes 2004) knowledge and information streams are exploding. Knowledge cannot be
managed like e.g. raw material or machines! Knowledge Management (KM) is mainly depending
on people and their willingness to learn and share knowledge. Cooperation and interaction is needed
in teams, in departments, between leaders and their employees, between employees and
customers/stakeholders to improve knowledge driven processes and procedures. Interventions in
systems are difficult because different types of systems demand different kind of actions regarding
their special characteristics. E.g. Banathy presents a map of system types in which he describes
extreme points for systems like open vs. closed, mechanistic vs. systemic, unitary vs. pluralist,
restricted/simple vs. complex and dominating vs. liberating/empowering (Banathy 1996).
According to this approach all of them have special rules and mechanisms and therefore
interventions have to be planned carefully. Changes in one system also lead to changes in other
interrelated systems and major structural changes, called “revolutions”, may lead to changes in
several system structures (Barber 1992).
166
Ideally structural change should lead to increased efficiency and a higher systems capacity to adapt
flexibly to the environment (Colomy 1990). Autopoiesis and self organisation (Maturana 1985;
Greyer/Zouwen 2001; Bolbrüge 1997; Zeleny 1981) as well as the controllability of social systems
are important questions in this context. KM plays a crucial role in social systems but it cannot be
holistically controlled, it depends very much on the interests of the involved parties and their needs.
For a successful implementation specific guidelines and methods are needed.
The case study outlines how knowledge and potential for improvement are revealed and how
synergies fruitfully used in social systems. Through a knowledge oriented concept the consulting
and implementation process of the partner company is optimised and the possibility to integrate
organisational change management know-how is given. In this process newcomers need to be
involved and a critical reflection of finished and ongoing projects is realised. To establish both, a
customer oriented software as well as high quality consulting and improvement processes KM
needs to be institutionalised on the long-run.
•
Reasons and pre-conditions for implementing KM
•
Concept and process of knowledge oriented supervision meetings
•
Best practices and lessons learned
•
Barriers and success criteria
•
Institutionalisation of KM on the long-run
Approach
The case study describes the concept and process of a successful institutionalisation to foster
communication and sharing among the consultants (e.g. learning and reflection about problems and
solutions as well as organisational change aspects are supervised by externals) through the use of
“knowledge oriented project supervision meetings”. (The term “supervision”is originally rooted in
the American linguistic area and means control and/or monitoring (Scobel 1995).) In this process
“best practices” and “lessons learned” are explored to foster a better planning and realisation of
projects. The consultants have the chance to improve their work and to practice critical reflection.
Internal facilitators are trained for a better future knowledge transfer. Furthermore the possibility for
holistic learning is enhanced through the documentation and transfer of the results to all involved
colleagues.
Because of the aforementioned problems and basic conditions in the working environment of the
consultants, knowledge oriented project supervision meetings are introduced in order to conquer the
increasing pressure of quality in consulting. Project risks like planning-, conversion- and
surrounding field risks (Redlefsen 1997) need to be made conscious and prevented purposefully.
External supervisors accompany the meetings, in order to ensure that relevant, usable final results in
the form of "best practices" and "lessons learned" are developed in the project. In this connection
supervision is not top down guided as a form of control, but is organized as a cooperative, critical
and steered process. Central aims of knowledge-oriented project supervision are: systematic
examination of the own work, the production of ideas and mental impetuses by the experience of
others and gaining insight of colleague’s work. The participants develop individual learning and
consulting competence through the detailed analysis of projects and the discussion in supervisiongroups, whereby their work is substantially facilitated and professionalized at the customer.
The success of knowledge oriented project management lies the balance of the organisational
framework and the supporting IT-infrastructure. Each company needs an own specifically designed
and adjusted knowledge management solution to foster innovation and the development of business
excellent strategies, structures, processes and products. The investment in knowledge management
167
needs to be directly interconnected with the consciousness and ability to change and transfer the
organisation step by step to foster that knowledge transfer processes are running smoothly and
efficiently. Referring to this we deal with organisational conditions and introduce a field-tested
process model to help organisations to foster knowledge transfer and sharing during and after
projects.
Case Analysis
Implementation of Knowledge Management
At the beginning of knowledge management intentions it is necessary to define the strategy and
targets of KM attempts. The strategic orientation regarding knowledge-based processes and
practices depends on considerations about the organisational purpose. A “strategic architecture”
(Hamel/Prahalad 1995) for the purpose of knowledge development has to be created which
determines how the specific company will meet the competition in future. This conception contains
perceptions about the future of the specific company, formulated in universal metaphors, analogies,
symbols and models, which represent the core concepts of the company and clarify the selfconception (Neumann 2000; Neumann/Stingl/Grillitsch 2002a).
Management of knowledge concentrates primarily on explicit knowledge which can be seen as one
of the slightest parts of knowledge inside an organisation. Through these circumstances the
efficiency and actual practicality of knowledge management attempts of this kind can be evaluated
as rather superficial. On the contrary the creation of the context for promotional, organisational,
general conditions (constraints) is the basis for the generation, transfer, actualisation and usability
of knowledge itself. This is why knowledge management should deal with the design of promoting
circumstances for the preferential treatment of the resource knowledge. This leads to the demand
for “Management for Knowledge” and the necessity of a composition of “contextually sensitive
organisational consciousness” and “management knowledge”. The principle of “Management for
Knowledge” follows the thesis of system- and structure-theoretical basis, which says that the
(re)producing self-developed order in the deep structure of the organisation continuously manages
the dealing with knowledge rather invisible through monitoring, interpretation, combination, giving
meaning and handling (Neumann 2000; Neumann/Stingl/Grillitsch 2000, 2002a; Neumann/Stingl
2003; Neumann/Stingl/Grillitsch 2004).
168
„Management for Knowledge“
Vision
Targets
Strategies
Leadership
„facilitator“
Knowledgeoriented
diagnosis
& redesign of
organisational
constraints
IT-InfraStructures
Implementation
Integration
Usage
Communication-, Feedback-, Reflection- & Learning processes
!
intensivation –extension –revision –regeneration
„Organisational knowledge“
!
Figure 1: Guidelines of “Management for Knowledge”
Active Learning by Project Reflection
Rosenkopf (2000) advises to focus on dynamic knowledge networks (e.g. among employees) to
realise competitive advantages. These networks influence technological and performance outcomes
whereby a systematic approach to information and knowledge is needed. (Rosenkopf 2000).
Company project experiences consistently show that projects mostly stay unquestioned. This can
primarily be seen as a reflection of the existing project culture of a company, e.g. a company jumps
from project to project without questioning relevant criteria for success or failure of specific project
steps. If anything is questioned than just who are the ones responsible for the failure: a play of
accusing and justification instead of learning from failures and improving systems, procedures or
processes in the organisation. Learning from each other seems to be out in the age of “distance
learning”via CD-ROM or e-learning tools (Pfeffer/Sutton 2001) but learning by doing and learning
from experiences is very effective. In many cases when problems have to be solved or “new
avenues to fast track thinking and innovation”have to be explored (Rylatt 2003) bringing different
people together helps.
For knowledge-oriented project-initiatives it is recommendable to start with the constructive
reflection of ongoing or already finished projects. In our case consultants who had to implement a
specific IT-system shared their knowledge in continuous project supervision meetings to discuss
their project experiences in a structured way. It makes sense to use projects and the consequentially
gained “lessons learned”and “best practices”in the sense of Knowledge Management to learn for
similar future projects. For that purpose it is only necessary to openly ask and honestly answer
critical questions during project meetings and accordingly during a concluding project-review and
to document the results: (1) What was good, what was bad? (2) What would we change as a project
team, if the project would start again? (3) What did we learn from the specific project? (4) What do
we know now and what didn’t we knew before? With questions like that project meetings get a new
structure. This certainly means a change in the sense of a knowledge-oriented handling of projects
that are always a mirror for the organisation and its organisational culture. The results of the
supervision meetings have to be documented and shared with the colleagues, information and
169
communication technologies are helpful in this context to foster knowledge transfer. The following
illustration shows how “Lessons Learned”can be included to improve new project activities.
Figure 2: Integration of “Lessons Learned in Project Management
Structure of Knowledge Oriented Project Supervision
At the beginning of knowledge-oriented project supervision the project which should be discussed
must be defined, whereby tasks, goals and strategies are considered and the basic conditions for the
project are clarified. On the basis of a prepared, during the meetings supplementable, problem list
the causes and background are analyzed. Resuming solution ideas and a process design are
developed, which are operationalized and concretized in an action-list. The negotiability and
practice fitness of the measures must be evaluated ("obstacle course") through a critical test of the
action-list, whereby simulations and worst case scenarios can be used as supporting tools. Due to
experiences in projects and reflection in the meetings "lessons learned" are documented and
possibilities for improvement are compiled.
170
“Learnings”
Project
“lessons learned“and
titel/topic/tasks
decisions, targets – strategies
background
potential for improvement
because of reflection
Realisation-efficiency
and practicability
of developed procedures
“obstacle course”
critical test of course of action
worst-case-ccenario, simulations
Course of action
Who What Till When Result
Competence
development
Increase of
realisation feasibility
of integration- and
change initiatives
Start !?
Trouble list
Causes
Possible solutions &
Process-design
Figure 3: Knowledge oriented promotion of projects through supervision
A structured and clearly communicated procedure for the introduction of knowledge oriented
project supervision contributes crucially to the successful process of the project. The transparency
of the process reduces uncertainty and contributes to the acceptance by the concerned employees.
Conditions and Culture in Knowledge Oriented Project Supervision Meetings
The supervision takes place in a multi-personal-setting, i.e. the consultants meet in relatively regular
intervals in a "group-system", which stands "differently connected to the institutionalized social
systems" (Schreyögg 1991), whereby each participant has the task to actively bring in a "learning
project". Per meeting a learning project is presented, discussed and analyzed by the supervision
group. “Learning projects”are respective problem fields and challenges as well as solutions from
the work of the consultants, whereby the following conditions should be given:
•
Work basis: The respective subject should be related to the field of activity during the
project.
•
Significance: The questions brought into the supervision group are important for the
respective participant (no "alibi" exercises).
•
Relevance: In the supervision circle behaviour, structure and process questions are raised.
•
Influence: Each supervision participant has direct influence.
•
Process: Problem solutions are developed through an ongoing process.
•
Interaction: Communication and co-operation are crucial preconditions for a problem
solution (link to KM: knowledge transfer processes).
The work of the project managers turns out to be particularly difficult because of the
aforementioned basic conditions. Even more important is their competence development, which
should cover five dimensions (Gregor-Rauschtenberger/Hansel 1993): successful innovation
management, structural, administrative and constructive project management and project leadership.
171
Particularly for project managers these transfer-meetings are therefore of high importance because
problems and solutions in the individual dimensions are represented and analyzed from an own
initiative. This contributes to the motivation of the department, since the individual persons feel
better up to their changing roles and tasks as well as challenging projects. The project participants
see that no project runs smoothly and that there are always challenges and difficulties to be
conquered; problems are addressed and conjoint solutions found.
The quality of the supervision meetings depends on the quality of the participant’s contributions and
particularly on communication and reflection culture in the supervision group. Supervision acting
forms the basis of the "supervision culture" (Petzold 1998), which is coined by perception
parameters, normative guidance concepts, interpretation work and action strategies for supervision
work. In order succeed the desired supervision culture must be coined and institutionalized from the
beginning. The definition of, for the respective context, suitable rules for the supervision meetings
is therefore essential during the supervision process. For the successful knowledge exchange
between the consultants it is crucial to establish a meeting culture, which promotes the open
exchange, the expression of constructional criticism and the development of solutions.
Consequently the consultants know the basic conditions of the supervision meeting process and are
able to act within this framework and are furthermore able to set interventions. They accept the
rules and understand their sense. Furthermore the participants are motivated to examine the
adherence of the basic conditions to contribute actively to the design of a productive culture during
the meetings. Qualitative team supervision provides the feeling to act in a "safe place", in which the
group approximates to sensitive topics guided by a reliable supervisor. Without “direct action and
decision pressure”an area for suppressed questions, conflicts and problematic topics is generated
where these issues can be addressed, reflected and clarified (Pühl 1998, p. 160). To guarantee a high
quality regarding content, the participants receive a precise workflow for the project supervision in
the form of a check list. This list assists the consultants in planning the reflection and presentation
of their projects whereby the following topics have to be considered:
•
Project based review
•
Personal review
•
Critical reflection of the project
•
Future prospects (for running projects)
•
Problem solutions and action list
•
Critical view of the action list
•
Learnings and discussion
The reflection of the projects along different criteria and the exploration of learning experiences are
central for a successful knowledge-oriented project reflection. In this regard the supervisor’s role as
a facilitator is demanded to give necessary mental impetuses to the group, without an anticipatory
solution or a too strong influence on participants. The use of different observation levels during the
supervision process is crucial, whereby e. g. the following three dimensions should be considered
(Scala/Grossman 1997, p. 70 ff.):
•
Personal characteristics of the supervisor
•
Personal relations and group dynamic
•
Formal and informal structures of the organisation
The supervisor is constantly reflecting the processes to be able to set interventions for the
promotion of a target oriented reflection and to foster a fruitful culture within the group during the
whole project supervision process.
172
Institutionalisation of Knowledge Oriented Project Supervision
Theory and practice show that employees spent about 30 minutes per day to search for needed
information to work effectively. This is approximately five per cent of the whole working time.
(Heck 2002). Initiatives to share knowledge or to show ways to use knowledge sources help to
minimise this search time and the related costs. To use the advantages of knowledge oriented
project supervision on the long run the supervision meetings have to be institutionalised in the
organisation. In our case the following steps were undertaken:
•
Commitment of top management for the project supervision meetings
•
Structured procedure for the implementation
•
Training of internal moderators
•
Collection and documentation of “Best Practices”and “Lessons Learned”
•
Reflection and evaluation of the project supervision meetings
•
Improvement of the meetings and the documentation according to participants needs
•
Supporting IT-infrastructure to foster knowledge transfer
•
Clear roles and responsibilities (in meetings and for documentation)
Important for the success of knowledge oriented supervision meetings are open communication and
the development of a productive meeting culture. The team has to know exactly what are the sense
and the outcome of the meetings and about their contribution. If the team is not yet familiar with
each other a team building phase has to be included. In our case two internal moderators were
trained and qualified to manage the process after the external supervision phase. To prepare for this
new, challenging role the two moderators attended an intensive workshop on moderating
knowledge oriented supervision meetings in which following topics are examined:
•
Basic rules of moderation
•
Preparation of the moderation
•
Phases in a moderation process and their characteristics
•
Role of a moderator
•
Success criteria for moderation
In the supervision meetings the new moderators have the possibility to experience their new role
and to further developing their competences. They are supported by the external supervisors, who
are mentoring the future moderators in their intervention and reflection processes. So the
moderators have the chance to get familiar with the procedure step by step (learning by monitoring
and doing) and have enough time to acquire the necessary skills for their new role in theory and
practice.
From the structure-theoretical point of view an integration and use of knowledge result only if the
involved actors (“knowledge agents”) reproduce their knowledge enriched actions, they have to use
“lessons learned”and “best practices”in daily work. Furthermore they refer in their interactions to
changing structures, sets of rules and resources. Through the spontaneous, in a way self-organised
creation of rules, the locally existing knowledge of the knowledge agents is used in the best way. In
rules about learning- and selection processes, the knowledge and the experiences of the different
experts are integrated. Only through the possibility of relating to knowledge in a current action,
knowledge is effective as an “accurate or valid awareness” (Giddens 1984) about a situation or
problem. In the collective reflection of activities/projects the problem solving potential is activated.
New or improved solutions can be found which leads in our case to process innovation. These
173
collective activities lead to a self-referential circle and act as a starting point for further actions,
which finally shape the identity of the system.
Innovation, improvement and new ways of thinking or reflection lead to systemic creativity.
Organisational culture/climate has an overwhelming influence on creativity and systemic creativity
asks everyone to be a leader. Everyone is responsible for “sparking ideas”and their transformation
into “useful innovation”. (Mauzy/Harriman 2003). Knowledge Management has always to do with
change whereby the degree of change should suit the aimed purpose to be effective. The most
important influencial factors for excellent companies’ performance results and according KMattempts are fruitful relationships among people, result-based leadership, communication and
teamwork. (Longenecker/Simonetti 2001).
Discussion
In the sense of “cultural change” knowledge management projects are always highly complex,
multi-dimensional, far reaching challenging areas with numerous impacts (looking right and left,
back and forth, etc.), this results in a mostly consequently, concentrated and adjusted way of a
cascading realisation of separate steps of change. Rethinking and reframing (change of
paradigms, change in consciousness) need to be started at the managerial top-level, because it is
transmitted and carried by them. This means a creation of organisational consciousness through
kick-off workshops and sensibilisation seminars for the realisation of the organisational processes,
the identification of problem areas and for the necessary changes.
The top and middle management activity and commitment is an important multiplier for the
success of knowledge management initiatives. Through the personal commitment and the
willingness to realise KM, multiply available single-activities of KM are constituted as a broad
movement.(Palass/Servatius 2001) The main task of managers lies in the creation of a promoting
work environment, in which many people have access to established knowledge and are able to act
according to it. Managers take the position of role models through their activities. They create
standards, through the desired frankly handling and transfer of knowledge (give a limit and
demand). Managers need to be the facilitators of knowledge management projects and they should
promote management for knowledge as an essential criterion for success.
Besides of an evaluation of “present” data, for an analysis and description of the company’s
situation, a specific data feedback and a resulting holistic company diagnosis in the sense of “dual
management”(hard facts resp. indicators for economic situation, market share, capacity, utilisation,
production course, product-market combination etc. and soft facts like character, identity, sense,
behaviour, communication, climate, culture, management, etc.), a target-oriented realisation of
developed measures and a concerning control of success is necessary. This leads to a cyclic,
iterative process in the sense of a rolling planning to realize the whole target “step by step”. Many
unplanned side effects, backlashes and blowbacks of separate process steps and systemic
interventions need to be played through (in the sense of worst case scenario planning) for the
reduction of unexpected effects. The process of diagnosis carries specific importance. The diagnosis
does not describe just symptoms, but moreover the actual causes for problems, nuisance, emotions
of discontentment and learning-barriers. One should find out, why the organisation is like it is; why
specific results are reached, etc. Every organization is perfectly designed (structures, strategies,
culture, behaviour, etc.) to get the results, it gets. There are also many different methods,
instruments and techniques. We think that the principle of “put the whole system in the room” is
the most useful. This means to work together with a representative community of organisational
members (picture the organisation with its characteristics and ways of functioning) in a workshop
174
on the possible causes of specific problems. This creates transparency; makes the problem
landscape more conscious; decreases one-sided interpretations, prejudices, attribution, fantasies and
projections and furthermore promotes an open discourse process.
Time plays a mostly underestimated and important role for the initiation of planned knowledge
management and change-programs as well as for the right timing (kairos) of interventions (window
of opportunity). The related relevant questions are: “How much change is necessary at what point in
time?”and “How much change is useful at the present development stage?”Through the ongoing
serious, open and collaborative answering of the questions an excessive demand as well as
unnecessary burden can be prevented (following the “right”company’s velocity).
Communication, exchange of information and media work plays an important, but mostly
neglected, role in the sense of „management of knowledge and change“. Communication is
essential in each step of the initiative: at the beginning (clear transmission of vision, target, and
strategy); during ongoing processes (feedback, report of actual things/intermediate results for
motivation). It is essential to consider the right time, intensity, use of media and an informational
and communicative language.
Team development and team supervision should help to build, develop and reflectively assist
teamwork. The separate task and target-oriented configured groups pass consequently the diverse
team-dynamic phases of development till they are a powerful, effective, target-oriented, learning
and result-responsible team which is based on functional trust. The primary interest of the task force
is the improvement of the cause, but they know about the interpersonal relations and their impact on
the task’s success probability. The team members need to know about each one’s strengths,
potentials, affinities, interests as well as weaknesses for the classification and acceptance of roles.
Besides this social competence the teams moreover need to have professional competence and
knowledge regarding project management, method and tool use, techniques of decision-making and
problem solving. If it is necessary this can be conveyed in separate training elements.
Primarily through team supervision the lonesome existence of the realization goal getter is reduced.
Different process- and result owners come continuously together in supervisory groups to expand
the project specific problems. This means they collectively diagnose the specific starting situation,
promoting and hindering framework, plan separate steps of action and alternative processes,
reflect and prove them reciprocally and go back to practice with concrete realisation arrangement
plans. They use the knowledge of many group-members in similar situations. They interchange
knowledge within a community, communicate and pick up role-specific problem areas, expectations
and misgivings out as a central theme. That way they can assist each other, what again leads to
incentives and staying power. Besides this supervisory meetings (escorted by external consultants in
the role of reflectors, process specialists and if necessary responsible for input) the specific teammembers meet each other in the mean time. This helps each other in the way of an “inter-vision”for
the specific project-work and leads to an identification and use of existing employee potential.
In this article we tried to give a name to premises and requirements, which we think are essential for
changes regarding knowledge-oriented initiatives. That this is always getting more important is not
questioned, because only fast (willing to change, learn and develop) organisations are able to meet
the environmental expectations always for anew with intelligence, innovation, knowledge and the
establishment of organisationally specific potential and ability of success.
Key Findings and Business Impacts
For a successful integration of KM in applied project management and furthermore for an
enhancement of excellent procedures cross-project supervision meetings are one possible solution
to transfer best practices and lessons learned. These supervision meeting structures crucially
contribute on the one hand to the social dimension of enterprises (formal and informal
175
communication and information) and on the other hand optimise project processes and give the
opportunity to learn from already realised projects (not reinvent the wheel!). The following listed
key findings are criteria for success if they are considered and treated in a fruitful way. If they are
neglected they turn out to be barriers and future trouble areas.
•
Analysis of the present-situation with self-assessment, employee interviews, management
assessment, etc.
•
Careful diagnosis of weaknesses core-causes
•
Make resources available
•
Ongoing observation, reflection and change of contextual conditions
•
Identification, seriousness and picking up of barriers
•
Define process-"ownership"
•
Search and development of promoters, critical mass and multiplications
•
Promote managerial re-thinking processes and changes in behaviour (e.g.: through
management development, behaviour-oriented organisational simulation, etc.)
•
"Awareness" for psycho-social and complex-dynamic processes within organisations
•
Promote cross-functional semi-autonomous project work
•
Create incentives for employee (self) motivation
•
Install and institutionalise feedback, reflection and learning processes
•
Develop target group-oriented personnel development
qualification, mediation of tools and knowledge)
•
Define measurement criteria for an evaluation of success (Through: what do we want to
realize a successful change process?)
•
Production of visual, vivid and realised results and transparency of success
programs
(sensibilisation,
ATTENTION: Companies have to be aware that there never exists a guarantee for success.
Success lies in the balance of individual, collective and organisational hard and soft facts with “a
hint of spirit”.
Conclusion
The cognitive comprehension about the necessity of an efficient utilisation of knowledge in the
company’s code of practice permanently increases, though the company’s pressure of troubles and
change in an increasing knowledge competition primarily provides the increasing probability for
serious steps towards realisation. For this case not only established theoretical models need to be
generated but also feasible ways of implementation. So that knowledge management does not take
the rash way from trend to myth.
Organisational consciousness, as the ability to think in structures and processes (Buchinger 1997),
is the core principle of knowledge oriented project supervision. This competence is developed
systematically, supported by external supervisors, through the analysis of projects. The internal
moderators as well as the consultants gain the possibility to develop their organisational analytic
way of reflection, which is also essential in project realisations at the customer.
176
But even knowledge management is only a kind of construction of reality among many others and
does not supply universal rules for the development of companies. It lays in the hand of the
company itself to put one’s money where one’s mouth is.
•
Managerial commitment to initiate KM-intentions and for the support during
all different stages ensures a successful implementation.
•
Rolling planning in the sense of incorporation of important learning loops to
avoid trouble-shooting in advance establish a designed process.
•
Agreement on objectives with key players in the organisation leads to a clear
vision and strategy.
•
Take and give time to think, plan, act, reflect and institutionalise for goaloriented and qualitative procedures to guarantee a sustainable result.
•
Team development and team supervision are essential to increase the teamefficiency and -effectiveness in reaching defined goals as well as in the
division and execution of tasks.
•
Contribute work for media, information and communication to give
orientation, ensure commitment and participation and to foster reflection,
pro-action and learning.
•
Use of external resources (e.g.: consultants/trainer) to integrate an external
view (e.g.: professional support) for the avoidance of operational blindness.
•
Identification and use of existing employee potential to motivate the right
people at the right time and to promote the process.
•
Train system thinking to be aware of principles, coherences,
interdependencies, interactions and to be able to plan suitable measures,
procedures, actors and time frames for the implementation and
institutionalisation.
References
Banathy, B. H. (1996) Contemporary Systems Thinking: Designing Social Systems in a Changing
World, New. York.
Barber, B. (1992) Dynamic Systems and Social Change: Neofunctionalism and the Theory of the
Social System. Colony, P. (ed.), The Dynamics of Social Systems. London.
Bolbrüge, G. (1997) Selbstorganisation und Steuerbarkeit sozialer Systeme. Weinheim.
Buchinger, K. (1997) Supervision in Organisationen: Den Wandel begleiten. Heidelberg.
Colomy, P. (1990) Revision and Progress in Differentiation Theory. Alexander, J. C.; Colomy, P.
(eds.) Differentiation Theory and Social Change: Comparative and Historical Perspectives.
New York.
Giddens, A. (1984) Interpretative Soziologie. Eine kritische Einführung. Frankfurt am Main.
Gregor-Rauschtenberger, B. and Hansel, J. (1993) Innovative Projektführung – Erfolgreiches
Führungsverhalten durch Supervision und Coaching. Berlin/Heidelberg/New York
177
Greyer, F.; Zouwen, J. (2001) Introduction to the Main Themes in Sociocybernetics. Greyer, F.;
Zouwen, J. (eds.) Sociocybernetics: Complexity, Autopoiesis, and Observation of Social
Systems. Westport.
Hamel, G., Prahalad, C. K. (1995 Wettlauf um die Zukunft: Wie Sie mit bahnbrechenden Strategien
die Kontrolle über Ihre Branche gewinnen und die Märkte von morgen schaffen. Wien.
Heck, A. (2002) Die Praxis des Knowledge Managements. Grundlagen – Vorgehen – Tools.
Braunschweig/Wiesbaden.
Logan, R. K. and Stokes, L. W. (2004) Collaborate to Compete. Driving Profitability in the
Knowledge Economy. Canada.
Longenecker, C. O. and Simonetti, J. L. (2001) Getting Results. Five Absolutes for High
Performance. San Francisco.
Maturana, H. R. (1985) Erkennen: Die Organisation und Verkörperung von Wirklichkeit. 2.
Edition, Braunschweig.
Mauzy, J.; Harriman, R. (2003) Creativity, Inc. - Building an Inventive Organization.
Boston/Massachusetts.
Neumann, R. (2000) Die Organisation als Ordnung des Wissens. Wissensmanagement im
Spannungsfeld von Anspruch und Realisierbarkeit. Wiesbaden.
Neumann, R. and Stingl, A. (2003) Knowledge Management Systems and Best Practices. Corporate
Management’s Way to Business Excellence. Proceedings of the IADIS International
Conference. Lisbon, Portugal.
Neumann, R. and Stingl, A. and Grillitsch, W. (2002), “Chances and Risks of Knowledge
Management Systems in Interorganisational Competence Centers”. Proceedings of the
European Conference on Knowledge Management (ECKM). Dublin, Ireland.
Neumann, R. and Stingl, A. and Grillitsch, W. (2002a) Best Practices and Lessons learned in
Knowledge Management Projects. Proceedings of the European Conference on Knowledge
Management (ECKM). Dublin, Ireland.
Neumann, R. and Stingl, A. and Grillitsch, W., (2004) Knowledge-Oriented Project-Initiatives or
How Companies Use their Knowledge to Create a Sustainable Order of Knowledge.
Proceedings of the Conference on Global Project and Manufacturing Management.
University of Siegen, Germany.
Palass, B. and Servatius, H.-G., (2001) WissensWert: Mit Knowledge-Management erfolgreich im
E-Business. Stuttgart 2001, pp. 68 –75
Petzold, H. G. (1998) Integrative Supervision, Meta-Consultig & Organisationsentwicklung –
Modelle und Methoden reflexiver Praxis. Paderborn.
Pfeffer, J., Sutton, R. (2001) Wie aus Wissen Taten werden. So schließen die besten Unternehmen
die Umsetzungslücke. Frankfurt am Main.
Pühl, H.: Team-Supervision (1998) Von der Subversion zur Institutionsanalyse. Göttingen.
Redlefsen, Ch (1997) Von der Supervision zu Projektberatung. Salzburg.
Rosenkopf, L. (2000) Managing Dynamic Knowledge Networks. Day, G. S. and Schoemaker, P. J.
H. (eds.): “Wharton on Managing Emerging Technologies”. New York.
Rylatt, A. (2003) Winning the Knowledge Game. Smarter learning for business excellence. Oxford.
Scala, K. and Grossmann, R. Supervision in Organisationen: Veränderungen bewältigen – Qualität
sichern –Entwicklung fördern. München.
178
Schmitz, Ch., Zucker, B. (2003) Wissensmanagement. Schnelleres Lernen im Unternehmen.
Regensburg.
Schreyögg, A. (1991) Supervision: Ein integratives Modell – Lehrbuch zu Theorie & Praxis.
Paderborn.
Scobel, W. A (1995) Leitfaden zur Supervision. In: Wilker F.-W. (Hrsg.) Supervision und Coaching
–Aus der Praxis für die Praxis. Bonn.
Zeleny, M. (1981) Autopoiesis: A Theory of Living Organisation. New York.
Assistant Prof. Mag. Waltraud Grillitsch is researcher/member of the
Knowledge Management project group of the department for eBusiness/
Business Technologies (biztec) and lector at the department for Organizational, Human Resources and Management Development at the “Alpen Adria”
University of Klagenfurt. She is writing her Ph.D. Master thesis in the field of
Knowledge Management in corporate networks and is studying journalism at
the University of Klagenfurt. She is working in specific company projects in the
region.
Assistant Prof. Mag. Mag. Alexandra Stingl is researcher/member of the
scientific staff at the department for eBusiness/ Business Technologies (biztec)
and lector at the department for Organizational-, Human Resources and
Management Development at the “Alpen Adria”University of Klagenfurt. She is
working on her Ph.D. master thesis in the field of knowledge oriented Change
Management on the way to Business Excellence. Besides research and
teaching activities she is working in different company projects in the region.
She studied Business Administration and English in Klagenfurt, Newcastle
(Australia), Beijing and Shanghai (China). She is member of the editorial board
of the Electronic Journal of Knowledge Management-Management Centre
International, London.
Prof. Dr. Robert Neumann is associate professor at the department for
Organizational-, Human Resources and Management Development as well as
senior researcher at the department eBusiness/ Business Technologies (biztec)
at the “Alpen Adria”University of Klagenfurt. He is the scientific head of the
General Master of Business Administration MBA program „Advanced Academic
Business Manager“at the “Alpen Adria”Universität of Klagenfurt and partner of
the Transformation Consulting Group (St. Gallen/München). His scientific and
consulting focus concentrates on Change- and Knowledge Management in
private-/public- and non-profit-organisations. He is member of the editorial
board of the Electronic Journal of Knowledge Management-Management
Centre International, London.
179
10
Sharing Knowledge
Between Generations: a
Case Study about
Managing Apprenticeship
in Italian Districts
Marco Ruffino
Sharing Knowledge between
Generations: a Case Study about
Managing Apprenticeship in Italian
Districts
Marco Ruffino, IRSEA - Istituto di Ricerche sulla Società e l'Economia dell'Apprendimento (IT)
and University of Bologna (IT) ([email protected])
Abstract
The case study analyzes objectives, methodology and outputs of a
tool (KAT - Knowledge Apprenticeship Tool) tested on a group of
SMEs in Emilia-Romagna Region (Central Italy), conceived to
evaluate the coherence of the learning resources owned and used by
the apprentices and the enterprise. The tool aims at supporting
(micro) decision processes related to labour organization and human
resources management, in order to improve contextual knowledge
transmission and to reduce the weight of critical factors. The audit
model is based on Nonaka’
s knowledge management processes
(socialization, exteriorization, combination, and interiorization) with
regard to three typologies of variables: capability, possibility and
willingness to share knowledge. On the whole, over 70 weighted
variables have been sorted out, and for each of them a Likert-like
measurement scale has been conceived. Besides operating the
model, the test highlighted SMEs’scarce attention to planning and
management of relations between expert workers and new recruits,
against an ever-less availability of resources (time, attention … ) to
invest on knowledge reproduction processes. Another critical factor is
the scarce integration between formal and non-formal learning, which
urges to think over pedagogies in knowledge management logic.
Keywords: Knowledge Sharing; Knowledge and Learning Audit; Competencies; Apprenticeship;
SMEs.
Background
Industrial Context
The role of small-firm systems – especially those organised in local production systems and
industrial districts –has attracted attention as a distinctive feature of Italian economic development
(Crouch et al. 2001), especially in the so-called “Third Italy” (Regions like Emilia-Romagna,
Veneto, Marche, Umbria, ..). In a few words, a district is a local production system based on a
“network of SMEs networks” characterized by a high division of labour (each firm focuses on a
specific process segment) and, at the same time, by a strong capacity of coordinating production
devices, thanks to low transaction costs. This is mostly due to the presence of not only marketdriven regulations, typical of the social and institutional features of the local communities, and
specifically to the ability of sharing common values and competencies.
183
Sharing Knowledge Between Generations
In terms of knowledge economy, a district is an economic and social area characterized by an
effective sharing of distinctive knowledge, mainly empirical and tacit, developed through collective
learning processes over time, largely based on cross-communities of practice (Wenger, 1998). This
allows competition (among enterprises) on one side and cooperation (among individuals) on the
other, through learning by interacting and learning by reusing processes, which in turn support the
development and diffusion of incremental innovation. The competitive capacity of a district is
therefore strictly connected to the capacity to manage its shared knowledge (Rullani, 2004),
maintaining a balance between reproduction and innovation over time.
The change of the competitive system, determined by market globalisation, by a different
international division of labour and by technological innovation, often has a critical impact on
districts. In terms of knowledge economy, it brings three negative effects:
•
higher key knowledge (re)production costs, caused by the loss of efficiency of transmission
processes (due to the higher flexibility demanded to labour force and to the reduction of onthe-job training);
•
probable less return on investment in learning, due to high staff turnover, especially those
with distinctive professional competencies;
•
less proportion of unemployed to invest in learning, preferring to achieve those “practical
skills”which are heavily losing social value.
In many cases the local intangible assets (knowledge, connections between enterprises, social
capital) are “consumed” before they can be reproduced and innovated: as to say, the speed of
collective learning processes is lower than the evolving speed of exogenous competitive factors.
The districts’spontaneous social capacities seem less and less adequate to guarantee an adequate
management of the requested knowledge (Albertini, 2002).
One of the reasons of districts’weakness is SMEs’reduced capacity to manage their knowledge, in
response to the broader trouble of evolving from a “material culture” to an “intangible assets
culture”. SMEs’knowledge is mainly informal and not stored; they have scarce awareness of its
value as a competitive factor; they have scarce economical, organisational and know-how resources
to invest on the creation and maintenance of knowledge capital (CEN, 2004). Meantime, SMEs’
social and productive characteristics are coherent with a knowledge-oriented vision, if they are
suitably managed: in theory, the strong weight of relational capital and of social networks reduces
the costs of professional resources integration; the organisational flexibility arising from the
“strength of weak links” may reduce the cost of day-by-day adaptation processes; in many cases,
SMEs are better than big enterprises in producing and managing the “multiplicity”(of situations,
resources, … ), with considerable effects in terms of creativity, innovation and use of cognitive
capacities.
The response to SMEs and districts’new needs requires knowledge-based economical policies,
integrated with technological and organisational innovation policies (Corso et al., 2003). An
important aspect is the qualification of education and vocational learning systems (formal learning)
and of apprenticeship, the lasting reproduction process of knowledge –largely tacit and empirical –
that gives rise to local productive capacity. Nowadays, it stands clear that expert workers often lack
of adequate capabilities of knowledge transmission and that new recruits often lack of learning
capacity (ENFEA, 2002). In particular, the cost of learning frequently seems higher than the
immediate economical value of transmitted knowledge, and this proportionally reduces the social
willingness to undertake knowledge transmission processes (Ruffino, 2001).
Problem
Considering the above situation, a key target of economical, educational and vocational training
politics is the reduction of distinctive knowledge transmission costs, locally and at an enterprises’
stage. Apprenticeship is central to this strategy, both in regulations (Italian labour contract reform)
184
and in methods (development of pedagogic models enhancing integration between formal and nonformal learning), and many innovative projects are being developed, some at an experimental stage.
In Emilia-Romagna Region the most important consortium managing apprentices’ learning
activities (Consorzio Formazione & Lavoro, “Training & Work Consortium”) wants to work out
and test a specific KM model for improving apprentices and enterprises capacity to manage their
cognitive relations, removing/diminishing hurdles in knowledge transmission. The model is part of
a broader strategy of quality improvement in vocational training offer (CF&L, 2005), which aims to
define customised vocational training plans for each apprentice. These plans are based on a better
integration between formal learning (delivering knowledge fundamentals in order to enhance the
quality of contextual learning processes) and non-formal learning (developing the practical teaching
capacities and the enterprise’s tutorship).
•
To define practical knowledge transmission problems in SMEs-apprentices
relationship;
•
To define and test a structured approach of knowledge audit, aimed to
understand the critical factors of the SMEs-apprentices relationship;
•
To define strategies for improving learning processes, especially in the relation
between formal and non-formal learning.
Approach
The case study reports the outputs of a test coordinated by the author during 2004, on behalf of
Consorzio Formazione & Lavoro, aiming to define and test an easy-to-use Knowledge
Apprenticeship Tool (KAT), addressed to trainers and expert workers and conceived to evaluate the
coherence of the learning resources possessed and used by the apprentice and by the enterprise with
the kind of knowledge to transfer. The final target of the test is to facilitate the initial diagnosis of
training needs and the following monitoring of accomplished learning, improving decision making
upon management of workers placement. An indirect output expected by using the tool is an
improvement of SMEs’capacity to think about its KM capacities in an organisational development
approach.
The obtained outputs have been analyzed and discussed also referring to the broader reform process
of educational and vocational training systems ongoing in Italy, with special regard to formal and
non-formal learning alternation models.
Case Analysis
The Knowledge Apprenticeship Tool development
The KAT has been conceived as a tool representing the apprentice-organisation relationship in a
more objective and transparent way, in order to support the micro-decisional processes of human
resources and production factors management. The development of KAT urged to face two different
problems: i) the definition of a theoretically solid model of KM audit and ii) the development of a
simple and quick-reference interface for the first users, i.e. the training and enterprise tutors dealing
with apprenticeship at a pedagogic level.
185
Theoretical Framework
The aim of the tool is to support the evaluation of:
•
the coherence of the learning relation between the apprentice and its organizational context,
in order to estimate the efficiency of cognitive resources use;
•
the intensity of the cognitive resources involved in the relation, in order to identify possible
constricting factors.
The objective difficulties in knowledge “measuring”(OECD, 2003) and the need to reduce audit
cost leaded to choose a comparative evaluation model, based on a comparison of the state of the
resources used by each member of the relation, rather than on a direct exam of the relation in itself,
which would require a longer time and more complex analysis tools. Coherently with a KM
oriented approach, the tools needs in fact to be directly and easily used by the persons in charge of
the apprentice’s placement (“enterprise’s tutors”), in order to facilitate “in context” decision
making.
The comparative evaluation brings two needs:
•
to have a set of common landmarks shared by the different players of the relationship (the
“touchstones”that allow the comparison between different objects);
•
to have distinct sets of observative variables, defined considering each actor’s
characteristics and position in the relationship (e.g. the subject who has to learn or the one
who has to transmit knowledge).
This exigency led to the architecture of the audit model, based upon the following choices:
1. Simplifying the relation up to just two “players”: an individual (the apprentice) and a “collective
subject”, as to say the specific part of workspace where the individual works, made by the sum
of the expert professionals working there and of the organizational resources (model of work
division and coordination; procedures; use of time; cultural values; … .);
2. Assuming as a common reference the knowledge management macro-processes described in the
well-known Nonaka and Takeuchi’s model (1995), evaluating for each of them the level of
cognitive resources involved. A correct approach of KM indeed requires looking at the relation
between expert workers and new entries not only as knowledge transfer from the first to the
second ones, but broadly as an opportunity of “hybridization”among practical and theoretical
knowledge. In fact, more and more apprentices show a good level of formal education, in many
cases better than the tutors’ one, and this potentially represents a resource for the entire
organization for supporting exteriorization and combination processes. It is therefore correct to
talk about “sharing”, even though asymmetric, rather than about “transmission”. This approach
focuses on the capabilities of i) using theoretical knowledge as a resource for representing and
learning practical knowledge and ii) using work experiences to give theoretical knowledge new
meanings, in order to “mobilize”and make them productive (CEDEFOP, 2004).
Evaluating the involvement of the apprentice and the organization in knowledge sharing processes
with regard to three kinds of resources (Figure 1.):
•
“learning/transmitting capabilities”, as to say:
a) for the apprentice, having those meta-competences which enable him to activate and
manage cognitive processes coherent with the kind of knowledge and resources
typical of formal and informal learning contexts;
b) for the organisation, having the competencies of managing its own knowledge
(explaining and representing capabilities, pedagogic skills, ..);
•
186
“learning/transmitting possibilities”, as to say:
a) for the apprentice, having the essential knowledge and personal features (physical and
mental abilities, … ) to understand instructions and to reprocess experience;
b) for the organisation, offering the organisational resources (time and working
conditions, expert workers, availability of information, cognitive division of work
stages, … ) necessary to accomplish learning processes;
•
“learning/transmitting willingness”as to say:
a) for the apprentice, the rate of involvement and willingness towards the specific
position and role assigned by the enterprise;
b) for the organisation, the coherence between behaviours and ways of knowledge
transfer (acknowledgment of knowledge value; importance recognised to socialisation
and externalisation processes, … ).
Learning/transfert
capabilities
Apprentice learning potential not
efficiently used in the
apprenticeship relation
Overlap area: effective
knowledge transfert
Learning/transfert
possibilities
Organization learning potential not
efficiently used in the apprenticeship
relation
Organization learning
characters
Learning/transfert
willingness
Apprentice learning
characters
Figure 1: The analytical dimensions of knowledge sharing
Audit Variables
The theoretical model was put in practice defining, according to literature suggestions, 77
observative variables, 50 concerning the apprentice and 22 the organization. A stronger emphasis
on the apprentice arises from the need to deepen two different aspects: i) the level of owned
knowledge, especially the one deriving from formal learning, and ii) his/her behaviour in
knowledge sharing process. On the other side, the organization is examined with almost exclusive
regard to the latter aspect.
Each variable has been assigned a theoretical weight and a measurement scale, conceived on a
Likert-like logic and accompanied by brief explanatory profiles for each assessment stage, in order
to accomplish a “realistic” evaluation by enterprise’s tutors. Table 1, summarizes the main
typologies of the used variables.
187
Table 1: Typologies of variables used for the audit
Knowledge
processes
Players
Apprentice
Organization
Socialization
Quality of: comprehension and management
of social relations, comprehension and
respect of procedures, formal and functional
hierarchy. Degree of loyalty and trust.
Degree of belonging to possible
communities of practices.
Formal and effective model of labour
division and coordination. Human resources
structure in terms of age, length in service,
skill and provenance. Capability and
willingness of social interaction. Knowledge
sharing trend. Intensity of the resources
made available to the apprentice.
Exteriorization
Quality of empirical recognition of the key
elements of the labour context. Quality of
reconstruction of one’s own work
experience as “cause-and-effect”relations.
Capability of oral and written
communication of experience to colleagues
and to other possible subjects.
Level of activities formalization. Existence
and use of quality systems and of non
conformities recording system. Frequency
of explanations to be given to third parties.
Level of causal knowledge in the
organization.
Combination
Quality of information sources research.
Level of use of theoretical and formal
knowledge in explaining phenomenons and
consequent adopted behaviours. Approach
towards formal learning.
Interiorization
Quality of contextual learning processes:
importance of improvements in the capacity
of using technological and organizational
resources, of understanding and controlling
variances; social relations management.
Frequency and quality of finding and using
written sources. Way of decision taking with
regard to innovation needs or solving
persistent problems. Presence and efficacy
of continuing training practices.
Quality of contextual learning processes:
importance of improvements in the capacity
of using technological and organizational
resources, of understanding and controlling
variances; social relations management.
Tool Output
The main target of KAT outputs is to encourage the cognitive activation –in the sense of “enacting”
given by Weick (1979) – of the actors involved in apprenticeship, helping them to become more
aware of the KM aspects coming into play. Rather than giving “answers”or “keys”to the best way
for knowledge sharing management, the approach intends to spark questions which can lead to a
less “obvious”– i.e. as much as possible independent of enterprise’s original culture - formulation
of the problem. This meets a theoretical trend sustaining a tight relation between knowledge
management and organizational learning (OECD, 2000), the second one being the cognitive
fundament of the first. Without an effective management of collective (i.e. common to all the
organizational players) learning it is impossible to manage enterprise’s knowledge effectively.
At the moment the tool provides two kinds of output:
•
the synthetic comparison between apprentice and organization’s profiles, which highlights the
intensity and the four macro-processes of knowledge management overlap degree (Figure 2.).
The graphical model, very simple and therefore scarcely precise, intends to represent in a single
image the audit’s main features, enabling a global and partly “implicit”perception of the result.
The elements striking the eye (without knowing well why, potential problems are immediately
detectable looking at the figure) are the extent level of the different processes and the different
profiles of apprentice and organization;
•
a degressive ordered list of the most critical factors, based on the intensity of their components
(the audit variables) and on the distance between the apprentice and the organization (as to say
the problems due to non-efficient use of learning potential). This second (analytical) output
wants to stimulate the players (starting from the enterprise’s tutor) to improve either the specific
188
features of the apprenticeship (e.g. apprentice’s work insertion) or, broadly, Knowledge
Management inside the enterprise (e.g. coordination processes, level of labour cognitive
division, presence of procedures and “memories”, etc.).
Exteriorization
Socialization
Interiorization
Combination
Apprentice
profile
Organization
profile
Figure 2: The KAT’s main output: a comparison between apprentice and organization
profiles
The outputs come out from a specific heuristic empirically defined, which revises the different
weighted variables re-conducting them – for each process – to the three categories of capabilities,
possibilities and willingness. The algorithm employed considers some effects of mutual
conditioning of the variables, using an “if … then”logic.
Testing the KAT
The tool was tested on 25 industrial SMEs belonging to 5 local productive systems of EmiliaRomagna Region, for a total amount of almost 40 apprentices, in order to verify:
•
the coherence of the methodology;
•
the intelligibility of the variables (especially the measurement scales) and of the outputs;
•
the usefulness of the audit results for managing the relations within apprenticeship.
The tool was presented to a group of enterprise’s tutors in a specific training seminar, where the
underlying KM principles and the audit variables were discussed. The seminar also represented an
opportunity to evidence the main problems of apprenticeship management, substantially confirming
the critical factors underpinned by literature.
Tutors were requested to: i) express an empirical and de-structured evaluation, independent from
the tool, of the state of the “apprentice-enterprise”relationship; ii) set value on each audit variable
for every “apprentice-enterprise”relationship; iii) appraise the intelligibility of the questions and of
their measurement scales.
Tutors’ answers for each “apprentice-organizational context” couple were elaborated by the
research group of Consorzio Formazione e Lavoro according to the given heuristic, and the audit
189
outputs were then compared with tutors’empirical evaluation, analyzing the reasons of possible
significant variances. The outputs of KAT (redefined, if necessary) were then given back to
enterprise’s tutors, thus verifying their intelligibility and their potential usability for improving
apprenticeship management.
In general, enterprises’tutors appreciated the tool, first of all because it offered the chance to
reconsider knowledge transmission processes to newcomers systematically and more objectively.
At the same time, the main negative remarks were: i) the considerable amount of time required to
survey all the audit variables; ii) in some cases, the uneasy comprehension of some item and
measurement scales; iii) the difficult full use of audit outputs in managing apprenticeship.
Two fundamental problems emerged:
•
on one side, SMEs still have scarce capability to “represent”their knowledge and the related
transmission processes. Many questions seem “difficult” because SMEs aren’t fully conscious
of all the factors that influence knowledge system reproduction. Therefore, in many cases, the
tool represented more a chance for “asking oneself new questions” rather than for “giving
oneself new answers”. In fact, even when tutors understand the problems, they have troubles in
rethinking the organizational management of apprenticeship;
•
on the other side, at the moment in Italy it is very difficult to provide apprentices with a
customized formal learning offer, able to meet the needs pointed out by the KAT. This reduces
the potential of the tool, in this case for reasons not dependent on its logics.
All the gathered informations led to the revision of KAT both in terms of model (reformulation of
some variables; change of the weighs determined by the heuristic) and of textual formulation (way
of proposing questions).
Key Findings
The test pointed out the following important statements:
1. The original working theory is confirmed: reading apprenticeship under a knowledge
management logic is possible and useful (also for SMEs), and it helps to have a broader vision
of the conditions enabling a better cognitive relation between the apprentice and his/her
organizational context. This implies the importance of regarding apprenticeship not only as a
transfer of knowledge from experts to new entries, but rather as an “asymmetric” knowledge
sharing, which implies the relation between formal and non formal learning.
2. The adopted audit model seems correct with regard to the method (logics, audit variables), but it
needs some further developments in users interface (“enterprise’s tutors”, management,
apprentices themselves) and especially in heuristic, still too much dependent on the analysis and
interpreting competences owned by the users. The introduction of a FCM - Fuzzy Cognitive
Maps (Kosko, 1998) - heuristic is under examination, in order to represent the relation between
the different variables in a less deterministic way.
3. Anyway, the primary aim of KAT is to draw the attention and the collective learning processes
of the enterprise on knowledge management topic, helping the players to examine the
apprenticeship relation in a “non obvious”way and to think about the cognitive working of their
production system;
Within the limits of the examined sample, the tested SMEs evidenced four critical factors:
190
•
tacit knowledge at the base of the productive process is not located and represented
enough, and this heavily increases transmission costs;
•
integration between formal and non-formal learning is considerably difficult, with
negative effects on the relation between empirical know-how and theoretical know-why.
Notably, theoretical explicit knowledge seldom facilitates a better understanding of
practical knowledge obtained with working experience, thus increasing learning costs;
•
expert workers find it difficult to adopt “pedagogical” behaviours consistent with the
kind of knowledge they have to transfer. There is an overall scarce attention to
knowledge management, due to the habit of considering learning processes between
generations as a “natural fact”, and possible problems as just “cognitive limits”of the
apprentice (rather than of the “apprentice-organization”relation);
SMEs have less and less organisational resources dedicated to their knowledge reproduction, due to
the need to reduce costs and time because of the new competitive scenario. “Cognitive economies”
which historically favoured SMEs settled in industrial districts are decreasing day by day because
of globalization and of changes occurring in local social dynamics.
Critical factors of apprenticeship and broadly of knowledge transmission between generations seem
much more understandable if a knowledge management approach is adopted (Ruffino, 2005). This
leads to reconsider current pedagogies with special attention to: i) the individualization of formal
learning; ii) the highlight of relations between know-how and know-why (Lundvall and Johnson,
1994); iii) the strengthening of local networks of knowledge transmission (Trentin, 2004), in
learning region logic (OECD, 2001).
Business Impacts
In perspective, two are the expected impacts:
•
costs reduction and higher effectiveness of apprentices’ vocational training outside the
enterprise, supporting their on-the-job learning. To achieve this goal, a more precise needs
diagnosis is requested, with special regard to representation capabilities and to work habits
necessary for learning-by-doing and learning-by-interacting;
•
higher enterprise’s attention towards KM, scheduling specific consultancy interventions and
support to apprenticeship management in order to improve skills in setting out tacit knowledge
and in managing contextual learning processes.
Expected developments of the project (referring to June 2006) are tuning up the tool and putting
into place training of enterprise’s tutors in order to achieve three goals:
•
developing an Internet version of KAT integrated with apprenticeships database in EmiliaRomagna Region, making audit virtually possible;
•
strengthening the actual heuristic, in order to have a more detailed output of audit, as a support
for planning apprentices’customised training activities;
•
having a standing observatory of the state of apprenticeship, regarded as KM practice, and
working in a benchmarking perspective on a middle turn, especially with enterprises of the same
district.
Conclusions
Transmission of knowledge from productive organization to the new recruits is the key process of
the reproduction of an enterprise’s distinctive knowledge. The usual apprenticeship practices, based
upon alternance of classroom training and on-the-job coaching by expert workers, look more and
more insufficient, also in network productive systems like districts. Reconsidering apprenticeship in
191
terms of KM, especially in SMEs, is the key to define new pedagogies aimed to integrate formal
and non formal learning, know-how and know-why. Within this logic, it seems useful and
productive to give the enterprise tools (like the experimented KAT) to diagnose the “cognitive
quality” of the apprenticeship, especially if the tools facilitates the activation of organizational
learning processes, at the heart of every knowledge management practice.
•
Apprenticeship can be analysed and managed in a knowledge management
logic, favouring a knowledge-management-centered approach rather than a
knowledge-transfer one.
•
The developed audit tool seems to be correct as regards the method and the
analysis capacity. The main limiting factor is SMEs’scarce level of competences
in human resources and knowledge management. By now, the tool works more
for “
asking questions”rather than for giving effective answers to knowledge
reproduction problems.
•
Knowledge management requires to activate and manage organizational
learning processes: the final aim is the overlap of enterprise’
s productive assets
(“
work”
) and cognitive assets (“
knowledge reproduction and innovation”
).
•
Even in districts SMEs show ever-higher difficulties in preserving and innovating
their knowledge capital, due to high transmission costs (problems of cognitive
efficiency), to scarce awareness of knowledge management exigencies and to
continuous reduction of resources (coaching time) invested in learning
processes.
•
Nowadays, a strongly limiting factor is the insufficient connection between
formal and non-formal learning. Explicit and formal knowledge is not much used
to reduce the costs of tacit and contextual knowledge learning. The relations
between know-how and know-why still remain difficult.
•
A revision of the methods of alternance and integration between formal and
non-formal learning is required, as well as a higher individualization of learning
offer in apprenticeship. The relations between education, vocational training and
enterprise have to be revised on territorial level. The “
learning region”is an
important reference in this sense.
Acknowledgements
Special thanks to Mr. Edoardino Cavalletto and Mrs. Maria Elena Romanini, Consorzio Formazione Lavoro,
for the role played in testing the tool, and to Mrs. Paola Morini, IRSEA, for the editing of this paper.
References
Albertini S. (2002), La gestione delle risorse umane nei distretti industriali, Etas libri, ISBN: 88453-1183-X
Cavalletto E. (2005), Consorzio Formazione & Lavoro: tre anni di attività nel settore della
formazione e dell’apprendistato, GMPR group (in press).
CEDEFOP (2004), European perspectives on learning at work. The acquisition of work process
knowledge, CEDEFOP, ISBN: 92-896-0273-2
Corso M., Martini A., Paolucci E., Pellegrini L. (2003), Knowledge management configurations in
Italian small-to-medium enterprises, Integrated Manufacturing Systems, Vol 14, No. 1.
192
CEN (2004), European Guide to good Practice in Knowledge Management. Part III – SME
implementation, CEN
Crouch, Le Galès, Trigilia, Voelzkow (2001), Local Production Systems in Europe. Rise or
demise?, Oxford University Press, ISBN: 0-19-924251-8
ENFEA (2003), Indagine nazionale sui fabbisogni formativi nella piccola e media impresa.
Fabbisogni professionali e sistemi di sapere, Neos Edizioni, ISBN: 88-88245-04-9
Kosko B. (1986), Fuzzy Cognitive Maps, International Journal of Man-Machine Studies, 24.
Lundvall B-A and Johnson B. (1994), The learning economy, Journal of Industry Studies, 1, No 2.
Nonaka I., Takeuchi H. (1995), The knowledge-creating company, Oxford University Press, ISBN:
0-195-09269-4
OECD (2003), Measuring Knowledge Management in the Business Sector, OECD, ISBN: 92-6410026-1
OECD (2001), Cities and Regions in the New Learning Economy, OECD, ISBN: 92-64-18568-2
OECD (2000), Knowledge Management in the Learning Society, OECD, ISBN: 92-64-17182-7
Ruffino M. (2005), L'apprendistato come pratica di gestione della conoscenza, in Cavalletto E.
(eds), Consorzio Formazione & Lavoro: tre anni di attività nel settore della formazione e
dell’apprendistato, Bologna, cit.
Ruffino M. (2001), Formazione continua e competenze delle PMI. Modelli, strumenti e standard al
tempo dell’economia dell’apprendimento, Franco Angeli, ISBN: 88-464-2906-0
Rullani E. (2004), Economia della conoscenza, Roma, Carocci, ISBN: 88-430-2839-1
Trentin G. (2004), Apprendimento in rete e condivisione delle conoscenze. Ruolo, dinamiche e
tecnologie delle comunità professionali on-line, Franco Angeli, ISBN: 88-464-5165-1
Weick K. (1979), The Social Psychology of Organizing, 2d ed., Random House, ISBN: 0-07554808-9
Wenger E. (1998), Communities of practices. Learning, Meaning and Identity, Cambridge
University Press, ISBN: 0-521-66363-6
Author Biography
Marco Ruffino is the President of IRSEA - Research Institute on Learning
Economy and Society; Director of Koinè – Knowledge in Network, both based on
Genoa (Italy). He is teaching: "Knowledge management" at the University of
Camerino; "Decision and cognitive processes theories" and "Socials Network
Analysis" at the University of Bologna; "Continuous vocational training needs" at
the University "La Sapienza" of Rome. Professional since 1985, he carries out
research, planning, advisory and training activities in the following fields: learning
organisation and knowledge management; learning region and network systems;
work analysis, professional needs diagnosis, planning and methodological
innovation of initial and continuing vocational training; policy and decision making
of education, vocational training and social politics.
193
11
Knowledge based
Methods in Cost
Reduction Processes
Roberta Cuel
Knowledge Based Methods
in Cost Reduction Processes
Roberta Cuel, Dept. of Computer and Management Science, University of Trento (Italy)
([email protected])
Abstract
This document describes some of the typical knowledge management
processes that a consulting company has to go through when dealing
with cost reduction. This analysis aims at understanding how
consultants manage knowledge in order to effectively produce value
from their activity. In particular, our study has been focused on social
aspects, namely the relations among individuals that allow people to
improve their productivity in terms of expertise in the service activity
and innovation of products and solutions. By considering the tacit
dimension of knowledge, the persistency of old procedures and
routines, the difficulty of unveiling knowledge from social networks, we
have attempted to analyze the firm through the ethnographic
methodology. In other words, we observed some work activities within
the firm (the author and some students have participated to the daily
activity of the company) and interviewed some important members of
the firm. The interviews have stressed the analysis of knowledge
management processes according to the “sense-making approach”.
Following this approach, we will describe how consultants develop a
project of costs reduction, how they manage knowledge in order to
obtain the maximum profit by making innovation, and how they will
reuse their knowledge.
Keywords: cost management in outsourcing; costs reduction processes; behaviours and processes;
sense-making theories.
Background
In dynamic markets (characterized by work specialization, outsourcing processes, just in time and
distributed productions, etc.) companies have moved from hierarchical structures to networked
models, based on both intra-organizational networks among strategic units, divisions, groups, and
so on; and inter-organizational networks, such as industrial districts and knowledge networks
(Hamel & Prahalad, 1990). In these last decades, firms are increasingly focusing their attention on
their core competencies, outsourcing a higher percentage of the total costs of their products. One of
the main explanations for this attitude derives from the fact that suppliers provide products with
lower costs, higher functionality, quality, and technologies, due to the partition of the production
chain to different operators. In this scenario, various specialized activities and groups (Ashby; 1956;
Numagami, Ohta & Nonaka, 1989) might not totally be controlled by a unique subject, and might
grow and differentiate their activities in an autonomous way. These subjects (buyers, suppliers,
brokers, etc.) manage knowledge according to personal interests, often in order to achieve
antagonistic aims. Thus they will cooperate, and at the same time compete, in various stages of the
product life cycle (design, production, and distribution phases), coexisting as in a bio-functional
system (Maturana & Varela, 1980) and creating unexpected combinations of processes and products
197
(Chandler, 1962). This coopetition (a blend of cooperation and competition) is practically managed
by various techniques such as franchising agreements, outsourcing processes, R&D collaborations,
participation to joint ventures and joined spin offs, etc. (Domberger, 1998). The complexity of these
processes, and the encountering of different perspectives and interests, increases the difficulty of
negotiation processes among different organizations (Cooper, Slagmulder and Drucker, 1999) and
cost management processes dependent from various actors and organizations. In this scenario, cost
management encompasses a broader scope than in the past, and the increasing trends of partnership
participation, outsourcing processes, de-localization of productions, etc. force organizations to
accurately manage the total costs of their products even if an increasing amount of items do not rely
upon their core competencies. Therefore, specialized expertises and competencies are necessary to
effectively manage costs, and new practitioners and consultant firms are offering their specialized
knowledge and skills.
Industrial Context
In this work, a paradigmatic case study is described. Creactive Consulting S.p.A. is an Italian
consulting company in cost management for medium and large firms (Creactive, 2005). Established
in year 2000, now Creactive Consulting S.p.A. is specialised in offering cost management services
such as: expense reduction projects for a specific cost area (e.g. logistics, tools), projects for one
specific expense category (e.g. express delivery) or special jobs for critical areas. In year 2004
Creactive Consulting S.p.A. has set up an effectiveness partnership with ACP (an IT company) to
develop semantic based technologies and tools aimed at supporting cost management processes,
managing catalogues, and comparing products and services.
The consultancy firm, composed by 6 partners and 40 experts in cost management, has carried out
many projects of costs reduction for over 100 different clients, both public organizations and private
companies which belong to Italian and international firms. Some of them are: Telecom, Pirelli,
Veronesi Group, Automotive Lighting, Banca Intesa, Siemens, Nestle, etc. At the beginning the
firm managed a huge number of projects for small and medium enterprises (almost 50 projects in a
year), now it focuses on large organizations and the number of contemporary managed projects was
progressively reduced to 10. These projects are long term activities, have a higher level of
complexity that requires stronger attention to human and technical resources allocation.
Flexibility and concrete approach, together with a steady activity to develop new solutions have
allowed Creactive Consulting S.p.A to operate in any area and on any category of cost: from the
easier general expenses to the more complex dynamics of core business costs. The keystone to
develop successful projects (with almost 15% of savings) is the correct knowledge of customer's
costs structure such as the series of products and services that are consumed and purchased within
the firm and their functional and technical characteristics. To achieve an effective expenses
reduction the company analyses the market, and proposes innovative solutions to the client,
maintaining or improving the level of service. Due to the expertise of qualified consultants and
partners, new solutions are unveiled and shared with the client.
In the 95% of its projects, the firm has guaranteed a media of 15% of savings, in particular it has
obtained the following percentages of savings on products and services: 26% on clothing, 15% on
insurance policies, 22% on paper and forms, 27% on hardware and software, 32% on copiers
machines, 30% on mailing services, 29% on telecommunication services, 25% on waste
management services, 16% on cleaning and logistic services, etc.
After that Creactive Consulting S.p.A. implements new solutions within clients’firms, consolidates
the services and supervises the achieved results in order to make the company completely able of
self-governing its expenses. The success of Creactive Consulting S.p.A. is sustained by its system
of reward (based on success fees on achieved savings), the active participation in multidivisional
teams, and the ability of consultants to understand the corporate cost structure of clients (spending
198
data models, demand analysis, strategy compliance, etc.), the capacity to find feasible deviations in
the cost structure of products and services, and the ability to propose and then test innovative
solutions (Creactive, 2005). All these features will be described in the following paragraphs.
Problem
The client’s cost reduction process is developed in a relational context that doesn’t fall anymore
into the simple dichotomy of markets and hierarchy (Williamson, 1975), but generates real complex
systems of negotiation and product/knowledge sharing. Instead, these relationships might represent
intermediate or hybrid forms of governance that enable firms to access the knowledge of their
trading partners in more efficient ways, and act on it. These hybrid relational contexts are
characterized by incomplete contracting since it is either impossible or unreasonable to develop
standard agreements that completely specify all of the probable interactions’outcomes (Cooper and
Slagmulder, 2002, 2002b). This causes a more complex system of cost management which should
take into account a set of knowledge that is buyer’s or supplier’s property, such as technologies,
functionalities, and qualities of products, semi-processed products, and services. Despite the
increased interest in interdependencies and information flows that transcend organizational
boundaries and their potential implication for management accounting, the topic has largely been
ignored in the accounting research literature (Cooper and Slagmulder, 2002, 2002b). In academy,
the debate on the relevance and future of Cost Accounting and Management highlighted by the
famous Johnson and Kaplan’s book (1997), has driven much attention towards the development of
the so called “new cost management techniques and approaches”.
This complexity increases the need of specialized expertise on products and cost management
methods, tools and organizational processes and encourages the proliferation of consultants and
experts in cost reduction and management that often are not employed in one unique organization.
In other words, it generates the proliferation of outsourcing processes of costs management which
increase the need for consultancy firms specialized in different kinds of cost management
methodologies, technologies, products and services. These experts collaborate with a growing
number of firms, have strong networks with producers and vendors, know innovative strategies of
negotiation processes, are aware of new production technologies, and know the preferences of
vendors and buyers. As a result, the role of knowledge becomes a very important matter and at the
same time a problem to overcome. Knowledge asymmetry arises among different actors (such as
vendors and buyers) because it is mainly caused by bounded rationality (for in depth analysis see
the work of James March and Herbert Simon (1958)) and by the information asymmetry between
organizations (the outsource) and consultants (outsourcers). At the moment consultants do not have
the same information that workers within organizations (i.e. purchasing officers, managers,
practitioner) have, and cannot autonomously decide on the selection of components and services
required by the firm. In other words consultants (outsourcers), together with organization’s experts
(outsource), have to think, analyze, validate and combine the most favourable set of semi-processed
products and services that minimize their costs. This continuous interdependency of knowledge
among external experts and purchasing officers generates a new need. Methodologies, tools and
processes of cost management should be analyzed from an innovative approach that should consider
knowledge as the discrete variable.
Therefore the aim of this work is to study some of the cost management practices through an
innovative approach (the sense-making approach deeply analyzed by Weick (1995)), stressing the
continuous interdependencies of knowledge among consultants, vendors and buyers.
199
•
The market complexity increases the need of specialized expertise on products,
cost management methods, tools and organizational processes.
•
The market scenario is characterized by the proliferation of outsourcing
processes of cost management which stress the need for emerging consultancy
firms specialized in different kinds of knowledge on cost management
methodologies, technologies, products and services.
•
Cost management consultants develop a new strategy, becoming knowledge
brokers among producers and vendors and conducting innovation among them.
•
As a result knowledge of both workers and organizations arises its strategic
importance and it becomes necessary to develop new cost management
methodologies based on knowledge.
Approach
Few theories should be mentioned in this work, with the aim of studying some outsourcing practices
of costs management. The methodology that has been used is ethnography and in particular
ethnographic interviews that have been used in order to understand what kind of knowledge is
important in consultancy firms, and in reduction cost processes. The analysis attempts to study how
knowledge can help consultant in sustaining the continuous interdependencies of knowledge among
vendors, buyers, and them, according to the sense-making approach. This new approach is based on
constructivism theories, which refer to the cognitive and social processes of developing a common
or shared understanding (Weick, 1995; Weick and Roberts, 1993; Daft and Weick, 1983). This
framework will be analyzed and discussed in this paragraph. Finally some important methodologies
of cost reduction will be described in order to understand how the cost management processes occur
among Creactive Consulting S.p.A. and other organizations.
The Ethnographic Methodology
There are a lot of methodologies that stress the social relations within firms, such as social network
analysis (SNA) (Wasserman et al., 1994), ethnography (Orlikowski and Gash, 1994), Tropos
(Castro et al., 2002), etc. Some of them (i.e. ethnography, ethnographic interviews) adopt an
anthropological point of view, using a very rigorous approach of participating in the daily activities
in order to understand the organizational cultures as a whole. Observers usually get a thorough
understanding of the circumstances of the events that are being studied, and determine the
significance of what they observe without gathering broad statistical information. The underling
theories of ethnographic methodologies refer to the symbolic interactionism which considers
meanings as handled in, and modified through, an interpretive process used by the person dealing
with the things she/he encounters (Blumer, 1962). In other words the organizational culture affects
the attitude of practitioners, and researchers might interpret these behaviors according to different
and incompatible interpretation schemas. Thus researchers should spend a lot of time with
practitioners, becoming members of the firm. Observers will understand the culture, the common
framework that people share, with the aim of giving the right interpretation on the observed events.
In this study, the strong collaboration among researchers and practitioners, allows us to participate
in the daily work, understanding the practitioners’points of view, beliefs, and frameworks. In order
to obtain more significant results, and more in depth analysis of specific procedures, we decided to
carry on a series of questionnaires and long interviews. Due to the high level of confidence among
researchers and interviewed, the style of interviews has been completely free, allowing us to discuss
significant open issues.
200
The Constructivism Approach
According to the constructivism approach (for in depth analysis see (Berger and Luckmann, 1966;
Weick, 1979)) members of a group give precise meanings to their activities and roles. These
meanings are strongly influenced by (and at the same time manipulate) their actions and
environment. In this scenario the sense-making process is defined as an interpretive action in which
people assign meanings to ongoing occurrences (Weick, 1995) through three main elements:
•
a framework: the system of mental constructions that individuals use interpreting objects or
reality. In a collective sense, it is the system of values, routines, stories that are embedded in
the organizational culture;
•
an issue: an object that only if connected with the framework has a meaning;
•
a cognitive shock: a gradual or radical interruption of the continuous flow of issues
interpretations. This interruption is caused by a significant deviation between expectations
an interpreted issues. Namely, individuals or groups recognize issues and experiences but
are not able to interpret them in a sound way as long as framework changes take place.
According to Weick, to make sense means to create order and understanding among experiences by
applying mental frameworks. The process of connecting a framework to an issue, which has been
introduced into the mind or into an organizational environment, is the very core of making sense.
Besides, routines and well established practices correspond to a specific framework, and only a
cognitive shock allows us to change them.
Some Background on Cost Reduction and Cost Management
The traditional volume-driver cost system does not fulfil adequately new emerging needs such as:
performance measurement, operational control or product costing purposes (Kaplan, 1990),
examination of relationships between a firm’s strategy and the design of information systems, cost
control and cost management (Yoshikawa, Tanaka, Innes e Mitchell, 1993; Horngren, Foster e
Datar, 1998), etc. Under the pressure of the emerging competition, accounting has moved to cost
management and has become an integral part of the planning and control system of manufacturing
operations (Brimson, 1991). Furthermore, under this perspective, cost management can be seen as a
set of techniques and procedures that support decision making (Kaplan, 1990), and overall, as an
“attitude and a way of life”that takes into consideration market strategies, value of products, etc.
(Cooper, Slagmulder and Drucker, 1999). Propelled by the search for solutions to the problem of
the inadequacy of “traditional cost systems”, researchers and practitioners have accepted the idea
that, in order to comply with a more complex set of needs, a wide set of possible solutions has to be
selected. Therefore, various, and sometimes not well coordinated, approaches and techniques have
been designed and proposed, taking a great deal of inspiration from “Japanes Cost Management
approaches”. The increasing complexity of cost management can be revealed by the analysis of its
methodologies and techniques and the increasing interest in processes and practices that sustain
effective processes of costs reduction.
For in depth analysis of these different techniques, see the Activity Based Costing (ABC) that
supports decisions making on products, analyzing the product life cycle according to Activity Based
Management and Activity Based Management Accounting (Turney, 1991; Brimson, 1991) the Life
Cycle Costing and the Target Costing (Berliner e Brimson, 1988; Susman, 1989) that aim to
enhance value and reduce costs of products throughout the product life-cycle and highlight
opportunities for cost reduction (Susman, 1989); the Target Costing, in which a target price and a
target profit should be achieved by a combination of products and their components (Sakurai, 1989;
Cooper, 2002); the Total Cost of Ownership that takes into consideration the overall costs of the
supply chain, analyses processes within the firm and also the relationships among supplier and
201
customers in terms of control and management costs (Ellram, 1995; 2002).
At the same time, various tools –such as Value Engineering, Variety Reduction Program, Cost
Table, Quality Function Deployment, Design For Manufacturing and Assembly (Monden and
Sakurai, 1989; Akao, 1989; Yoshikawa and Innes, Mitchell,1989; Yoshikawa, Innes and Mitchell,
1990; Koudate and Suzue, 1990; Gerthardt, Hutchinson and Mistry, 1991)– have been used to
define the most acceptable combination of components which sustain a cost reduction (Cooper and
Slagmulder, 2002; 2002b; Monden and Sakurai, 1989). All the combinations of products, semiprocessed products and services, should satisfy the functional requirements explicitly or implicitly
needed by the firm and its departments, even if the combinations of various components could
change products functionalities.
Finally, a lot of studies and works have been devoted to the creation of some organizational norms,
such as the active participation of purchasing officers and vendors, the participation of members
who operate in different organizational divisions (marketing, production, administration), the
creation of multidisciplinary teams, etc. In some studies it has emerged that the continuous
interaction between experts of different functions and disciplines helps to generate innovative
solutions; the vertical interactive control between top management and middle management
supplements the top management’s monitoring activities and helps to formulate their strategy; the
continuous interplay of multiple “local technologies”, and the interactions at the boundaries
between different units are critical factors for innovation and for creation of new knowledge (Brown
and Duguid, 1991); the encounter of different perspectives sustains innovation (Boland and
Tenkasi, 1995), etc. However, to achieve good and innovative results, it is not sufficient to organize
multidisciplinary meetings and inter-functionality teams since knowledge, values and trust are not
shared.
Case Analysis
In cost management processes, knowledge concerning strategies, products, functionalities, should
be trustfully shared, in a multidisciplinary and cross functional team. Even if the team members
work using different tools – such as Value Engineering, Variety Reduction Program, cost table,
Quality Function Deployment, Design For Manufacturing and Assembly (Monden and Sakurai,
1989; Akao, 1989; Yoshikawa and Innes, Mitchell, 1989;1990; Koudate and Suzue, 1990;
Gerthardt, Hutchinson and Mistry, 1991) – people need to share perspectives, create a common
agreement and an understanding of promising combinations of innovative products. These
significant issues are even more important if the cost management processes are in outsourcing. In
this case clients and outsourcers should share knowledge and strategic vision on products and
services functionalities and characteristics. Therefore, a new knowledge based approach aiming at
analyzing these processes should be used (Collini, Cuel, and Fabrello, 2005).
According to constructivism approach and in particular to the sense-making process, our
ethnographic analysis (which has been developed through daily participation, and qualitative
interviews) allows us to participate in the interpretive action in which practitioners assign meanings
to ongoing occurrences (Weick, 1995). In order to clearly describe how sense-making processes
allow us to understand these interpretive actions, the typical process of cost management, and some
paradigmatic examples will be described in the following paragraphs.
A Typical Process of Cost Management
Thanks to consultants’experiences, Creactive Consulting S.p.A. has generated and maintained a
system of best practices with the aim of offering effective services. This system of routines and
processes is usually applied to both general expenses (stationery, travels, telecommunications, etc.)
and core costs (logistics, ICT services, maintenance, raw materials, etc.), and it is part of the
202
organizational value chain depicted in Figure 1.
Figure 1: Creactive Consulting S.p.A.'s Value Creation
It is important to notice that each project, carried on by Creactive Consulting S.p.A., takes few
months (from 6 to 18 months) to be completed, and requires the involvement of the customers’
knowledge, particularly in the phases described in the value chain. Forms and times of intervention
are established according to personalized purposes, in order to produce savings in a very short time,
keeping the level of service absolutely unchanged – indeed, if possible, making it better. After
sharing the best solution with the client, Creactive Consulting S.p.A. proceeds to its concrete
implementation starting then a check period of consolidation and supervision of the results achieved
and making the company completely able of self-governing its expenses.
In particular the Creactive Consulting’s Value Chain consists of these main phases:
1. Data acquisition: composed by data collection, data validation, and calculus of expense
perimeter. This phase is aimed at unveiling the products and services systems, organizations
buy during a certain period of time (usually one year). Consultants should analyze all the
purchasing processes that have been carried out, obtaining data from different sources:
databases, paper receipts, purchasing orders, interviews. Sometimes, consultants can obtain
data directly from the information systems of the client, but some other time they have to
copy (by hand) all the paper purchasing requests. After this first analysis as a result there
will be a report on the expense perimeter of the firm.
2. The Consumption model is evaluated according to knowledge and purchasing practices on
the market, and a common view on the needed product (the benchmark) is shared among
consultants and practitioners. In other words, consultants analyze the expense perimeter in
order to understand which technical and non technical functionalities are really required
within the firm, and create the consumption model. This model will contain all the
information that experts need when searching for products and services. All the functional
and non functional requirements are analyzed by experts who operate in the market, aiming
at finding innovative products and services. Besides, the system of purchasing processes is
analyzed and some purchasing best practices are unveiled. In the benchmark phase this
knowledge is shared with practitioners allowing consultant to perfectly understand
organizational needs, market constrains, purchasing practices, and some possible lacks in
products and services functionalities.
3. Once the internal benchmark is set, and appropriate product categories (often the category of
products that presents more cost reduction opportunities) are chosen, it is essential to
evaluate the maximum achievable saving comparing it with the market benchmark and the
best practice needed to achieve this goal. New solutions are proposed, shared and evaluated
by both practitioners and consultants, and only the solution that seems more compatible with
the client’s strategy is accepted.
203
4. After the approval, the solution is implemented within the firm as a normal self-governed
purchasing phase, reducing impacts on the company's efficiency. The monitoring phase is
aimed at evaluating the benefits these new solutions have brought to the clients.
In the next paragraphs two paradigmatic activities will be presented, in particular all the phases
described above will be analyzed through the sense-making framework (Collini, Cuel, Fabrello,
2005).
The Case of the Food Industry
In this first case it is described how Creactive Consulting S.p.A. operated in one of the biggest
Italian food industry. This industry is the result of several operations of acquisition and fusion that
generated a proliferation of purchase offices, each one with different culture, policies, and routines.
In the first phase, consultants collected and analyzed data from all the various purchasing divisions,
finding solutions for cost reduction in no core products area such as packing, office materials, tools,
etc. As explained above, the major cost reduction chances can be obtained only for products that
present the better saving opportunities and one of these is the packing area, especially the poultry
pack boxes. The second phase attempted to share a common understanding (framework) among
consultants and practitioners, unveiling that:
•
all packs were very expensive, because they were specially made for food conservation, but
their primary use wasn’t always for food, often they were used as secondary packaging;
•
all packs had the same dimensions and didn’t always fit with the product characteristics
(height, length, width);
•
packs had some problems: they weren’t always filled and were difficult to handle because of
the fragility of packing structures and the very shoddy material composition.
The purchasing officers had never realized these problems, and continued to buy packs in the
traditional way. Workers believed that these products had satisfied their needs for a long time and
buying the same products was a consolidated practice that in the past never changed. Thus why did
they need to change it? When consultants explained them the real situation, they started to think that
their knowledge on packs was not complete. The second phase, the internal benchmark, was
developed, but clients still didn’t legitimate new solutions. This caused the first step of cognitive
shock that changed the organizational frameworks, made acceptable an improvement in purchasing
procedures and products, and legitimated consultants to find new ideas and solutions. Considering
the system of organizational requirement, packing houses, and the market, consultant detected
various solutions composed by a series of products, contracts with vendors and service level
agreements (third phase). One of the more innovative solutions has been the adoption of a plastic
boxes rent system, which promised excellent percentage of cost reduction for the food industry, but
managers preferred to maintain the status quo, because the proposed solution required some
changes in the logistic system, and they were not ready to afford that. Other solutions have been
analyzed and tested by the interfunctional team composed by Creactive consultants and
organization officers, but in this particular case no kind of packing satisfied the users’requirements
and frameworks. Thus, through the active collaboration with vendors, practitioners and purchasing
officers, a new kind of packing was developed, then introduced, and finally adopted in the firm.
This is made of a more resistant and less expensive material and also with different architectural
characteristics that allow easier filling and transportation and is available in various measures
suitable with standard pallets sizes. In any case organizational purchasing officers and practitioners
didn’t adopt immediately the new solutions. They accepted to introduce these new ideas only when,
in one special meeting, the CEO wanted to test the resistance of the packs. He is a very big man,
and jumped on the packs that traditionally are bought by the firm. The packs broke. Then he jumped
204
on the new packs, and they didn’t break. After that, the CEO was able to persuade all the
purchasing officers and practitioners to trust consultants, and buy the new packs. After this event,
consultants have been legitimated by the CEO, allowing a more open mind attitude of practitioners.
The Case Study of an Automobile Firm
As the previous case, the cost reduction issue of this case study concerned packing solutions for
headlights and taillights. These are very fragile parts and require very expensive and high quality
packing and secondary packaging. Creactive Consulting S.p.A. was asked to find on the market, or
to study ex-novo, a new type of packing that could guarantee high protection at lower costs. From a
market analysis Creactive discovered that a reasonable number of suppliers already produced
suitable packing systems; the problem was to compare these systems because vendors didn’t
provide information on functionalities and requirements that allow products evaluation, and added
every time additional special costs (second phase). Therefore consultants of Creactive Consulting
S.p.A. decided to create an interdisciplinary and interfunctional team, composed by purchasing
officers, shipping agents, producers, and vendors. Through continuous trustful interactions, the team
found innovative solutions and a new taillight packing system has been proposed (third phase). This
presented the same, or even better, characteristics of protection, with a production process that is
less complex and expensive then the other packing systems proposed by traditional suppliers. This
new packing has been introduced only after very hard suitability tests which have been carried on in
order to obtain a design patent. After that the commission gave the patent, and provided an
objective and trustable judgment of the product, officers convinced themselves that these new
taillight packing were the best solution, and now they use these within the automobile firm (fourth
phase).
Key Findings
The two case studies depicted in the above paragraphs can be analyzed through different traditional
perspectives. The cases can be studied according to the cost methodologies that have been used, in
particular the Total Cost of Ownership or the Target Cost methodologies in which functional and
non functional requirements are defined a priori, and a target cost should be achieved through
various combinations of components, services and semi-processed products. Another kind of
analysis could be focused on tools and instruments (using for instance Value Engineering, Variety
Reduction Programs systems) that have been used to design innovative products. Finally a more
organizational analysis could study the relevance of having multifunctional and multidivisional
teams. Although, these analysis explain many important aspects of cost management processes,
none of them can explain why and how people share perspectives, create a common agreement on
promising combination of products, adopt one solution instead of another sustaining innovation. As
a result a knowledge based analysis is needed; it should be aimed at understanding how people
trustfully participate, share strategic knowledge, achieve a common goal, and adopt an effective
solutions, in particular in outsourcing processes of cost management. Analyzing Creactive
Consulting’s value chain through the sense-making framework, we can obtain Figure 2.
205
Figure 2: Creactive Consulting’s value chain and the sense-making framework
Looking at the cases of food and automobile industries, the active participation of consultants in the
daily activities and the revenues based on success fee, facilitate the creation of a relaxing and
informal environment in which people feel free to share knowledge and help consultants to make
understandable the outsource’s framework (mainly understood through the analysis of practices and
needs). The scientific analysis of the sets of products, components, and services purchased by the
firm, allows consultants to unveil some asymmetries between the firm needs and the functionalities
of purchased services and products. This asymmetry can be used as a real information instrument
that generates some inclinations to accept cognitive shocks and pushes people to think about what
they really need and want, and how they can achieve these goals. Particularly, in the case of food
industry, people didn’t recognize that the packs were never fully laden and difficult to handle even
though consultants provided a cognitive shock. The cognitive shock can be caused by the unveiling
of non interpretable information or by some lacks in the framework interpretation, that makes
people uncomfortable with their system of meanings. These status of feeling push people to think
about their practices, often giving a new meaning to some of the activities they automatically carry
on. But only a very strong event brings changes in the framework, and encourages the acceptance of
innovative solutions. In the examples above mentioned the change happened, in the first case when
the CEO jumped on new type of packs and didn’t break them; in the second case when the
commission proved the quality of taillight packing.
In the particular case of Creactive Consulting S.p.A., these cognitive shocks occur because of the
consultant’s ability of being legitimated to look for new solutions, that will be discussed and tested
by the outsource. But not all the hypothesized solutions are automatically adopted in the firms. In
the case of food industry, the less expensive and more innovative solution has not been adopted, but
instead, a less convenient solution was chosen by the organization. This could be explained through
sense-making theories. The most innovative solution corresponds to an issue that is not compatible
with the decision maker’s framework. In other words the solution was too innovative, required to
many changes not only in the procedure and practices but in particular in the manager framework. It
could be that the decision maker wasn’t able to understand the great potentiality of this solution, or
wasn’t ready to radically change her/his system of meanings.
Concluding we can say that even if consultants provide some alternatives, the adopted solution will
be the one that is at the same time innovative and compatible with the decision makers’
frameworks. In the case of the automobile company the shared framework between consultants and
outsource was clear from the beginning. The principal purpose was to find out an innovative
206
product, that satisfied users, and had lower costs. Therefore a multidisciplinary team was settled up
with the objective of creating innovative solutions. When the new taillight packing system was
proposed, puzzlement was diffuse among members of the team. The new solution didn’t fit
immediately with the members’frameworks, and a cognitive shock was needed. The qualitative
tests on the product represented the cognitive shock, in fact after that, members literally changed
their mind, and started to ask for the new taillight packing within the firm.
In the analysis, this theory has been applied to the Creactive Consulting S.p.A.’s value Chain, and
consultant practices have been studied according to the sense-making processes. In particular during
meetings and focus groups participants (experts and consultants) are stimulated to perceive the real
situation within the firm (combination of products, strategies, contracts and service level
agreements (SLA), etc.), confirm their framework or make new meanings (understand and change
the framework), help members in changing their perceptions, unveil problems in product
combinations and SLA, etc. In this context, and in particular in processes of cost management
outsourcing, different frameworks (the one of consultants and the other of purchasing officers)
meet. This generates cognitive shocks which are explored and adapted to new interpretations, thus
frameworks change and innovation is sustained (Kezar and Eckel, 2002).
Business Impacts
In this paper the processes of reduction costs and the Creactive’s value chain have been analyzed
according to the sense-making approach. As explained above sense-making is the reciprocal process
where people seek information, assign it some meaning, act (Weick, 1995), and collectively give
sense to uncertain and ambiguous organizational situations (Weick and Roberts, 1993). These
sense-making processes become more important when cost reduction processes are developed in
outsourcing, different perspectives and cultures are encountered, and information asymmetries
persist. Even if in further works, other cases should be analyzed and stronger theories should be
defined; in this paper we want to highlight some main characteristics, that according to the sensemaking approach could be used as guidelines for effective processes of cost management. These
characteristics are:
•
the creation of a collaborative partnership. Outsource and outsourcers should become
accomplices that share knowledge and strategies in a cooperative way;
•
the establishment of a collaborative environment in which consultants and practitioners feel
free and trustful to share knowledge;
•
the agreement on a common initial situation in which some profitable solutions can be
developed;
•
the emerging of new meanings and solutions caused by a cognitive shock, which pushes
members of the inter-functional and interdivisional group to search for new solutions;
•
the adoption of the best suitable solutions. Even if innovative and suitable solutions are
proposed, other less convenient might be implemented within the firm, only because these
last ones are much more coherent and consistent with the decision makers’frameworks.
Conclusions
In these last decades cost management processes evolved into very complex approaches, which take
into account the organizational strategy of the firm, the knowledge and expertise of vendors, the
ability of brokers in negotiation processes, etc. This complexity arises in particular when the
product value chain depends on a huge number of specialized subjects which cooperate through
outsourcing processes, join ventures, and other kind of networked relationships. It derives that cost
207
management approaches, processes, and instruments should explicitly adopt a new perspective
which considers knowledge as a valuable matter. According to the constructivism approach,
knowledge is a social construction affected by organizational cultures, routines and procedures that
practitioners follow in their daily activity, etc. Thus in cost management approaches all these
aspects have to be considered.
One of the main focuses of this research is to identify some new processes and tools that could
easily allow practitioners and cost management consultant to deal with this kind of knowledge.
Through the analysis of a consolidated practice (the value chain of Creactive Consulting) we have
unveiled some main critical issues and variables that should be taken into account in cost
management processes. In particular, it emerged that consultants have to create a friendly
environment through which knowledge can be easily exchanged (a kind of “ba”defined by Nonaka
and Konno (1998)), client’s points of view, perspectives and needs clearly understood, and finally
purchasing practices deeply shared and analyzed. This can be done adopting the sense-making
framework based on the following phases: framework sharing, issue creation, and cognitive shocks.
An important result of our investigation, is to consider cost management consultancy firms, as
brokers, namely superpartes subjects who allow the convergence of both clients’and vendors’
needs. The consultants’expertise is based on knowledge about the client’s consumption model,
purchasing processes, the network of specialized vendors and their production technologies. In
some cases, vendor’s technology cannot satisfy the client’s needs, thus the encountering of clients’
and vendors’ needs should be facilitated in order to obtain innovative solutions. The role of
facilitator is carried on by the consultancy firms, and due to the consultant’s ability to enable free
discussion on open issues, innovative solution are made possible.
208
•
Sense-making framework provides a useful point of view particularly when cost
reduction processes are developed in outsourcing, perspectives and cultures are
very different, and information asymmetries persist. This has two different
impacts on both research and industry. Research should take into consideration
that decision processes on cost management are based on four phases:
framework sharing, issue creation, cognitive shock, and approval of final
solution. Industry should develop decision processes and best practices based
on these 4 steps, and should consider differences on cultures and perspectives
as strategic values instead of problems to overcome.
•
Knowledge sharing among practitioners and consultant is a very important
matter and can be done through trustful and collaborative partnerships. Thus
vendors and clients have to share a common (physical, virtual or mental)
space. In order to build a trustful collaborative environment, they have to
periodically share the same space (meet in conferences, meetings, etc.), share
communication channels (adopt common tools to share knowledge on
purchased product and services), compare their knowledge on functional and
technical product and services characteristics, etc. In other words, vendors
should be allowed to understand and interpret the emerging clients‘needs and,
at the same time, clients should be allowed to see how new products and
services can be adopted and used for their business.
•
New solutions can be developed thanks to the convergence of vendors and
clients’perspectives and cultures. Consultants, as brokers, allow the
encountering of various perspectives by stressing on cognitive shocks and
forcing people to look for innovative solutions. Therefore the role of a broker
becomes a strategic issue to study and apply in practice. Researchers should
better analyze the role of brokers among organizations, and in particular their
main competences and abilities in facilitating knowledge exchange.
Organizations should consider investments in a periodical analysis of
consumption models, vendor portfolios, and negotiation processes as very
strategic for their competitiveness on the market. For instance, it would allow
them to be periodically aware of how the market changes, what innovative
products and services are sold, and what purchasing strategy should be
adopted.
Acknowledgements
The author gratefully thanks Roberto Biscaro for actively contributing to this research activity, and Creactive
Consulting S.p.A. for funding the project OASI (Ontologies of Artifacts and Services to Industry). A special
thanks to Prof. Paolo Collini who helped me analyzing data and improving my results, and last but not least a
credit to Linda Fabrello who did the “dirty”work.
References
Akao, Y. (1989) “Quality Function Deployment for Product Development”, Union of Japanese
Engineers and Scientists, Tokyo.
Ashby, W.R., (1956) An Introduction to Cybernetics, John Wiley & Sons, New York.
Berger, P.L., and Luckmann, T. (1966) The social construction of reality New York, NY: Anchor
Books.
Berliner, C, and Brimson, J.A. (1988) Cost Management for Today’s Manufacturing: the CAM-I
209
Conceptual Design, Harvard Business School Press Boston.
Blumer, H. (1969) Symbolic Interactionism: Perspective and Method. Los Angeles, CA: University
of California Press.
Boland, J.R., and Tenkasi, R.V. (1995) “Perspective Making And Perspective Taking In CommuniTies Of Knowing”; Organization Science, Vol.6.4, pp.350-372.
Brimson, J. (1991) Activity Accounting. An Activity-Based Costing Approach, Wiley/NAA
Professional Book Series, New York.
Brown, J.S., and Duguid, P. (1991) “Organizational Learning And Communities-Of-Practice:
Toward A Unified View Of Working, Learning And Innovation”; Organization Science,
Vol.2.
Castro, J., Kolp, M., and Mylopoulos, J. (2002). Towards Requirements-Driven Information
Systems Engineering: the TROPOS Project. Information Systems.
Chandler, A.D. (1962). Strategy and structure: chapters in the history of the industrial enterprise.
MIT Press, Cambridge.
Collini P., Cuel R., Fabrello L. (2005) "Sense-making in target costing". Proceeding of “7th
Manufacturing Accounting Research Conference”, Tampere, Finland, 30th May-1st June.
Cooper, R. (2002) “Target Costing for New-Product Development”, in Journal of cost management,
Vol.16, No.3, May/June.
Cooper, R., Slagmulder, R., and Drucker P.F. (1999) “Develop Profitable New Products with
Target Costing”, in Sloan Management Review, Vol.40, No.4.
Cooper, R., and Slagmulder, R. (2002) “Target Costing for New-Product Development: ProductLevel Target Costing”, in Journal of cost management, Vol.16, No.4, July/August;
Cooper, R., and Slagmulder, R. (2002b) “Target Costing for New-Product Development:
Component-Level Target Costing”, in Journal of cost management, Vol.16, No.5,
September/October.
Creactive (2005) Creactive Consulting S.p.A. web site, http://www.creactive-consulting.com/ (date
accessed: 15 September, 2005).
Domberger, S. (1998) The Contracting Organization, A strategic guide to Outsourcing, Oxford
Univerity, Oxford.
Daft, R.L., and Weick, K.E. (1984) “Toward a Model of Organizations as Interpretation Systems”,
in Academy of Management Review, Vol.9, No.2, pp284-295.
Ellram, L.M. (1995) “Total Cost of Ownership, An analysis approach for purchasing”, in
International Journal of Physical Distribution & Logistics Management, Vol.25, No.8.
Ellram, L.M. (2002) “Target Costing and Supply Chain”, in European Journal of Purchasing and
Supply Management, Vol.8, Issue 4, pp.235-244, December.
Gerthardt, D., Hutchinson, W., and Mistry, D. (1991) “Design for Manufacture and Assembly”, in
the International Journal of Advanced Manufacturing Technology, Vol. 6.
Hamel, C., & Prahalad, G. (1990) The core competences of the corporation, Harvard Business
Review, May-June.
Horngren, C.T., Foster, G., and Datar, S. (1998) Cost Accounting – A Managerial Emphasis, 9th
Ed., New York.
Kaplan, R. (1990) Measures for Manufacturing Excellence, Harvard Business School Press.
Kezar, A., and Eckel, P. (2002) “Examining the Institutional Transformation Process: the
210
importance of Sense-making interrelated Strategies and Balance”, in Research in Higher
Education, Vol. 43, No.3, June.
Koudate, A., and Suzue, T. (1990) Variety Reduction Program, Productivity Press.
March, J., and Simon, H. (1958) Organizations, Wiley, New York.
Maturana H.R. & Varela F.J. (1980) Autopoiesis and Cognition: The Realization of the Living
Dordrecht: D. Reidel.
Monden, Y., and Sakurai, M. (1989) “Japanese Management Accounting: A World Class Approach
to Profit Management”, Productivity Press, Cambridge (Mass).
Nonaka, I., Konno, K. (1998) “The Concept of "Ba": Bulding a Foundation for Knowledge
Creation”. California Management Review, vol.40 no.3. Spring.
Numagami, T., Ohta, T. & Nonaka, I. (1989) Self-renewal of corporate organizations: equilibrium,
self-sustaining, and self-renewing models. Working paper, University of California at
Berkeley, No. OBIR-43.
Orlikowski, W. J., and Gash, D. C. (1994). Technological Frames: Making Sense of Information
Technology in Organization. ACM Transactions on Information Systems, Vol.12, No.2,
pp.174–207.
Sakurai, M. (1989) “Target Costing and How to Use it”, in Journal of Cost Management for the
Manufacturing Industry, Vol.3, No.2, Summer.
Susman, G.I. (1989) “Product Life Cycle Management”, in Journal of Cost Management for the
Manufacturing Industry, Summer .
Turney, P.B.B. (1991) “How Activity-Based Costing Helps Reduce Costs”, in Journal of Cost
Management for the Manufacturing Industry, Winter.
Weick, K.E. (1979) The Social Psychology Of Organizing. Mcgraw-Hill Inc.
Weick, K.E. (1995) Sense-making in Organizations, Sage Publications.
Weick, K.E., and Roberts, K.H. (1993) “Collective mind in Organizations: Heedful interrelating on
Flight Decks”, Administrative Science Quarterly, Vol. 38 pp. 357-381.
Williamson, O.E. (1975) Markets And Hierarchies: Analysis And Antitrust Implications, The Free
Press, New York.
Wasserman, S., Faust, K., and Iacobucci, D. (1994). Social Network Analysis : Methods and
Applications. Cambridge University Press.
Yoshikawa, T., Innes, J., and Mitchell, F. (1989) “Cost Management through Functional Analysis”,
in Journal of Cost Management for the Manufacturing Industry, Summer.
Yoshikawa, T., Innes, J., and Mitchell, F. (1990) “Cost Tables: A Foundation of Cost Management
in Japan”, in Journal of Cost Management for the Manufacturing Industry, June.
Yoshikawa, T., Innes, J., and Mitchell, F. (1993) Contemporary Cost Management, Cima, London.
211
Author Biography
Roberta Cuel has obtained a Phd in Organization and Management of the Firms
in 2003, the title of the dissertation was: “Trade off between organization and
technology in the knowledge management. A case Study”. Her main interests are:
the impact of innovative technologies on teams, communities, and organizational
models; the study of distributed tools and processes that allow organizational
learning and knowledge management; knowledge representation systems and
tools (such as ontologies, classifications, taxonomies) as mechanisms for
knowledge reification processes. She works within various projects, two of them
are:
INCONSPES
(http://fandango.cs.unitn.it/net-economy/InConSPES/),
managed by the University of Trento and KnowledgeWeb, a FP6 Network of
Excellence that aims to support the transition of Ontology technology from
Academia to Industry (http://knowledgeweb.semanticweb.org/). She teaches a
course on “Marketing and E-commerce”at the Faculty of Economics and one on
“Science Technology and Entrepreneurship” at the Faculty of Science of the
University of Trento.
212
12
Supporting Mathematical
Modelling Management
by Guiding and
Monitoring
Huub Scholten
Adrie J.M. Beulens
Supporting Mathematical Modelling
Management by Guiding and
Monitoring
Huub Scholten, Wageningen University ([email protected])
Adrie J.M. Beulens, Wageningen University
Abstract
Multidisciplinary, model-based problem solving for water management
is a complex process. Projects that have to follow this process may
encounter many problems, related to miscommunication, malpractice,
misuse of models, insufficient knowledge of the problems modelled
and overselling of model capabilities. This leads to modelling projects,
which are not transparent and difficult to audit.
The knowledge based system (KBS) discussed in this chapter
provides guidelines on what to do in such projects, derived from an
ontological knowledge base (KB), which contains state-of-the-art
knowledge on ‘
best modelling practices’ for teams of which its
members have different disciplinary backgrounds and play different
roles in a project. Furthermore, the KBS monitors what each team
member actually does and helps generating project reports for various
audiences and purposes. The last component of the KBS consists of
multimedia training material helping novice users to find their way in
the KBS.
The KBS (knowledge base and support tool) has been developed in
the context of a European Research and Technology project partly
funded by the European Commission.
The developers of this KBS learned that arriving at consensus on a
process KB (here for model-based water management) has an
intrinsic value. Professional users tested the KBS in two series of ten
test cases. They learned to like it and their inputs and comments have
significantly improved the KBS. But they also expect that it will be hard
to persuade large parts of the professional modelling community to
using it in their daily practice. Students found it complex and
cumbersome, but the quality of their modelling work increased.
Keywords: mathematical model, multidisciplinary knowledge, ontological knowledge base, model
based decision support, water management.
Background
Context
In many management areas, including water management, model based decision support systems
are and have been used by managers to help them to solve decision making problems. During the
last decades the scope and nature of the problems that managers are confronted with have changed
substantially. More and more an integrated approach taking into account technical, socio-economic
and environmental aspects are currently called for. Further, managers that commission modelling
215
Supporting Mathematical Modelling Management by Guiding and Monitoring
projects in which integrated and model based decision support systems are used also call for quality
assurance and transparency throughout the modelling project. These enhanced requirements have
partly been caused by problems encountered with model based decision support in the recent past.
The HarmoniQuA project, in which the KBS has been developed, was partly funded by the
European Union. The project had 12 partners in 10 countries, each with their own language and
practices.
Problem
The first problem that a project manager is confronted with in a normal project context is to arrive
at a shared vision on the nature and extent of a modelling project to support in finding solutions to a
stated management problem in the domain of water management. That vision entails amongst others
the scope of the study, the solution approach, expected results, duration, costs and resources used.
Thereafter, for a commissioned project the problem is to execute it in compliance with
specifications agreed upon with associated quality assurance. One part of that quality assurance is
ensured by transparency of activities executed thus guaranteeing that the work is auditable and
reproducible (Beulens and Scholten, 2004, 2005).
These current requirements are caused and fuelled by a multitude of problems and bad experiences
with model based studies in the past. Scholten et al. and Refsgaard et al. give a multitude of reasons
for these problems, including ambiguous terminology, a lack of mutual understanding between keyplayers, malpractice in regard to input data, inadequate model set-up, insufficient
calibration/validation, model use outside of its scope, insufficient knowledge on some processes,
miscommunication of the modeller to the end-user, overselling of model capabilities, confusion on
how to use model results in decision making and a lack of documentation and transparency of the
modelling process.
The responses of the modelling community to cope with these problems consist mainly of
guidelines, but these are usually nationally based and focused on single domains/disciplines
(BDMF, 2000, Middlemis, 2000, Van Waveren et al., 1999, Scholten, 1999, 2000, Scholten et al.,
2000, 2001). Resulting model outcomes and decisions based on them are often still non-transparent,
irreproducible, non-auditable and not fully comparable among different countries.
An additional complicating factor is related to the changing character of model-based problem
projects from monodisciplinary, single person and academic oriented research model studies into
multidisciplinary, decision support oriented projects, in which teams consisting of members with
different background and different roles have to co-operate to complete the complex job. Modelling
in multidisciplinary modelling teams facilitates exploring more complex questions, but it also
makes co-operation in teams more difficult. Team members with different scientific backgrounds
encounter more communication problems, which makes managing multidisciplinary model-based
water management projects a cumbersome affair.
The HarmoniQuA project aimed at lowering many of the hurdles encountered in present simulation
modelling by providing modelling guidelines, structured in a knowledge base and by developing a
tool to support projects that use models for problem solving. The context of the HarmoniQuA
project with European Commission funding, 12 partners in 10 countries and 10 different languages,
introduced new problems. Partly these were associated with language issues, which were beyond
the scope of the project as professional modellers were assumed to understand sufficiently English,
and partly because of the variety of modelling cultures in various countries ranging from very
professional and mature in northwest Europe to a novel and ad hoc state-of-the-art in some central
and south European countries. These discrepancies in expertise were further enhance by the level of
maturity of the scientific disciplines behind the water domain models, ranging from very mature for
groundwater modelling to immature for ecological and socio-economic models.
216
•
Multidisciplinary projects are instances of processes in which multidisciplinary
teams have to cooperate to achieve a result effectively, efficiently, with quality
assurance and transparency.
•
Successful execution and completion of multidisciplinary projects can be
supported by a knowledge-based system (KBS).
•
Throughout the lifecycle of a project team members get guidance on what they
have to do customized for their disciplinary background and their role in a
project.
•
What team members actually do is monitored by the KBS, which makes the
process transparent and enables auditing.
•
The KBS helps teams to generate customized reports for various audiences and
purposes.
•
The KBS facilitates communication between team members and helps them to
report what has been done, formatted for various purposes and audiences.
•
For the case study in this chapter the knowledge base is filled with knowledge
on model based problem solving in the context of water management.
•
The knowledge-based technology (knowledge base structure and support tool)
is re-useable for other types of mathematical modelling and other complex
cooperation processes.
Approach
To support and facilitate the work of multidisciplinary teams and its project manager in modelbased water management the following approach has been followed. The mathematical modelling
process has been described and decomposed in elements. That means that a generic model of a
modelling process has been developed. Based on this decomposition and experiences with other
processes, an ontological knowledge base (KB) has been designed with levels of increasing
specialization i.e. from rather abstract descriptions of process models to a description of modeling
processes for water management in all details.
This KB and its editor are the backbone of the knowledge-based system (KBS), which further
consists of the Modeling Support Tool (MoST) and training material. MoST helps a project
manager for a specific project to filter relevant pieces of knowledge from the KB in a structured
manner, depending on the problem at hand to arrive at the specification of the work to be done, by
whom (experts with differing disciplinary background) and using what (model and tool) resources.
Further, in the execution phase of the project, MoST monitors, what all team members do, and helps
generating ‘smart’reports for various audiences and purposes. Finally, the training material has
been developed to shorten the learning curve of the KBS for novice users. This training material
consists of written material, presentations, exercises that encourage using MoST in a training test
case project and many screen-recording movies on MoST, its knowledge base, a case study and
some background information. The movies are the core part of the training material and aim at
helping users to work with MoST and act as a sort of animated help facility.
The complexity of the KBS (KB, support tool, training material) sets high demands to the
development process. Therefore, a three stage building process has been followed, interrupted by
two extensive functional test periods, in which the KBS has been developed and tested.
217
Case Analysis
Knowledge Base
Development process
In the previous section we introduced the ontological knowledge base as a core element of MoST.
In this section we deal with the development in phases of that knowledge base. In this context we
use a combination of the definitions of an ontology as given by Gruber (1993, 1995) and Borst
(1997) leading to an operational definition of an ontology also emphasizing the negotiation aspects:
an ontology is a formal specification of a shared conceptualization, referring to what can be
represented in terms of concepts and the relationships among them.
A small team consisting of persons responsible for the knowledge base (KB) has decomposed
modelling processes into concepts and relations between these concepts in increasing levels of
specialization (level 0: a bootstrapping terminology, level 1: a generic process ontology, level 2: a
modelling process instance and level 3: specialized for water management). Subsequently, this
decomposition has been used to design and implement that ontology in Protégé2000 with modelling
knowledge obtained from the modelling experts in the team as instances of this ontology, together
forming the knowledge base. Details of modelling knowledge for domains of water management
have thus been uploaded to these instances. Within a larger group of professional modellers for
water management, this knowledge base has been discussed in several rounds to achieve consensus,
leading to a body of knowledge, shared by larger parts of the modelling community.
Ontological structure
An ontology can be seen as a formal framework to represent the structure and content of a body of
knowledge on a domain that can be processed by a machine. In order to describe the structure and
the content of knowledge, ontologies must have one or more standard vocabularies at hand, which
define the terms (concepts) and relations between the terms to describe this specific knowledge
domain (subject area). Concepts can be discussed and have to be represented. Concepts encompass
abstract and concrete things, processes, tasks and ambitions or goals. Concepts are used to define
and explain terms. Relations organize concepts in a hierarchical or in some self-defined structure.
Often ontologies contain other elements e.g. properties, functions, axioms, but these are not
essential to understand what ontologies are. Instances are also parts of an ontology, as they contain
the actual knowledge. If task is a concept in an ontology, the instances of task can be go shopping,
cook a meal, eat the meal. A comprehensive and clear introduction on what ontologies are and why
we need them is given in Chandrasekaran et al. (1999). In short: an ontology can be seen as a model
of organized domain knowledge and its content.
Protégé2000 (Protégé, 2000) has been chosen as tool for the development of the ontological
structure and for filling the KB with collected knowledge. The latter has thus been archived as
instances of the ontological structure. In order to improve the usability and user friendliness of
Protégé2000 in the context of an international project with many participants, who have to provide
for the knowledge to be incorporated in the KB, two extensions have been built that expand the
functionality of Protégé2000. Firstly, a plug-in for XML-export has been built, according to a
predefined XML-format, interpretable for the KB and for the support tool that has to co-operate
with the KB. It facilitates the exchange of knowledge between partners and can be automatically
processed. Secondly, a web based front-end knowledge base editor has been built that allows the
modelling experts without knowledge engineering skills to view and edit the contents of the
knowledge base (Kassahun et al., 2004), because even an easy and user-friendly tool like
Protégé2000 could form an obstacle for experts in a specific knowledge domain that are not using
218
knowledge engineering tools in their daily practice. The knowledge base editor reflects the
Protégé2000 ontology developed for the modelling knowledge base. When the ontology structure
changes, the editor adjusts itself to the new ontology carrying along knowledge items into the new
structure.
The knowledge base editor presents the knowledge in the KB as web forms. Domain experts can
add, edit and delete KB content (i.e. the instances of the ontological structure) within their own field
of expertise. The authorization mechanism of the knowledge editor minimizes conflicting updates
and errors. Finally, the knowledge base editor provides all interested individuals the possibility of
providing their comments and suggestions.
The KB distinguished three decomposition levels. At the highest level the modelling process has
been divided into steps, which are groups of tasks. To perform a task one or more activities, taking
into account the associated precedence constraints, have to be carried out. Activities are
associated with the role of a team member and are the smallest ‘doings’in the process. A task is
related to what has to be done and it refers therefore to the modelling process. By performing all its
activities in a correct manner thus producing deliverables with an appropriate quality and
documentation level a task will be completed. Each task has well defined deliverables. Steps are
logical groups of tasks and have primarily an organizational purpose for human actors involved in
the process and also have well defined deliverables.
description
name
is_a_property_of
has
step
task
property
isa
isa
has
name
definition
applicability of
method
input
output
description of
method
explanation
sensitivity and
pitfall
reference
software
aspect
has
activity
next
generic task
has
previous
is_property_of
isa
task
has
method
is_property_of
isa
feedback to
decision
has
feedback from
customization
property
has
isa
review
Figure 1. The ontological structure of the KB
(Rectangles are concepts, ovals are properties and arrows relations. The relation-type is a
stands for ‘is a … ’. The concept customization property is not specified here.)
A modelling process has been divided into 5 steps and at a lower decomposition level into 45 tasks.
Each task was further decomposed in the following task describing components: name,
definition, explanation, one or more activities (most with one or more associated methods, tools
and their applicability), references, software aspects, links to other tasks (precedence constraints)
and input and output (deliverables) of that task. In the decomposition three types of tasks are
distinguished: normal tasks, decision tasks (to decide on advancing to the next task or going back
219
to a previous one) and review tasks (i.e. special decision tasks emphasizing the negotiating
interaction between water manager (i.e. client) and modeller). Figure 1 presents the ontological
structure used for the modelling knowledge base.
Modelling experts from each of seven water management domains filled the KB by providing their
expertise to each task and indicated the relevance for user type and job complexity. The KB has
been tested in several ways, which is discussed in section ‘Testing’. Results of the tests have been
used to change the KB structure and content, which appeared to be easy because of the chosen
ontological structure of the KB and the functionality developed.
Furthermore, domain experts delivered a prototype glossary of almost 1000 entries, which has also
been included in the Protégé2000 KB. The glossary entries can be accessed from the guidelines
presented by the support tool through hyperlinks.
Guidelines for Model-based Water Management
The knowledge base is too large and complex to be discussed here in detail (see
www.HarmoniQuA.org). See for some details Refsgaard et al. (2005). In the KB the modelling
process is divided into the following steps: (1) MODEL STUDY PLAN, (2) DATA AND
CONCEPTUALIZATION, (3) MODEL SET-UP, (4) CALIBRATION AND VALIDATION and (5) SIMULATION
AND EVALUATION. Steps are decomposed into tasks. In the KB tasks belong to the work of one or
more user types, i.e. roles that team members have in a modelling project. The following user types
are included: water manager, modeller, auditor, stakeholder and concerned members of the public.
The first step (MODEL STUDY PLAN) starts with three tasks for the water manager: DESCRIBE
PROBLEM AND CONTEXT, DEFINE OBJECTIVE and IDENTIFY DATA AVAILABILITY. The water
manager also has the main responsibility for the fourth task, DETERMINE REQUIREMENTS, but here
the opinions of possible stakeholders and concerned members of the public on the overall objectives
and context of the modelling study should be taken into account before it is all compiled in the fifth
task PREPARE TERMS OF REFERENCE. In the sixth task (PROPOSAL AND TENDERING) there are roles
for the water manager, the modeller(s) and the auditor. The seventh task in this step is the decision
task AGREE ON MODEL STUDY PLAN AND BUDGET, where water manager and modeller have to
discuss how the model study has to be continued. The tasks just mentioned are concerned with the
first problem of the project manager of a project as described in the introduction of this paper, that
is to arrive at a shared and agreed upon vision on what has to be done. If parties come to an
agreement to continue, one may carry on with the execution of the project as described by the steps
numbered 2 through 5. The second step DATA AND CONCEPTUALIZATION is then the one to be
started with. That starts with the task called DESCRIBE SYSTEM AND DATA AVAILABILITY. This task
is typically a modeller’s task (Figure 2). If one decides in the decision task AGREE ON MODEL
STUDY PLAN AND BUDGET to redo some of the previous tasks, one has to return to the task DEFINE
OBJECTIVES. Figure 2 shows the first two decomposition levels of the modelling process.
After the first step MODEL STUDY PLAN the next steps consist mainly of tasks to be done by
modellers, but each step concludes with a review task, in which other user types (water manager,
auditor, stakeholder and concerned members of the public) participate.
The tasks briefly presented in figure 2 consist of activities. Part of which have simple associated
methods to used and part of which have complex methods to be used. The latter is for instance the
case for the complex task UNCERTAINTY ANALYSIS OF SIMULATION in the step SIMULATION AND
EVALUATION. This task consists of several activities to do and a number of methods to use. The
guidelines for model-based water management describe several methods for this task. The user may
select a method in order to comply with the requirements of the project. Users are not enforced to
use one of these methods provided by the guidelines, but can also use other, more appropriate
methods. The methods provided by the guidelines for this task include MONTE CARLO
SIMULATION, SENSITIVITY ANALYSIS, REGRESSION TECHNIQUES, NUSAP, EXPERT ELICITATION,
DATA UNCERTAINTY and ERROR PROPAGATION EQUATION. In addition to short descriptions of
220
Figure 2. The steps and tasks of the modelling process in MoST’s knowledge base
(from Refsgaard et al., 2004 with permission).
221
methods and their applicability the guidelines contain references to where the user can find more
thorough descriptions of the methods and supporting software tools to use the methods.
An exhaustive glossary of relevant terms completes the content of the KB. Glossary entries give
definitions with reference to the source of the definition, synonyms and – if required – additional
remarks in footnotes. The need for a consistent and explicit terminology has been discussed
extensively in Refsgaard and Henriksen (2004) and Refsgaard et al. (2005) state that the
terminology should be a part of the modelling KB and that it should be presented to guideline users
as a glossary.
Support Tool
Design
The support tool MoST (Modeling Support Tool) has been built in a prototyping process, starting
with providing just (modelling) guidelines and followed by adding functionality to monitor and
store what team members do in so called (modelling) journals. This version with a limited (still
complex) functionality was then tested in 10 test cases of normal commercial model projects. Based
on the results of these tests, the functionality of MoST has been extended to be able to assist
multidisciplinary teams co-operating online (internet) in multi-domain modelling projects for water
management. This has been tested in a second series of test cases and the latter results are used for
final improvements.
Functionality
The overall system consists of the KB previously discussed and the support tool, discussed in this
section. This system is presented in Figure 3. The support tool consists of four components, together
forming the Modeling Support Tool, MoST. These functional components (1) provide guidance
from the KB to specify what has to be done, (2) monitor what the project team members actually do
in a modelling project and store this in a model journal, (3) help users in reporting what is stored in
a model journal and (4) supply advice based on model journals of previous projects. The first three
functional components have been developed and are integrated in MoST, while the fourth –advice
–component is in the final design stage of development.
MoST will typically be used to support practical work of multidisciplinary teams in modelling
projects. Modelling projects have normally two phases, an initialization phase and a project
execution phase. In this respect MoST resembles workflow management tools, which distinguish
the same phases in business processes management (Fischer, 2003). MoST corresponds to
workflow management software in many aspects, but it differs in other aspects. Opposite to many
business processes, consisting of relatively simple tasks, which have to be executed frequently,
modelling is a process consisting of many complex tasks, requiring various types of expertise and
undergoing improvements based on scientific progress and increased computer power.
In order to facilitate working in distributed groups, the overall software product has been designed
in a client-server architecture. To some extent the KB and the tool can be considered as groupware.
Modelling teams will often work in a distributed configuration, connected by a LAN or the Internet.
A MoST-server contains all shared information (i.e. KB and the work done by a team and stored in
the project’s model journal). To speed up browsing in the KB, users will work with a local copy,
updated when necessary from MoST’s central server. The results of a multidisciplinary modelbased water management project, i.e. the work actually done, should be safely stored and available
for team members that are authorized to use the work of others. Therefore model journals are stored
at the server side. MoST-servers can be connected to users by LANs within an organization that
want to keep their multidisciplinary model-based problem solving work confidential or connected to
222
users through the Internet for organizations that want to share their work with other distributed
persons and organizations.
Knowledge Base
Model
Archive
Guidelines
Software capabilities
Glossary
Domains:
Model journal, Project A
Model journal, Project B
Model journal, Project C
Groundwater
Precipitationrunoff
Hydrodynamics
Flood forecasting
Water quality
Biota (ecology)
Socio-economics
Model journal, Project D
MoST
Guidance
Advise
Reporting
Monitoring
Generic + specific
for:
- model domain
- user
From previous
model projects
Specific for types
of users
Generic + specific
for:
- model domain
- user
User
Model Team
Single/multiple domain
Model Study
Plan
Data and
Conceptualisation
Model
Set-up
Calibration
and
Validation
Simulation
and
Evaluation
Reporting and client review take place in each step
Figure 3. The overall modelling support system consists of KB (upper left corner), MoST
(middle part) and model archive (upper right corner). The steps in the work of modelling
teams are shown in the lower part. The customization aspect Application Purpose is left out in
the guidance and monitoring tool components, as this item is not implemented in MoST.
MoST can also be used as a smart and powerful browser for the KB guidelines. Browsing is
supported in three ways. In a linear (textbook) format or according to the flowchart structure of a
modeling process represented in the KB or in way most users are familiar with by using a tree
structure as is used by most operating systems nowadays. This is realized in the guideline
component by providing three panels: a TREE VIEW (with an layout similar to MS Windows
Explorer), a FLOWCHART VIEW (similar to the structure of Figure 1) and a (textual) TASK VIEW (with
‘tabs’to facilitate browsing). The guideline component of MoST with these three panels is depicted
in Figure 4.
As a modelling project starts, the team member that initiates the project is the initial project
administrator. The administrator’s first responsibility is to give a name to the project and to specify
one or more subprojects, each consisting of one or more domains (not shown). Subsequently the
administrator defines which tasks have to be done in each of the subprojects, using the job
complexity templates from the guidelines as starting points (not shown). Next, the project
administrator composes the staff of persons to do the project (not shown). Users are added to the
223
team and their roles specified. Then, the administrator will give authorization rights, i.e. reading,
writing, decision making, to all persons involved, per subproject and per role and appoint – if
functional – extra project administrators (not shown). Finally the administrator can add and edit
questions and criteria, to be used by auditors (not shown). A project administrator can change all
project settings of an initialization phase during the project.
One or more team members with a role, associated with a specific task for that subproject can start
a task or skip a task, preferably with arguments why the task is skipped. All team members can
see, which tasks are completed, skipped, not yet started or in progress.
Figure 4. Screenshot of MoST’s guidance component with the TREE VIEW in the left, the
FLOWCHART VIEW in the upper right and the TASK VIEW in the lower right panel.
In the second phase of a modelling project, team members have to execute the tasks according to
their role and their authorization. MoST has to collect information about the execution of these
tasks for monitoring, control and auditing reasons (called keeping a model journal). The main
window to perform tasks in a modelling project (using the monitor component of MoST) is
depicted in Figure 5. In the top-left part of this window, below the main menu, a dropdown menu
enables users selecting one of the subprojects to work on. In Figure 5 the GW study (GW)
subproject has been chosen, dealing with the groundwater modelling issues in this project. In the
lower part of the screen the TREE VIEW is shown in the left panel. This can be used to browse
through the tasks to do and also to inspect the status of each task in this subproject. Different icons
are used to show this status, being a transparent rectangle for tasks not yet started, a yellow
rectangle for tasks started, a green OK sign through the yellow rectangle for tasks completed and a
red cross through the yellow rectangle for tasks skipped. The lower right panel is similar to the
TASK PANEL in the guiding component of MoST, providing adequate guidance on the task selected
in the TREE VIEW with tabs to select different views on the guidance for that task.
224
The upper right panel is called the ACTIVITY VIEW and it is designed to perform the task at hand by
executing, skipping or completing the activities associated with this task. This panel is the most
complex one to handle and it needs therefore more explanation. At the top of this panel MoST
keeps track of the start date and time of an activity. Below this time stamp box, users can select to
start an activity, if they are authorized to work on this task. The next user, that also wants to work
on this activity, is allowed to do so, but he/she will work in a different copy of the activity. The
server application of MoST will later merge all work done in a single and consistent model journal
of this project. When an activity in a task has been started, the user will go to the JOURNAL VIEW,
where the user has to describe in text what he/she did to perform this activity, select a method to
help completing the activity, and attach documents in a variety of formats. Such attachments are
implemented by providing a hyperlink in the model journal and sending attached documents to the
server for later use (reading, editing, printing, etc.). Time stamps of activities are also automatically
set or updated when an activity is later reworked. The list of activities at the task level uses similar
icons to indicate the status of the activities as are used for tasks in the TREE VIEW (the left panel).
Figure 5. Screenshot of main 'work' window in the project component, where the tasks in a
project will be monitored. See text for further explanation.
Decision tasks, presented in the TREE VIEW with a diamond shaped icon, require a decision instead
of simple completion as in normal tasks. Users, authorized to make the decision for a subproject,
can choose to continue to the next task or go back to a previous task as shown in the FLOWCHART
VIEW of the guiding component of MoST. Reviews tasks are special decision tasks and require a
decision too, but an essential element of review tasks is the discussion between team members with
different roles, including water manager, modellers and sometimes others.
Decision and review tasks allow users to accept the results of a task or redo one or more tasks. In
this way users can redo a model calibration, if they expect to get better results with new settings of
the calibration or they can use it for analyzing alternative scenarios with the model at hand or with
225
an alternative model. The process of (re)doing a task with a set of models and tools is called a
model-scenario. If we do the same task with other datasets we call it a data-scenario.
Switching to other tasks, starting, skipping and completing tasks and activities, triggers exchange
of information between the client application of MoST and MoST’s server application. In that case,
the model journal on the MoST-server will be synchronized with the local copy of the model
journal that has been changed. At the same time, all other local copies of that model journal will be
synchronized with the content of the model journal on the server.
Model journals also have an ontological structure, which resembles that of the KB, but with some
extra concepts and relations regarding the time-stamps and the information provided by team
members on what they actually have done (scenario information). This information is stored in
instances of the model journal ontology on a MoST-server. The content of a model journal is
exchanged between the model journal on the MoST-server and a MoST client (and vice versa) as
XML-file.
Next to the guidance component and the monitoring functionality in the project component, MoST
provides a powerful reporting functionality. This reporting facility filters the myriad of logged
information, monitored and stored in a model journal, and transfers it into a readable and compact
report according to requirements set by the user. In this way reporting is one of the ways to get an
overview of the project progress.
The purpose of the reporting functionality is to provide a report in a form customized for different
types of users, i.e. water managers, modellers, auditors, stakeholders and concerned members of the
public. The benefit of the audit trail recorded by the monitoring tool will be poor, without a
functionality to filter many recorded decisions made, methods and data used and other information
such as time stamps at the task level and at the activity level. Moreover, some information may be
confidential, e.g. some part of the audit trail may be restricted to specific types of users. The
reporting component ‘understands’the ontological terms from the KB and from model journals, e.g.
step, task, activity, method, users and user types, domains, job complexity, and other task
describing components. Therefore, the reporting component can handle information requests,
which depend on these ontological concepts. These requests are the result of filtering by team
members, who select what they want in a report. Subsequently the reporting component collects
requested information from a model journal and enables printing or exporting reports to HTML- or
PDF-format. In this way the reporting component facilitates generating audience-specific reports.
Training Material
MoST is a complex tool and novice users need training before they can benefit from all its features
to support modeling teams in their daily routine. Therefore, comprehensive training material has
been developed for students and professional modelers in water management. This training material
consists of written material, presentations, exercises that encourage using MoST in a test case
project and many screen-recording movies on MoST, its knowledge base, a case study and some
background information. The movies are the core part of the training material and aim at helping
users to work with MoST and act as a sort of animated help facility.
Testing
Scholten and Beulens (2005) have summarized the criteria to test such a KBS, including
verification (determining that the KBS is built according to its specifications) and validation
(determining that the KBS actually fulfills the purpose for which it was intended). Many of these
criteria are difficult to test, as there are no tools or general applicable methods. Here verification
and validation will be combined and referred to as testing.
226
The ontological structure of the KB and the process decomposition have been discussed by all
partners at project meetings and commented on by various (internal) testers and (external)
reviewers.
Reading and using the KB were the methods that the testers used to evaluate it. In this way the
content of the decomposition elements have been tested according to the following criteria:
correctness (i.e. capturing intuitions of domain experts?), completeness (i.e. no gaps regarding
steps, tasks, activities, methods and tools), redundancy (i.e. no unintended synonyms), consistency
(i.e. different ways of treating the same concept and does the content of the KB include
contradictions) and meaningfulness (i.e. intended users should be able to understand). These tests
were conducted at three levels: (1) internally reviewed by experts (modellers for water
management) not involved in developing the KB, (2) used in two series of test cases by professional
modellers, where MoST has been used in normal, commercial projects and (3) commented by three
external reviewers.
MoST (software) was tested at four levels: (1) by a project partner not involved in the design and
implementation of the software on reliability, correct functioning according to requirements, (2) in
both test series by professional modellers on adequacy (i.e. does the support tool adequately support
daily practice of professionals), (3) by the external reviewers on correct functioning, reliability and
adequacy and (4) by professionals attending workshops on adequacy and students attending master
degree courses on appropriateness for novel users. Furthermore, all users of the KBS were asked to
fill in an online questionnaire. In workshops and courses the training material has been used, which
tests its correctness and usefulness for professionals and students. Testing of training material by
using it was further evaluated with the online questionnaires.
A final test is scientific publishing. Many aspects have been published (Refsgaard and Henriksen,
2004, Refsgaard et al., 2005, Olsson et al., 2004) or are accepted for publication (Scholten et al.,
2006). Furthermore, the KBS is presented at many conferences and workshops.
Usefulness for Other Applications
MoST and its KB are designed to support model-based water management, but a claim that it can
beneficially support other complex processes too, seems realistic. Unlike workflow management
systems, MoST is designed to assist in complex, science related processes, which will not be
repeated many times. However, we expect that components of these processes are reusable in other
processes within the same domain. This idea is supported by expert opinions from the domain and
by the collected experiences in the HarmoniQuA project. At present this claim is further tested
within the EC-funded project AquaStress aiming at decision support in a stakeholder participation
centred process to mitigate water stress problems (shortage, excess, quality problems, etcetera)
throughout Europe. If this is successful, other fields of application will be assessed. Finally, it is
considered to extent the functionality in order to support hierarchies of processes too.
Key Findings
The case study described in this chapter resulted in a knowledge-based system (KBS) that provides
guidance and support for multidisciplinary teams working in complex projects, here instantiated for
model-based water management. The KBS consists of several components.
The first component is an ontological knowledge base (KB) with levels of increasing specialization,
i.e. from rather abstract to describing modelling processes for water management in all details. The
227
more abstract levels can be re-used for other processes than model-based water management, as is
tested at present. This is realized by filling the KB structure (the more abstract levels) with expert
knowledge on the new process (dealing with participatory water stress mitigation support), using
the second component, i.e. the knowledge base editor. The third component is the support tool,
MoST, which has several functions. It presents guidance from the KB to team members, customized
for their disciplinary backgrounds and their role in the project at hand. Next it facilitates to keep
records on what is actually done by each team member in its monitoring function. Further, it helps
generating reports, customized for various audiences and purposes. Finally, it aims at deriving
advice from previous projects, but this function is in the development phase and not yet
implemented. The fourth component of the KBS consists of training material to help novice users to
get familiar with MoST and employ it easily in their daily work.
Test results are promising, as they show – next to some bugs and minor shortcomings – that most
users (whether professionals or students) like the KBS and perceive it as useful. Nevertheless, the
same users expect significant problems with clarifying other potential users the costs and benefits of
using the approach and system and convincing them to use it. The best way to overcome this hurdle
is to clearly define the incentives of using it to all participants. First of all managers in their role of
clients for projects, for which the KBS is designed, have to understand how they can benefit from
its use. That clearly pertains to matters of effectiveness, quality assurance, auditable, and efficiency.
For participants in the modelling process incentives may have to do with clear and effective support
and guidance in the execution of tasks as enabled by the knowledge base and the modelling log. If
participants are convinced about associated benefits and incentives, if they are supported in going
through the learning curve, and if there are no hidden agenda’s then managers may enforce its use.
One of the major objectives of the KBS is facilitating communication and understanding between
members of a multidisciplinary team. The KBS uses English as single language (KB, support tool
and training material). In some countries, especially in northwest Europe, the foreign language,
English, is hardly seen as hurdle hindering its use by the professional community. In many central
and south European countries English will not obstruct researchers to use it, but many consultants
that do most of the (routine) modelling work will and cannot use it, unless translated in their
language. This is amusing, as the KBS intends to facilitate communication. This language problem
may even obstruct the usefulness of the KBS for team members with less professional roles, e.g.
stakeholders and concerned members of the public. Translating the KBS will only partly be a
solution, as it will introduce a new problem. Maintenance of the KB is now easy, but in the
multilingual case, it will be tedious endangering the consistency between the language versions.
The KB structure, implemented in Protégé2000, the KB editor and the support tool MoST
represents a promising new technology to support projects dealing with other types of complex,
techno-scientific processes. This is the result of the use of proper design criteria, which ensures a
complete split-up of the content of the knowledge base on one hand and the structural aspects of the
support tool and the KB on the other hand. In this way, filling the knowledge base with knowledge
related to other types of processes, results in new systems which can be used to run projects by
multidisciplinary teams for these other processes. The implementation of the KBS for
multidisciplinary model-based water management is a ‘proof of principle’of the correctness of the
followed approach. The full potential of this technology has still to be proven.
Impacts
The knowledge-based system (KBS) discussed in this chapter has different impacts on various sets
of persons involved. Individual domain experts (the providers of the knowledge) organized and
reviewed their expert knowledge before making it explicit and upload it to the KB. In this way, their
daily routine has been made explicit and improved in the direction of ‘best practices’. The same
domain experts, but now considered as a group discussed these pieces of knowledge, negotiated it
228
to reach agreement on the KB content, which resulted in a further progress towards ‘best practices’
and consistency in the use of terminology and methodology in the KB. The ontological nature of the
KB aims at such consensus and achieving consensus is therefore a major result of this case study.
The professional community of modellers for water management (in the Netherlands estimated to be
300-600 persons on a total population of 16 million) appeared to be attracted to the idea of making
the knowledge they daily use explicit. Initially they found the support tool cumbersome, but after
using it, they perceived it to be beneficial, as it increases the quality of their work without
decreasing their efficiency. Managers in their role as clients of modelling projects appeared difficult
to interest, but when interested they evaluate the KBS as valuable and often enforce its use.
Undergraduate students found the KBS complex and laborious, but teachers found that the KBS
increased their efficiency and the quality of the students’work. The multimedia training material
helped all novice users significantly in familiarizing with the KBS, especially the students. Some
professionals expected initially that the KBS is a cumbersome burden, forcing their daily practice
into a straitjacket. Training sessions with the multimedia training material helped them to get a
more positive view on its usability.
The problems at the start of the HarmoniQuA project (partners from 10 countries with different
level of modelling maturity and working in 7 water management domains with different levels of
maturity in their disciplinary background) was felt as a confusing starting point, hindering a
straightforward approach in the development of the knowledge base and support tool. But during
the project the condition sine qua non of an ontological approach, i.e. coming to agreement on
expert knowledge, facilitated to overcome the problems. In order to contribute their own
disciplinary expertise project partners had to understand the expertise of the other domain experts
and formulate their expert knowledge in a clear and understandable way for experts from other
domains. Partners from countries with a shorter history in professional model application for water
management experienced a fast intensification of their professional skills. In this way, the
HarmoniQuA project was also a process in which paradigm islands (countries and disciplines) were
connected to a firm network of mutual understanding, opening up challenges for international
cooperation and integrated multidisciplinary studies.
Conclusions
The development of the knowledge based system (KBS) including its ontological knowledge base
for multidisciplinary modelling in water management combines knowledge engineering technology
with modelling expertise. Modelling team members with different roles and working in different
domains of water management get guidance customized for their profile (domain, role, job
complexity). This technology seems promising for modelling projects in other application areas
than water management and shows potential for supporting other complex, multidisciplinary
projects too. This has yet to be proven.
Making modelling knowledge explicit and representing it in an ontology are the major benefits of
the KBS discussed in this chapter. The ontological approach makes improving, changing and
updating the KB easy. This is an essential prerequisite for ensuring that the KB can become a body
of knowledge shared by major parts of the professional modelling community. Its specificity for
model-based water management does not hinder extending it to other disciplines or application
domains, thanks to the flexibility provided by its ontological set-up. The present state of the KB is a
cross section in time of relevant and sufficient modelling knowledge. It is an intermediate product
open for improvements and updates, if scientific progress requires.
The ontological approach of MoST’s KB passed all stages of an ontological knowledge base
development with the design of an ontological structure, the development of a knowledge
acquisition tool, the implementation of the KB as instances of the ontological structure and the
229
building of a tool to use the KB for some purposes. This chapter does not focus on knowledge
engineering aspects of ontologies and on developing domain factual ontologies, because here the
ontological approach was only instrumental.
The development of the KBS and its use in two series of ten test cases seems successful. It solves
many of the problems for which it was designed, but several questions are still unanswered.
Technically the KB can easily be maintained and updated, but reaching agreement on its content,
outside of the shelter provided by an externally financed project as HarmoniQuA, is a hard job.
Persuading substantial parts of the professional water management community to use the KBS is
another hurdle to overcome. Developing this KBS is worthwhile in itself, but it will be worthless if
this tool cannot be found in the toolbox of professional modellers for water management.
•
Implementation of a complex knowledge-based system (KBS) like MoST and its
knowledge base can be facilitated by clarifying costs, benefits and incentives to
clients of multidisciplinary model-based problem solving projects for water
management, as they can enforce its use. In addition, all participants of such
projects must be convinced about the incentives associated with the use of
KBS.
•
Set-up of an ontological knowledge base requires knowledge engineers that
speak the language of domain experts instead of domain experts that are
enforced to do the work of knowledge engineers.
•
End-user involvement in the early stages of KBS development helps avoiding
crucial design blunders.
•
Communicating basic concepts of a KBS with intended users encourages them
to use it.
•
A Babylonian language problem is an often-neglected issue in knowledge
management research.
•
The complete KBS (KB, support tool and multimedia training material) can be
obtained from www.HarmoniQuA.org/.
Acknowledgements
The present work was carried out within the Project ‘
Harmonizing Quality Assurance in model based
catchments and river basin management (HarmoniQuA)’
, which is partly funded by the EC Energy,
Environment and Sustainable Development programme (Contract EVK1-CT2001-00097). The authors thank
all persons involved in this project (see http://harmoniqua.wau.nl/public/credits/credits.htm). They are also
grateful to Pasky Pascual (Environmental Protection Agency, Washington, USA), Nils Ferrand (Cemagref,
Montpellier, France) and Hugh Middlemis (Aquaterra, Kent Town, Australia), who acted as external
reviewers of MoST and its knowledge base. The constructive comments of an anonymous reviewer are
acknowledged.
References
BDMF, 2000. Protocols for Water and Environmental Modeling. Bay-Delta Modeling Forum. Ad
hoc Modeling Protocols Committee. www.sfei.org/modelingforum/.
Beulens, A.J.M. and H. Scholten, 2004, An ontological framework for structuring process
knowledge specified for the process of model-based problem solving. In: M. Makowski (Ed.),
18th Workshop on Methodologies and Tools for Complex System Modeling and Integrated
Policy Assessment, 6 –8 September, 2004, Laxenburg, Austria, 4-8.
230
Beulens, A.J.M. and H. Scholten, 2005, Towards a process ontology for a model based support
system for problem solving: the ontology bootstrap problem. In: M. Makowski (Ed.),
Knowledge Creation and Integration for Solving Complex Problems. The 19th International
Workshop on Complex Systems Modeling (CSM) jointly with the 6th International
Symposium on Knowledge and Systems Sciences (KSS), August 29-31, 2005, Laxenburg,
Austria, 3-8.
Borst, W. N. 1997. Construction of engineering ontologies for knowledge sharing and reuse.
University of Twente, Enschede, The Netherlands.
Chandrasekaran, B., J. Josephson, V. Benjamins. 1999. What are ontologies and why do we need
them? IEEE Intelligent Systems 14, 20-26.
Fischer, L., 2003. Workflow Handbook 2003. Future Strategies Inc., Book Division, Lighthouse
Point, FL, USA, 362 pp.
Gruber, T. R. 1993. A translation approach to portable ontology specifications. Knowledge
Acquisition 5(2), 199-220.
Gruber, T. R. 1995. Towards principles for the design of ontologies used for knowledge sharing.
International Journal of Human-Computer Studies 43, 907-928.
Kassahun, A., H. Scholten, G. Zompanakis, and C. Gavardinas. 2004. Support for model based
water management with the HarmoniQuA toolbox. In: C. Pahl, S. Schmidt, and T. Jakeman,
editors. iEMSs 2004 International Congress: "Complexity and Integrated Resources
Management". International Environmental Modeling and Software Society, Osnabrück,
Germany, June 2004.
Middlemis, H. 2000. Murray-Darling Basin Commission. Groundwater flow modeling guideline.
Project no. 125, Aquaterra Consulting Pty Ltd., South Perth. Western Australia.
Olsson, J., H. Scholten, B. Arheimer, and L. Andersson. 2004. Quality Assurance support tool for
catchment-based modeling: a test on the HBV-NP model for eutrophication assessment. In:
8th International Conference on Diffuse/Nonpoint Pollution, 24-29 October 2004.
International Water Association, Kyoto.
Protégé (2000). The Protégé Project. http://protege.stanford.edu (date accessed: 14 September,
2005).
Refsgaard, J. C., and H. J. Henriksen. 2004. Modeling guidelines - terminology and guiding
principles. Advances in Water Resources 27:71-82.
Refsgaard, J. C., H. J. Henriksen, B. Harrar, H. Scholten, and A. Kassahun. 2005. Quality
Assurance in model based water management - review of existing practice and outline of new
approaches. Environmental Modeling and Software, accepted for publication.
Scholten, H. 2001. An ontological approach in quality assessment of modeling and simulation. In:
A. W. Heemink, L. Dekker, H. d. S. Arons, I. Smit, and T. L. v. Stijn, editors. PROCEEDINGS
OF EUROSIM 2001, SHAPING FUTURE WITH SIMULATION, The 4th International
EUROSIM Congress, in which is incorporated the 2nd Conference on Modeling and
Simulation in Biology, Medicine and Biomedical Engineering. TU Delft, 6 pages on CDROM, ISBN: 90-806441-1-0, Delft, The Netherlands.
Scholten, H. and A.J.M. Beulens, 2005. Testing ontological support for multidisciplinary modelbased problem solving for water management. In: M. Makowski (Ed.), Knowledge Creation
and Integration for Solving Complex Problems. The 19th International Workshop on
Complex Systems Modeling (CSM) jointly with the 6th International Symposium on
Knowledge and Systems Sciences (KSS), August 29-31, 2005. IIASA, Laxenburg, Austria,
pp. 51-55.
231
Scholten, H., 1999. Good Modeling Practice. In: M. Makowski (Ed.), Abstracts of the 13th JISRIIASA Workshop on methodologies and tools for complex system modeling and integrated
policy assessment, IIASA, Laxenburg, 57-59.
Scholten, H., A. Kassahun, J.C. Refsgaard, T. Kargas, C. Gavardinas and A.J.M. Beulens, 200x. A
methodology to support multidisciplinary model-based water management. Environmental
Modeling & Software accepted for publication.
Scholten, H., R.H. Van Waveren, S. Groot, F. Van Geer, H. Wösten, R.D. Koeze and J.J. Noort,
2000. Good Modeling Practice in water management, Proceedings HydroInformatics2000.
International Association for Hydraulic Research, Cedar Rapids, Iowa, USA.
Scholten, H., R.H. Van Waveren, S. Groot, F. Van Geer, J.H.M. Wösten, R.D. Koeze and J.J.
Noort, 2001. Improving the quality of model-based decision support: Good Modeling Practice
in water management. In: A. Schumann, J. Xia, M. Marino, D. Rosbjerg (Eds.), Regional
Management of Water Resources. Proceedings of Symposium S2 of the 6th Scientific
Assembly of the International Association of Hydrological Sciences, Maastricht, The
Netherlands, July 18-27, 2001. IAHS, Maastricht, The Netherlands, pp. 223-230.
Van Waveren, R.H., S. Groot, H. Scholten, F. Van Geer, H. Wösten, R. Koeze, J. Noort, 1999.
Good Modeling Practice Handbook, STOWA, Utrecht, The Netherlands, in Dutch, English
translation: http://www.informatics.wur.nl/research projects/pub-pdf/gmp.pdf (date accessed:
14 August, 2005).
Huub Scholten is employed as senior scientist and assistant professor in Applied
Systems Science, in the Information Technology Group of Wageningen
University. His professional interests lie in the field of simulation model quality,
ontological knowledge engineering, model development, calibration, uncertainty
analysis, validation and simulation modeling methodology and also in the
development of software to support these activities. He participated in many
research projects, of which some as project leader. He is involved in the EUprojects HarmoniQuA as coordinator and AquaStress as partner responsible for
knowledge management issues. He has authored more than 75 publications in
domain specific model development, model application and in more generic
methods and tools in his fields of interest. Besides research he gives four
courses in Modeling and Simulation at Bachelor and Master level, supervises
Master thesis and PhD thesis work at Wageningen University.
Adrie J.M. Beulens (1948) is currently Professor of Information Systems, head of
the Information Technology Group at Wageningen University in the Netherlands.
He was the founding director of Wageningen School of Management at
Wageningen University and its dean until 2002. He is (co-)author and editor of
many articles in journals on Operations Research and Information Systems. He
is active as a management consultant for a large variety of companies especially
in the areas of ECR, SCM, logistics, information systems and E-business. From
2002 till 2005 he was scientific director for the focus area of Transparency for
KLICT. He is member of a number of (inter)national professional organisations.
Finally he is member of the board of ACC (Agri Chain Competence) and AKK
(Agro KetenKennis).
232
Design and
Implementation of
KM Solutions
13
for Organisations Dealing
Andrew Lewis
Ben Diamond
Knowledge Management for
Organisations Dealing with Critical
Incidents
Andrew Lewis ([email protected])
Ben Diamond, West Midlands Fire Service ([email protected])
Abstract
The RIMSAT project’
s goal was to develop and implement knowledge
management tools and processes for use by fire fighters. Advanced
technologies such as case-based reasoning software and new
knowledge formats such as the “unit of experience” enabled the
delivery of appropriate, easily comprehensible knowledge to fire
fighters in a very short amount of time. The unit of experience is a
way of communicating lessons learned in the field in a simple,
structured way without necessarily defining them as best practices.
The most difficult barriers to acceptance of these methods and tools
were the fears that the widespread discussion of errors would lead to
liability claims and that computer tools would interfere with the rapid
decision-making that characterises incident command. The key to
building support for the tools was to clearly define their role in the
decision-making process. They had to be introduced as auxiliary
resources rather than as obligatory tools that could interfere with
normal command processes. The liability issue required a careful
analysis of the fire service’
s responsibility with regard to the
extinguishment of fires and a review of jurisprudence in this area. The
conclusion following this review was that the fire service is indeed
permitted to make non-negligent mistakes and that discussion of
those mistakes should be allowed and encouraged.
Keywords: critical incident, knowledge management, fire fighter, liability, trust
Background
Industrial Context
Fire fighting has always been a job for which hands-on experience is essential. Though some
aspects of the job can be taught through training courses, many must be acquired “the hard way”,
i.e. through participating in incidents. For this reason, fire fighters have a great deal of respect for
the knowledge of colleagues who have been fighting fires for many years.
Nevertheless, while the importance of hands-on experience remains true, there has been a growing
sense that more could and should be done to share knowledge and experience about fire fighting
operations. Industrial fires and the associated procedures and regulations are becoming increasingly
complex and diverse. Even the most experienced fire fighter cannot expect to know how to deal
with all of the different hazardous materials and equipment potentially present at industrial
incidents.
237
Knowledge Management for Organisations Dealing with Critical Incidents
Problem
The RIMSAT consortium sought to explore how knowledge management (KM) processes and
technologies could be used in the fire service to improve training for and management of these
diverse, complex incidents. The challenge was to develop KM techniques and tools that would be
usable and useful in an environment teeming with time pressure, high stakes and stress on the user.
If the system proved to be useful in the fire service, it would likely have applications in other
safety-critical domains such as medicine, the transportation and energy industries, the military, and
the police.
If the system were to be used, it was imperative that it not contribute to a common problem in
critical incident management: “information overload”. An incident commander can generally cope
with no more than five spans of control (i.e. individuals, processes and/or systems that he must
supervise). Beyond this number, tasks should be delegated so that he does not lose control. KM
tools for incident commanders had to be able to deliver highly relevant knowledge without adding a
span of control and slowing down the decision-making process.
In addition to the technical difficulties of the task, there were cultural challenges as well. Fire
fighters can be distrustful of computer systems that claim to give advice about how to respond to an
incident. Many believe that such systems are more likely to get in the way of a good commander’s
split-second decision-making. How is a computer system going to give the appropriate solution to
incidents that are almost always unique? Every fire has its own particular characteristics and takes
place in unique environments and circumstances. To make matters even more complicated,
knowledge must be delivered quickly enough to be taken into account in the five minutes or less
that a commander has to make his/her decisions. The prospective users also asked themselves
whether the system could potentially diminish the need for experience and allow inexperienced
officers access to a ‘fast track’promotion?
The Learning Objectives for this chapter (below) summarise the questions that the project faced at
its inception.
•
How can a computer system deliver relevant knowledge in an appropriate time
frame for critical incidents that are highly complex, unique and quickly
evolving?
•
How can critical incident managers integrate outside knowledge into their
decision-making without experiencing “
information overload”
?
•
How can KM methods and tools gain the trust necessary for their deployment in
domains where lives are on the line and where trust is generally built through
long-standing relationships?
Approach
The RIMSAT project consisted of researching the user needs (in terms of the form and sources of
knowledge) and then developing and testing custom KM methods, processes, and a prototype
computer tool. In order to establish the user needs, a series of twelve focus groups were run with
the Staffordshire, West Midlands, London, South Wales and Hampshire fire brigades from the
United Kingdom. The project team included several members of the West Midlands and London
fire brigades and the final user trials were performed at these two organisations. A modified
version of the prototype was ultimately deployed at the West Midlands Fire Service, which is
served by 400 full-time operational fire fighters who attend roughly 60,000 incidents per year over
238
an area including three major cities (Birmingham, Wolverhampton and Coventry) and four
metropolitan districts (Sandwell, Dudley, Solihull and Walsall).
Early in the project, it was clear that a knowledge base could never contain solutions for all of the
myriad industrial fire scenarios. The goal was to deliver applicable knowledge to the fire fighters,
not just information, but these incidents required complex solutions that would take so long to read
and understand that the fire would have completely changed character by the time anyone could
determine whether a particular solution was appropriate. Furthermore, a lot of fire fighting
knowledge is inherently tacit – it is based on experience and cannot be applied in any simple,
logical manner. Often it is difficult to explain why fires behaved in certain ways or why certain fire
fighting actions worked or didn’t work.
For these reasons, the project team decided to implement tools and processes based on the following
key elements:
•
•
•
Unit of experience –a self-contained atom of knowledge including a single, specific
lesson, its important context details, and an explanation. Rather than addressing
entire incidents, a unit of experience applies to a punctual, specific set of
circumstances within an incident.
The storytelling technique. Within the “Explanation” section of the unit of
experience, the author can simply tell the story of what happened in the original
incident from which the lesson came. This allows fire fighters to record their
experiences without necessarily having a complete understanding of the reasons for
observed phenomena.
Integrated case-based reasoning (CBR) and model-based reasoning (MBR)
technologies. The CBR tool allows for highly structured, precise searches of
knowledge bases while MBR further refines the search and quickly performs
otherwise time-consuming calculations and predictions.
The Unit of Experience
The unit of experience is a lesson learned with an explanation of how it was learned and in what
contexts it is applicable (Delaitre 2000). Rather than addressing an entire incident, each unit of
experience focuses on one particular task within the incident. This compartmentalisation, i.e. the
breaking down of complex situations into smaller, more manageable parts, helps the user to quickly
concentrate on the most relevant information. For instance, if a fire fighter returns from a fire and
lists the lessons that he/she learned, one lesson might concern the identification of entry and exit
points for a building. Such a lesson could apply to a variety of different incidents, whether they are
fires, chemical spills, rescue operations, or building collapses, so it is important that the lesson not
be buried among irrelevant details of the particular incident from which it came. The incident
might have gone on for hours, generated a number of lessons, and involved many different fire
fighters, hazardous materials, and actions, but the particular unit of experience will address only the
lesson about identifying entry and exit points. It outlines the basic limits of the lesson’s
applications, but the user is expected to be able to judge whether it is useful in another set of
circumstances (Lewis 2004).
Following an incident, a fire fighter enters units of experience into a database through a standard
quality control process. During incidents or training exercises, fire fighters can describe their
current circumstances using pre-defined descriptors (e.g. “tunnel, underground”, “non-breathable
air”, “complex building design”) and retrieve relevant units of experience using the CBR search
engine.
Unit of Experience - A lesson drawn from an incident, presented with an explanation and the key
context information from that incident, so that a recipient can understand when and how to apply
the lesson.
239
Figure 1 shows an example of a unit of experience. The context details are organised according to
attributes and values of each attribute.
Attribute
Value
Breathable Air
Quality
non-breathable
Tactical mode
offensive
Objectives
save life
Risk Assessments
trapped
Incident Type
generic
Complex Building
Design
yes
Narrow Corridors
yes
Type of premises
tunnel, underground
Current factors
affecting
access/egress
miscellaneous access/egress complications
Rescue Actions
evacuation, search and rescue of civilians, search and rescue of fire
fighters, withdrawal of breathing apparatus (BA) team
Lesson
When committing BA crews from a plan of a site into an incident,
confirm the exact location of entry points in relation to the plan and
actual site prior to committing personnel into a potentially hazardous
area
Lesson Level
basic
Explanation
When committing BA crews into an underground complex the
committing officer confused the entrance and exit points with those
marked on the plan and as a result briefed crews incorrectly on
internal layout and search patterns. The officer believed he was
committing crews into the entrance when in fact he was committing
into the exit of an underground complex.
Address
Black Country museum
Preferred recipient
BA Entry Control Officer, BA Main Control Officer, Incident
Commander, Sector Commander
Incident Date
April 9, 2003
Author
Barlow
Validator
Martin Clark
Authoriser
Andrew Lewis
Source and
History
Training exercise in simulated underground mine museum.
Figure 1: Unit of experience
Storytelling in the Unit of Experience
Communicating fire-fighting knowledge is not always easy. Following an incident, it can be
difficult to recount the exact sequence of events or provide in-depth explanations of why certain
phenomena occurred. The whirlwind of activity during an incident can make it hard to remember
240
precisely what happened and often the witnessed phenomena are not understood by even the most
advanced experts. Expert knowledge in fire fighting is often “tacit”because it cannot be explained
through simple rules.
Despite all of these complications, fire fighters can almost always communicate their experiences
and lessons learned in a coherent way through stories. One example of such a case is a unit of
experience that recounted a tunnel fire in which the fire fighters were exposed to falling debris
being projected from the ventilation shafts of the tunnel. Some experts doubted that dangerous
debris could travel up the long ventilation shafts and endanger fire fighters. Based on what was
known about the incident and physical laws, it couldn’t be explained. Nonetheless, the author of
the unit of experience had witnessed the phenomenon and it was thus considered worthy of note.
For receivers of knowledge, the story is often the form of knowledge that allows them to understand
and relate most quickly to others’ experiences (Denning 2002). This mode of knowledge
transmission is particularly appropriate to domains such as fire fighting. Studies have shown that
the best decision-makers in safety and time critical domains tend to make their conclusions by
matching current situations with past experiences. They do not usually go through deductive
reasoning, i.e. cause-effect analyses (Klein 1998). For this reason, training in the fire services often
relies largely on the exploration of case histories.
The unit of experience contains an explanation that justifies the lesson. In many cases, the
explanation will tell the story of the incident from which the lesson came. This was an important
feature of the unit of experience because it allowed fire fighters to express themselves in a way that
was comfortable and effective for them. Fire fighters are accustomed to recounting on-the-job
adventures in informal settings like the coffee room, the dining hall or the gym. The readiness with
which they take part in such knowledge sharing was an advantage that had to be harnessed.
Like fire fighters, industrial technicians often prefer to explain lessons learned through stories rather
than through structured presentations. Many assembly line workers are not comfortable explaining
a lesson in an academic way, i.e. defining a problem context, taking a position, supporting it with
evidence, addressing counter-arguments, etc. In many cases, the storytelling method is more
effective, particularly in such domains where there is little time for exploring the causes of every
event. The same people that will not contribute lessons learned when they are asked to do so in an
academic manner will often readily contribute valuable experience in the form of on-the-job stories.
The unit of experience could be applied in such domains.
Integrated Case-Based Reasoning (CBR) and Model-Based Reasoning (MBR)
Technologies
CBR is a technology that allowed the storage and retrieval of units of experience (the “cases”) in a
structured way according to their context details. A set of generic context details for industrial fires
was defined (called the “domain model”) and the units of experience were stored according to these
pre-defined terms (attributes and values). Users can select the values related to their current context
in order to access similar units of experience and compare them in a structured way with their
current situation (Bergman 1998). Figure 2 demonstrates this case retrieval process.
241
Search Query
Attribute
Number of Stories in 3
Building
Case 1(x% similarity) Case 2(y% similarity)
3
6
Smoke Colour
Red
Red
Red
Rain Condition
Heavy rain
Light rain
No rain
Preferred Recipient
Incident Commander
Sector Commander
Safety Officer
Lesson
Na
It is important… .
Beware of… .
Explanation
Na
In this incident, the When there is
red smoke… .
rain… .
no
Figure 2: Case retrieval and comparison
For aspects of critical incident management that are well understood and can be distilled into
reliable models, it was considered worthwhile to do so. The RIMSAT MBR models were
integrated with the domain model used for CBR. Without delving into the mechanics of RIMSAT’s
model-based tools, their basic functions are to:
•
•
•
•
Suggest potentially important context details based on the current incident
description (thereby speeding up and optimising the case-based search) (MéridaCampos 2003)
Prioritise risks based on the specified context
Refine the search according to risk priorities (Klein, R. 2001)
Calculate resource needs and safety margins (thereby providing precise answers for
a range of circumstances within well-understood aspects of incident management)
(Mérida-Campos 2003)
Recognition-Primed Decision-Making
It is widely accepted that the dominant form of decision-making for fire service incident
commanders is Recognition-Primed Decision-making (RPD), as defined by Gary Klein (1998). In
RPD, commanders search their memories for experiences that are relevant to their current situation.
They test these experiences mentally, envisioning whether they are truly applicable. When they
find the most relevant experience, they adapt it in order to develop a proper course of action.
Case-based reasoning, as it is implemented in RIMSAT, proposes relevant experiences based on a
structured search query. In doing so, it draws on the collective experience of the entire network of
users. It provides the user with access to an organisational memory. Such technology can be seen
as feeding the RPD process with additional experiences (Lewis 2004). This relationship has been
described by military theorist Thomas Killion (2000). Killion proposes that a computer system
support the RPD process by widening the experience base of the decision-maker. Figure 3 outlines
this process.
242
A2
A2
C1
C2
1
2
3
UoEs reside in RIMSAT
experience base
CBR engine supplies
UoEs based on user
request
User evaluates and
adapts the most relevant
UoEs using RPD
Figure 3: Units of experience in the Recognition Primed Decision-making (RPD) process
Case Analysis
The Importance of the User Role Vis-à-vis the Computer System
Though the RIMSAT team had worked in close collaboration with the fire service to adapt the tools
and methods to the specific constraints of fire fighting, many fire fighters still felt that a knowledgebased decision support system would complicate the incident commander’s task and that it would
never be able to provide appropriate solutions as quickly as an experienced commander. They said
that they would have much more confidence in an experienced commander than in a computer
system.
The project team completely agreed with this point of view but suggested that it was not
incompatible with a well-constructed, well-used KM system. The team proposed to run a series of
training sessions that would explain the role of the incident commander vis-à-vis the system and the
difference between KM and artificial intelligence. There was a sense that KM was being confused
with artificial intelligence and that the role of the KM system had not been made sufficiently clear
to the user population.
During the half-day KM training session that was organised, fire fighters readily admitted that there
was not enough dissemination of lessons learned following incidents. As mentioned earlier, they
were also very aware of the risk of information overload. When presented as a means of
overcoming precisely these two problems, KM appealed to many fire fighters who had previously
viewed it as another dot-com era catch phrase.
In addition to providing an introduction to KM, the training session also explained the envisioned
roles of the tool and the user in the decision-making process. There was a tendency among the user
group to view the KM system as an artificial intelligence tool that would actually make decisions in
place of the incident commander. It was important to explain that the system was, on the contrary,
only intended as a supporting resource for the commander. It was explained that the system was
only expected to give a set of possible solutions to a given problem, thereby complementing the
RPD decision-making process.
Ironically, this evoked a concern over the fact that the system would sometimes propose incorrect
advice. This is indeed the case: no matter how good the validation process and how complete the
knowledge base, the knowledge proposed by the system will not always be appropriate to the
current situation. Herein lies the role of the incident commander. He/she must have the experience
and ability to recognize the proper solution among those proposed… and to know if none of the
243
advice is appropriate. This is comforting to the user because it implies respect for and preservation
of the incident commander’s primary role in decision-making.
The fire fighters accepted this breakdown of roles and responsibilities. The tool is expected to
provide the user with the most relevant cases in response to a search query. It is the user’s
responsibility to make appropriate decisions, regardless of the cases proposed by the system. The
system brings the collective memory of the entire network of users to the aid of the decision-maker.
It provides him/her with additional resources. However, the system advice does not change the
commander’s accountability for the decisions taken.
The fire service decided that the decision support system would be operated by one of the incident
commander’s staff because the operational commander of the incident has no spare time to deal
with a computer system. However, future evolutions of such decision support systems might be
voice activated and incorporated into helmets, perhaps allowing them to be used directly by
commanders1.
It is particularly important to emphasise these roles and responsibilities in the fire service because
of the potential consequences of poor decisions. If someone does not take the time to evaluate
advice given by the tool, lives could be lost. Anyone using such a decision-support system should
be certified to use it through a training program that ensures a full understanding of these concepts.
Making clear distinctions of roles and responsibilities can also avoid a phenomenon called
“malicious compliance” that can occur in high-stress hierarchical organisations. Malicious
compliance is when a person obeys the order of a superior despite knowing that the order is
inappropriate. The failure of the action is then attributed to the superior, while the lower rank is
free of responsibility. This phenomenon can also happen with unwelcome computer tools when
users purposely follow erroneous advice in order to prove the inefficacy of the tool or to shirk
responsibility. Such phenomena are less likely when the users have been taught that they are
expected to make good decisions with or without the decision support tool, and that they are
accountable for their decisions regardless of the knowledge presented by the tool.
Building Trust
Over the course of the project, there were several trust-related barriers working against effective
KM in the fire service. One barrier that nearly prevented the launch of the project was a fear that
the discussion of errors or the delivery of erroneous advice would expose the organisation or
individuals to criticism or liability claims. Once these fears were conquered, there remained distrust
of the system’s ability to achieve its stated objectives.
Liability Fears
Liability is one of the most sensitive issues surrounding KM in safety-critical domains. Almost all
critical incident managers will agree that discussion of errors is one of their primary learning
methods. That is why there is a debrief after almost every fire fighting intervention, military
operation, and surgery. However, such discussions are generally not communicated to an audience
beyond the actual participants, despite the obvious utility of such communication. The reason is
that the legal establishment (laws, governments, judges) has not evolved to protect this key learning
process (Cole 2003).
Though fire brigades have not been targeted to the extent of doctors in malpractice suits, such
lawsuits are still a concern that causes some legal departments to seek to restrict the discussion of
errors. Any wide inter-organisation discussion of errors is seen as running the risk of exposure to
1
The EC-sponsored AMIRA project is currently researching the viability of such a system:
http://www.amira.no.
244
liability claims. Before such open discussion can take place, there needs to be an acceptance that, in
the heat of critical incidents, poor choices will be made and accidents will happen. Every industrial
fire cannot be handled perfectly, just as every surgery cannot be a success. The goal is to learn
from errors, not repeat them, and minimise their frequency and consequences.
The RIMSAT project was nearly grounded because of such liability fears. It was extremely
difficult to find fire fighting organisations that were willing to participate in such a project. One
local government was worried that any system that led to poor decisions could be a source of
liability. Another group did not want any debrief reports to be shared, regardless of confidentiality
agreements within the project, because it was worried that past errors (if there were any) might be
exposed. There appeared to be a deeply ingrained policy of limiting liability exposure, even at the
cost of potential learning. Similar mentalities can be seen in business environments where
confidentiality is sometimes pursued excessively or blindly to the detriment of knowledge sharing.
Though many key people in critical incident management recognize that the path must be cleared of
legal barriers to make way for better organisational learning, such changes take a long time and
involve many players. A lot of the organisations are under the control of government entities that
are slow to evolve. There is a need to convince upper management, legal departments, and even the
operational users themselves. All levels of the organisation must agree to change.
The RIMSAT project and the participating fire brigades managed to overcome these liability
concerns by taking a close look at previous legal judgements in this area and thereby elucidating
their risks. This analysis brought clarity to the fire service’s role and set the stage for a much more
open and confident sharing of knowledge.
First of all, it should be noted that, in the event of litigation regarding a fire or other event, an
organisation is obligated by law to divulge all information and knowledge that it has about the
event. A litigator will obtain damaging information if it exists, and attempts to ‘hide’any such
information would be detrimental to the defendant in a court case. It is rather pointless to prevent
the open discussion of errors so as to not divulge compromising information if that information can
be requisitioned at any time.
When lawyers try to recover damages from the fire service for allegedly poor fire fighting
performance, the fire service’s defence is often based on the idea that when the fire service is called
to a building on fire, something catastrophic has already happened. The efforts of the fire service
are always well intentioned, and it is very rare that actions taken by the fire service actually make
matters worse than they already are. Fire fighters do their best to put fires out and they should not
be blamed for interventions that do not go perfectly. In fact, it has been widely accepted that
legally, fire brigades are not obligated to put out fires, but merely to respond to calls. As long as
they do not make the fire worse, they are not responsible for the outcomes.
The fire brigades in the RIMSAT project concluded that there was enough legal protection and
sufficient anticipated benefits to push forward with more open sharing of knowledge. Nonetheless,
legal departments of other organisations are still limiting the extent of knowledge sharing, so there
is a need to communicate to a wider audience on this subject. The ultimate goal should be complete
transparency, in which the practitioners, public, and legal authorities have a common understanding
of the risks associated with critical incident management and the expectations of each person
involved. When such an understanding exists, lessons learned can be discussed freely and errors
properly distinguished from malpractice. Associated legislation would allow critical incident
management organisations to work under the assumption that they could freely exchange lessons
learned without exposing themselves to unjust lawsuits. Meanwhile, they would be aware of their
responsibilities and know that they would be accountable for errors resulting from negligence –i.e.
malpractice. The prevention of such errors would serve as an incentive for everyone to implement
and use the most effective knowledge sharing processes.
245
Fire services like those participating in the RIMSAT project have taken a number of steps to
encourage a knowledge-sharing culture. They use terms like “learning points”and “development
opportunities”rather than “errors”or “mistakes”in order to emphasise the fact that errors can be
foundations for progress and should be accepted as inherent in their jobs. The brigades feel that this
encourages fire fighters to be more open and frank about discussing what happened at incidents. It
was very important in the RIMSAT project to reassure fire fighters that the system was not intended
as a means of increasing scrutiny of their job performance. For this reason, video recording of
incidents was ruled out for the present time, despite its potential benefits for the system. The fire
service unions feel that video recording would interfere with the decision-making process and bring
about unreasonable post-incident scrutiny of individuals’performance.
Trust in the Tool
A number of people in the fire service had seen too many big ideas come and go and/or had become
entrenched in an anti-technology mindset. Such a state of mind has many parallels in the industrial
world. How many companies have adopted complicated tools and then seen them neglected and
forgotten because they don’t fit the user needs and constraints? The same thing has taken place in
the fire service.
In industry, one sometimes comes across line managers who have been doing their job for 30 years
and are going to be damned before they let a computer system tell them how to do it. Many of these
companies work in a state of constant crisis in which “putting out fires” (figurative term for
managing business crises) is considered part of the everyday job of a line manager. His ability to
manage such crises is rooted in his experience, creativity, and years of hard work. These attitudes
are very much present in the fire service, where the literal putting out of fires is also seen as rooted
firmly in the experience, skill, and talents of the incident commander.
One of the only ways to convince these sceptics is through concrete examples of the computer
tool’s utility. For example, a powerful argument in favour of the RIMSAT tool was made via a fire
involving plastic crates. During the user trials, the prototype system was taken to a difficult fire
involving plastic crates. With the particular crates involved, the fire spread very rapidly, did not
appear to be retarded by water, and radiated intense heat. Several mistakes were made: the fire
fighters placed themselves and their vehicles too close to the fire (it was spreading rapidly and they
could have been trapped between the fire and the outer fences) and they actually accelerated the
spread of the fire by use of water jets instead of foam. They entered units of experience into the
database that said that, for fires involving such plastic crates, it is important to immediately call for
backup resources, place vehicles and personnel a good distance from the fire, and use fine sprays
and foams to extinguish the fire rather than water jets.
A year later, there was an incident involving the same type of plastic crates, but on a much larger
scale. Unfortunately, the units of experience from the previous incident were not available to the
attending fire fighters (being only a prototype, the tool was not yet deployed). The same mistakes
were made but with more serious consequences. The incident commander said that the units of
experience from the previous incident would have been an enormous help. Stories like this one
have led to significant pushing by senior management to implement a knowledge-based decision
support system.
Certification of Users
A certification process is essential to guarantee the integrity of such a KM system. Requiring
certification of users ensures a certain level of competence and instils confidence in the user
population.
As the system is often consulted in high-stress situations, a user’s lack of familiarity with the
system quickly translates into panic and frustration. This was witnessed in the first trials of the
246
RIMSAT prototype, when users were unfamiliar with the interface and had a difficult time
describing their situation with the RIMSAT domain model (structured search query). If users are
unable to get appropriate results within an extremely short period of time (one or two minutes), they
consider the system to be a failure.
By requiring that users go through a training process in which they familiarise themselves with the
domain model, many of these complications can be avoided. Such training prevents them from
making pre-mature judgments of the tool and ensures that they are well acquainted with the tool
before they use it during incidents. Certification programs also define roles and responsibilities
(thereby avoiding the malicious compliance phenomenon mentioned earlier) and provide a strict set
of procedures for contributing to and managing the system. Attainment of the certification provides
a sense of achievement and should also be a positive point in employee performance evaluations.
The human resources department can help provide incentives for certification. Nonetheless, the
training should be mandatory for all of those using the decision-support system during incidents.
Those who use the tool in their spare time for self-training purposes do not necessarily have to be
certified but should be encouraged to do so.
Key Findings
The user trials of the decision support prototype demonstrated that incident command teams could
indeed retrieve relevant knowledge quickly. Nonetheless, for budgetary reasons and technical
issues (no network capabilities on the command vehicles), the tool was at first only deployed for
use in the training of operational fire fighters. Fire fighters could explore a library of units of
experience at the firehouse and add new ones based on their experiences in the field but they did not
have access to the CBR search interface.
These deployment obstacles turned out to be blessings in disguise. By introducing it slowly in a
low-stress training environment, the West Midlands Fire Service was able to generate acceptance
and appreciation of the tool. After a year and a half of using the system to improve training, the
contribution and consultation of units of experience has become a regular part of the fire fighters’
jobs. The creation of units of experience has been integrated into the established business processes
such as the debriefing and reporting carried out by monitoring officers. Among the 400 operational
fire fighters, there are over 150 consultations of the case base per month and 1-2 new units of
experience created per month. The support for the initiative has helped resolve the budgetary and
technical issues and the complete CBR system will soon be used for real-time incident support.
For training, the West Midlands Fire Service has used a simplified version of the unit of experience.
Each unit of experience in the current system contains only a title, the lesson learned and an
explanation. The titles are listed alphabetically, allowing users to peruse the list and open those
units of experience that interest them. Fire fighters are apparently willing to take the time to dig
around in the library of units of experience. These unstructured units of experience and the
“perusing” search method would not be quick enough for use during incidents, but they are
sufficient for self-training. The basic process of building and reading units of experience has
important benefits for organisational learning (Prencipe 2001).
The fire fighters have shown a great deal of appreciation for the units of experience that have been
gathered. Many of the units of experience deal with circumstances and lessons learned that could
never have been anticipated before the actual event. They often treat issues that are not covered by
procedures or other training resources. Once the implementation took place, there was very little
reluctance or negativity with regard to the system. While contributors have been proud to share
247
their achievements and good practices developed on the job, they have also shown a great
willingness to communicate about their mistakes.
Though the deployment of the system and processes at West Midlands Fire Service was carried out
almost entirely by a single officer, it was essential to have the full support of the Chief Fire Officer
and to constantly communicate about the initiative. Reminders about the system were sent out
monthly in addition to alerts about new units of experience. Though the current use is already
satisfactory, the project leader in the fire service expects a large increase in use when the new
interface has been implemented.
Business Impacts
Organisations dealing with critical incidents have many of the same requirements and constraints as
regular businesses, but in an amplified form. For example, businesspeople want to be able to access
and understand information and knowledge as quickly as possible; in the fire service, every second
of delay can mean lives lost. When deploying a new system in a corporation, buy-in needs to be
won through the rapid servicing of problems; in the fire service, if the system hinders the
management of one incident, trust in the system might be irrecoverable. Thus, the application of
KM to the fire service provides interesting insights into problems dealt with in the business world.
High Expectations and Gradual Deployment
KM systems can often be victims of the high expectations that they generate. In the fire service, it
was essential to introduce the system as a tool for fire fighters to store, share and use their
accumulated knowledge. By design, the system is only as good as the people contributing to it and
using it. If the fire fighters had expected to be able to sit back and let RIMSAT tell them what to
do, the initiative would have been doomed. The introduction of the system as a training resource
was a perfect way to make its usefulness and limitations clear. The same measured expectations
and gradual introduction would likely foster wider acceptance of systems in some corporate KM
initiatives.
Domain Models can be Built for Complex Domains
The RIMSAT case also demonstrates that a domain model for a case-based reasoning tool can be
created for a complex domain like industrial fires. The same could likely be done for other complex
domains – markets, behaviours, medical conditions, etc. The resulting system allows for rapid
retrieval and comparison of relevant knowledge.
Units of Experience as an Alternative to Best Practices
A common problem with best practice systems is that the time and energy necessary to define and
validate a best practice can be a barrier to its use. Furthermore, the best practice can only be
defined for a generic context and will always be vulnerable to differences in circumstances, tools or
environment. This is particularly true in the fire service and other critical incident domains.
Procedures (i.e. best practices) have been established for years, but they really only form the tip of
the iceberg in terms of shareable knowledge.
The unit of experience method provides a way of formatting and sharing good practices, new ideas,
warnings, and lessons learned. There are very few cases when fire fighters can determine with
certainty whether their course of action was the best one or whether it would be the best in another
set of similar circumstances. However, they can still profit from the experience. By being simple
to create and not raised to the level of a best practice, the unit of experience can alleviate peoples’
inhibitions to using and contributing to knowledge sharing systems. They also allow people to
248
express “bad practices”, warnings, and other forms of advice. These benefits have applications to
both safety-critical domains and more mainstream domains.
Conclusions
Though this chapter has focused on the cultural barriers that needed to be overcome to institute
knowledge sharing in the fire service – namely, the complexity of the domain and the mistrust of
technology and KM –it is important to note the aspect of fire service culture that greatly facilitates
knowledge sharing: fire fighters want to learn and share their knowledge with colleagues beyond
the bounds of their own organisation. This is true for many critical incident domains and should be
capitalised upon.
Those who work in the KM field are used to dealing with employees who guard their knowledge
like state secrets and organisations that are progressing by baby steps towards the paradigm of the
knowledge-sharing extended enterprise. In fields like medicine, emergency services, and
transportation, there is a refreshing willingness to collaborate across typical frontiers. This opens
up enormous possibilities. In the UK, a KM system linking all of the 55,000 UK fire fighters is
currently under consideration. Afterwards, why not connect them with fire fighters in other
countries? Could the same be done for doctors in order to help with patient diagnosis and
treatment? For police trying to solve crimes?
While there are cultural barriers and trust issues like the ones mentioned in this chapter and
budgetary and bureaucratic difficulties associated with public service institutions, they are certainly
surmountable. If introduced in a careful, well-planned manner that takes these obstacles into
account, KM should soon make important contributions to the improvement of critical incident
management.
249
•
In high-stakes, high-pressure domains, slow deployment of KM systems (e.g. in
a training environment) can be a good way of avoiding resistance.
•
The unit of knowledge exchange has to have the appropriate “
granularity”
, i.e.
it must have a breadth or specificity that corresponds to the types of questions
posed by the users.
•
In complex domains, many people find it easier and more effective to explain
lessons learned with real stories from the field rather than with logical
arguments. Such stories can be captured and shared through a variety of KM
systems.
•
KM has to help reduce “
information overload”
, not exacerbate it.
•
For safety-critical decision-makers, it is important to make the distinction
between KM and artificial intelligence (AI). KM supports human decisionmaking by organising knowledge and information while AI actually takes an
active part in the decision-making process.
•
The best weapon against “
malicious compliance”with the KM tool is clear
communication of the fact that the user is accountable for his/her decisions,
regardless of the knowledge delivered by the tool.
•
The discussion of errors is one of the most valuable sources of knowledge, but
in order for it to be effective, it is important to associate errors more with
learning than with fault. Using terms like “
learning points”or “
development
opportunities”rather than “
errors”or “
mistakes”can be helpful.
•
Evidence of the repetition of mistakes is a powerful argument for organisations
to develop their KM processes and tools.
•
Mandatory user certification programs, organised in conjunction with the
department of human resources, are a good way of building trust and
understanding of KM systems, particularly in safety-critical domains. They
prevent pre-mature judgments and improper use of the system by ensuring
that all users have completed the required training.
•
Within critical incident management and public service communities, there is a
refreshing willingness to share knowledge between organisations.
Acknowledgements
We would like to thank all of the members of the RIMSAT project team, their organisations, and the
European Information Society Technologies Program for making this research possible. In particular, we
would like to thank Eric Auriol, Andrea di Trapani, Axel Antoniotti (Kaidara Software); Peter Cox (TMPL
Consulting); Martin Hunn, Graham Stagg (London Fire Brigade); Carlos Mérida-Campos, Emma Rollon,
Ignasi Rodriguez-Roda, Ulises Cortès (Lequia – University of Girona); Ben Ward, Dave Bowen (Teradyne);
Frank Sheehan, Martin Clark (West Midlands Fire Service); Claude Koemmerer, David Faure (Nemesia);
Paul Hearn (European Commission) and Patricia Arundel.
References
Bergmann, R., Breen, S., Göker, M., Manago, M., Wess, S., (1998) Developing Industrial Case
Based Reasoning Application, The INRECA Methodology.
250
Cole, M. (2003) The accretion of error : Professional Knowledge and the Making of Mistakes in
Health Care. KnowledgeBoard, http://www.knowledgeboard.com/cgi-bin/item.cgi?id=98601,
2003.
Delaitre, S., Moisan, S., Mille, A. (2000) Instrumentation d’un processus de retour d’experience
pour la gestion des risques. Proceedings of the conference IC2000, pages 3-10, N. AussenacGilles Editeur, Toulouse, May, 2000.
Denning, S. (2002) Advance text of Squirrel Inc. : A Fable of Leadership Through Storytelling.
Accessed via : http://www.knowledgeboard.com/cgi-bin/item.cgi?id=95935, November,
2002.
Killion, T. H. (2000) Decision-making and the levels of war. Military Review, Fort Leavenworth:
Nov/Dec, 2000.
Klein, G., (1998) Sources of Power: How people make decisions. MIT Press.
Klein, R. (2001) The Information Gap – Risk Assessment and Operational Command at Incidents
Involving Hazardous Materials. Where Should We be Going? VectorCommand Information
Bulletin, Summer 2001, International Edition 0.4, page 11.
Lewis, A., ed. (2004) Learning from Critical Incidents: Knowledge management methods and
technologies for supporting critical incident managers. Editions Sapientia, Paris: 2004.
Mérida-Campos, C., and Rollon, E.(2003) "Using a Relevance Model for performing Feature
Weighting.” In Proceedings of CCIA 2003, IOS-Press 2003.
Prencipe, A., Tell, F. (2001) Inter-project learning: processes and outcomes of knowledge
codification in project-based firms, University of Sussex.
Andrew Lewis is currently Knowledge Management Coordinator for Accenture
Financial Services in France. During the RIMSAT project, Andrew was working for
Nemesia, a French consulting company specializing in knowledge management. He
is co-author and editor of the book Learning from Critical Incidents: Knowledge
management methods and technologies for supporting critical incident managers
(Editions Sapientia 2004). He is also editor of the KM & Critical Incident
Management special interest group (CIM SIG) on the KnowledgeBoard website
(www.knowledgeboard.com).
Ben Diamond is a Station Officer with West Midlands Fire Service in the UK
(www.wmfs.net). He has worked in this busy urban emergency service since 1988,
gaining a wide experience of this domain. He contributed to the user trials and
evaluation during the RIMSAT project and is now taking this work forward with the
AMIRA project (www.amira.no) in which West Midlands Fire Service are users.
251
14
prodela
–On-demand Training for
Liza Wohlfart
Dorothee Frielingsdorf
prodela: On-Demand Training for
Liza Wohlfart, IAT, University of Stuttgart ([email protected])
Dorothee Frielingsdorf, IAT, University of Stuttgart ([email protected])
Abstract
As the modernisation process of the German public administrations
proceeds, theoretical concepts have to be transferred into real life
practice, a time-consuming process that seriously threatens the
productivity of public servants. What is urgently needed is ad hoc
support in the work process that helps to learn about basic ideas, to
apply them and to find fast solutions for upcoming problems and
challenges.
Aim of the German project prodela was the development of an ondemand training for tackling this problem. It was realised by combining
e-learning, traditional seminars and knowledge management in an
innovative mixture that enables on-demand support and strengthens
community aspects. The basic platform adapted for providing the
prodela on-demand training accordingly was not an e-learning but a
knowledge management system, ready to provide easy browsing and
communication facilities.
Apart from developing a technical prototype, the project also led to
scientific insights concerning both the necessary requirements for
designing on-demand support for public servants and for
implementing it.
Keywords: on-demand learning, e-learning, knowledge management, public administrations
Background
Industrial Context
The public sector in Germany employs around 4.8 million people, 1.57 million of whom work in
the 13,800 communities of the country (Bundesministerium des Inneren, 2002). The working
conditions of these communities have dramatically changed in the last decades. The demands of
citizens have risen, customer-orientation is now considered to be an essential part of the service
mentality. At the same time, the financial situation of public bodies has dramatically changed, as
their budget has become tighter. As a consequence, public administrations are currently confronted
with a modernisation process that focuses on the implementation of business management
instruments and challenges both the traditional working procedures of public bodies as well as their
cultures. In a survey of 297 German cities, 72% of them reported to have optimised their processes
in the last five years in the course of re-structuring projects. The use of Internet, e.g. for
eGovernment, is one of the big issues in this process, for improving internal, as well as governmentto-business (G2B) and government-to-citizens (G2C) relationships (Kommune21, 2005).
Learning about new processes and adapting to them is a time-consuming process. The learning
processes related to the modernisation of the German administrations seriously threaten the
productivity of the public servants, as they have to try to keep updated on changes and
255
developments, search for necessary information and transfer it into their daily practice. The German
project prodela, which was funded by the German ministry for economy and work (BMWA) from
January 2002 to July 2005, focused specifically on reducing this loss of productivity: "prodela" =
Produktivitätssteigerung durch problemorientiertes on demand-Lernen am Arbeitsplatz (increase in
productivity through problem-oriented on demand learning in the work place).
Problem
A main problem of the modernisation process of public administrations in Germany is that its
realisation is oriented on theoretical models, such as the "Neue Steuerungsmodell" (new steering
model), whose implementation in real-life practice has to be done by the single public
administration. The related problems that subsequently occur are then experienced by different
people in different administrations – but not shared. Solutions that have proven to be successful for
a specific case are not documented and made public, so that others can profit from them. This fact is
even more important as there is already a growing awareness of the importance of inter-communal
exchanges and collaborations, in areas such as eGovernment (Kommune21, 2005), IT
infrastructure, business support or tourism (Städtetag Nordrhein-Westfalen 2005).
Apart from the lack of a community perspective, the available training for public servants often fails
to provide ad hoc support on up-to-date contents. Traditional face-to-face seminars e.g. can be a
good introduction to a topic, they however cannot help to cope with the upcoming problems and
challenges thereafter and cannot keep the learners updated on new standards and regulations.
In addition, although there is a range of scientific findings on didactical models, many traditional
trainings do not make a real use of them. Traditional seminars often base on purely instructionalist
approaches, which can be useful for teaching the basics, but are not sufficient for on-the-job
support. The knowledge thus acquired is stored and can be reproduced, but it is difficult to apply it
for coping with daily tasks. If learning is to be a continuous process, integrated into work, other
approaches have to be favoured that provide more of an on-demand support.
This on-demand support can be related to the didactic approach of situated learning, which is based
on constructivistic ideas (see e.g. Mandl, Gruber & Renkl, 1997). It considers knowledge as
context-dependent, i.e. as a construct that is build up individually by each learner. The situation the
knowledge is created in is thus of great importance and should not be too different to the situation it
is applied in to ease the transfer.
E-learning can be an effective means to provide training in the work process. It however challenges
the discipline and self-organization competences of learners, so an effective support e.g. via online
tutoring is important. The exchange in online communities can provide a useful frame for this, as
well as for the exchange with other learners, which is crucial for motivation and knowledge
transfer. These elements are, however, neglected by many e-learning applications, which are based
on a traditional course model, sometime supported by glossaries and FAQ (frequently asked
questions) lists.
A general problem of public administrations in Germany today is that their available budget is
shrinking and shrinking, despite the high costs for trainings and experts. What they thus need is a
training that is both low prized and provides practical ad hoc support in the work process.
256
In this case study you will learn about
•
the main characteristics of on-demand training;
•
the specific requirements of the target group public administrations;
•
the critical success factors for creating and implementing on-demand training in
public administrations;
•
how to evaluate the efficiency of on-demand training.
Approach
The prodela on-demand training was designed by creating a first prototype, based on an assessment
of the user requirements, which was then optimised through various user tests. The user
requirements analysis was mainly done by means of work observations in public administrations. In
addition, experts were interviewed and a literature review was done. On basis of the results of the
user requirements analyses, standard learning scenarios, i.e. the most common situations where ondemand support is needed during the modernisation process, were identified and the necessary
means of support derived. The standard learning scenarios then formed the frame for designing the
didactical concept of the prodela on-demand training with its mix of traditional seminars, e-learning
and knowledge management.
The evaluation of the prodela prototype was done on the basis of a four-step evaluation model. This
model was mainly informed by the level-based evaluation models of Kirkpatrick and Schenkel 1.
Kirkpatrick's four level evaluation model is one of the most known evaluation models. It
distinguishes between the levels "reaction", "learning", "behaviour" and "transfer" for evaluating
the efficiency of a training. Schenkel added two levels to Kirkpatrick's model to consider as well the
evaluation of the learning solution by experts during its development ("product" level) and the
economic success of the training measure ("return on invest" level). As the timeframe of the
prodela project was restricted, it was not possible to assess the long-term return on investment. In
addition, the levels "behaviour" and "transfer" were merged into one ("transfer"). The evaluation
model of prodela thus consisted of the following four levels (see Figure 1):
•
Product: how is the training designed, what are its features?
•
Reaction: how is the training used, how many users make use of it, how often do they do so
and when?
•
Learning: what learning effect can be observed, which knowledge could be acquired,
deepened, expanded?
•
Transfer: how strongly is the acquired knowledge and the ad hoc support used in the daily
work processes, how has the behaviour changed?
The evaluations were then done by means of interviews (especially for the "product" level) and
surveys ("reaction" and "learning" level) with test users as well as through observations of the
learners' behaviour in the Fraunhofer IAO learn lab2, a specific laboratory designed for analysing
the learning behaviour of e-learning users ("transfer").
Schenkel, P. et al. (2000). Qualitätsbeurteilung multimedialer Lern- und Informationssysteme. Evaluationsmethoden auf dem
Prüfstand, BW Bildung und Wissen Verlag und Software GmbH, pp. 59.
2 More details on the IAO learn lab are provided in later sections.
1
257
End of project
Focus of the
evaluation
Transfer
Learning
Reaction
Product
Project start
Figure 1: prodela evaluation model
In addition to the feed-back of the users, the developed prototype was also evaluated by
interviewing the trainers leading the face-to-face seminars and providing help for the users as well
as the call center tutor. The focus of the evaluation changed from product and reaction to learning
and transfer in the course of the development.
Case Analysis
User Requirements Analysis
The observation of the work processes in the public administrations produced some concrete
requirements for the prodela on-demand training and showed some standard learning scenarios,
from which further user requirements could then be derived.
The main concrete requirements for the prodela on-demand training the observations revealed are
the following.
•
Most public servants did have an own computer at their work place. E-mail was used more
and more regularly, while not all of them did have an Internet access. The experiences of the
employees with the Internet varied accordingly.
•
Especially the public servants working in direct contact to customers need short,
independent modules for learning, as they cannot interrupt their work processes for a long
time.
•
A stronger exchange of experiences with employees of other administrations was strongly
requested. This exchange of experiences was already realized by means of informal work
groups, which however were rather short-termed due to the lack of moderation and
management. In addition, these work groups were not transparent to other people, so they
were rather closed with a more or less casual choice of members.
In addition to these concrete requirements, four main standard learning scenarios could be identified
during the work observations the prodela solution would have to support. From these learning
scenarios, further requirements for the solution design could be derived.
258
•
1st learning scenario: learning for a concrete effect. The aim in this situation is to learn
about a concrete process or procedure, such as the design of a controlling report. This
scenario can e.g. be supported by checklists and FAQs-lists.
•
2nd learning scenario: learning about new guidelines. The challenge for learners in this
situation is to familiarise themselves with a new area or task. Useful support material here
includes case studies from other public servants and the exchange with others in forums and
chats.
•
3rd learning scenario: learning for rationalisation and effectiveness. Public servants are
confronted with more and more complex tasks, so learning how to rationalise things and
how to be more effective in one's work processes is crucial. Tutors and professional experts
can be a good guide in this respect.
•
4th learning scenario: learning for acquiring implicit corporate knowledge. In the course of
the modernization process, it is indispensable to inform public servants about the strategic
goals and vision of the organization and to help them see the impacts for their daily work. A
restricted forum for employees of one administration can be a good place to discuss these
things.
Community Building
The first prototype of the prodela solution was then designed in the following way. A typical
prodela course would start with a face-to-face seminar providing an introduction to the topic; the
second phase was then based on e-learning modules (web-based training, WBT); the third was a
face-to-face seminar again, focusing on certain aspects of the topic in more detail and the practical
application in the daily work. The fourth and final phase then was the self-managed on-demand
phase, where learners would apply their acquired knowledge in the work process, supported through
a rich document data base of so-called "on-demand contents" (articles, case studies, etc.; see also
below).
offline
elements
face-to-face seminar
(introductory)
online
elements
(platform)
e-learning modules
(WBT)
face-to-face seminar
(details/application)
on-demand documents
(ODD)
tutoring services: call center tutor supported by platform provider and
seminar trainers; contact details (mail, phone) online
forums (moderated by tutor respectively seminar trainers)
time
Figure 2: Overview on a typical prodela course to a specific topic
Right from the start of the course, the learners were supported by a call centre tutor (mail, phone,
forum contact), who supported them in organisational, technical and topic-specific issues.
259
Furthermore, an on-line community of practice (CoP) was build up by means of topic-related
forums (moderated by the seminar trainers), a general forum (moderated by the call-centre tutor)
and contact lists showing the contact details of all learners in a course as well as their function and
position.
An important aspect of the prodela solution was the support of the collaboration of the learners of
the prodela community (peer support). This aspect had a specific importance for the on-demand
focus of prodela, as many problems in the course of the modernisation process arise from the
practical implementation of modernisation concepts. These problems can often be solved much
faster by peer contacts than be professional experts. In addition, it was important to pay respect to
the limited budget of the administrations, so peer support was considered as an economical
alternative to expert tutoring and a useful means to update the content database (as learners
themselves were asked to provide case studies, checklists and so on).
The Platform
The platform that was used for prodela was an adapted knowledge management platform, not an elearning platform. This decision was made on basis of the specific requirements of the on-demand
approach, such as the strong interlinking of the learners and the dynamic and heterogeneous
character of the offered contents.
Figure 3: The prodela platform
260
Design of the Contents
prodela was conceptualized as a dynamically growing system, where new contents are regularly
added. The identification of the main learning scenarios here helped to identify, on the one hand,
the most crucial contents (e.g. e-learning modules, checklists, case studies). On the other hand, they
informed the concrete design of these contents, as e.g. the e-learning modules were primarily
designed as basic introduction to a topic, while other contents such as checklists and FAQ-lists were
offered as add-on for on-the-job support. The main content types offered by prodela were:
•
web-based modules: e-learning modules designed to provide basic information and training
for a certain topic. Every topic was introduced by means of 6-10 single modules, which
were kept short (10-15 minutes needed for each) to enable working with them during short
breaks in the work process, but which related to each other to provide a holistic introduction
to a topic.
•
on-demand contents: all contents ready to provide practical on-demand support in the work
process were summarized as "on-demand content". This included e.g. FAQ-lists, case
studies, articles and templates. On-demand contents were offered by professional experts, as
well as by the learners themselves, who provided case studies on tasks they had realized and
templates they used in their daily work.
The quality of the prodela contents was ensured by means of a quality management process. All
contents provided by learners were checked by experts for professional correctness and by the
prodela call center tutor for design aspects and correct spelling. The quality inspection status of
each document was indicated.
Figure 4: prodela e-learning module (on “controlling”)
261
Tutoring
Tutoring can have a whole range of different functions, from the active pushing of information to
the passive pulling of support. It can be personal and electronic. In prodela, there was online- as
well as offline tutoring, the first being provided by online documents (e.g. administrative FAQ, user
guide), forum and mail contacts to professional experts and the call center tutor. Offline support
was offered during the face-to-face seminars and via phone.
Evaluation and Feed-Back
The main evaluations in the course of the prodela project focused on the levels product, reaction
and learning and were done by means of a survey done with all course participants. The survey
questionnaire was formulated on the basis of detailed interviews with single learners, who tested the
very first prototype of the prodela on-demand training and provided important feed-back for its
further adaptation.
Some of the feed-back of the survey evaluation pointed to the contents' lack of focus on the specific
target group public administrations at the time. Some of the first contents still e.g. featured
examples from private companies or offered standard templates that were hard to transfer for an
application in the public servants' daily job. To prevent this problem in the future, a detailed
authors' guideline was developed that clearly explained the focus, style and quality expected for
prodela contents.
Other feed-back tackled the functionalities of the platform in general as well as of the forums and
the e-learning modules in specific. The user guide that explained all technical features and the
available contents and tutoring offerings was also a feed-back result. An important output of one
specific survey on technical problems and requirements revealed that many learners could not make
use of the audio function of the e-learning modules that was at first implemented as a standard
feature for using them properly. As a consequence, the e-learning modules were then equipped with
a text that could be blended in as substitute for the audible explanation.
The design of the on-demand contents was another source for further improvement. A detailed
design template was elaborated to ensure the homogenous presentation of all contents. In addition,
content templates were realized for all important content types to help learners offer their own
contents to the learner community in an appropriate form (e.g. how to write a case study).
Concerning the tutoring, most learners wanted to be regularly updated on new contents and courses.
A newsletter was thus designed, which was prepared and sent out by the call center tutor. Apart
from providing help (pull function), the prodela call centre tutor thus also got an active push role as
he informed the users of new courses and contents as well as events and articles related to public
administrations in modernisation processes.
Scientific Insights
The learn lab observations and the interviews with the trainers near the end of the project mainly
served for validating the real effects of the developed training solution (transfer level), i.e. the
application of the contents and the use of the platform in the daily work. In addition, they helped to
obtain more scientific insights concerning the general requirements of on-demand training for
public servants.
262
Learn Lab Observations
The learn lab consists of two rooms, separated through a semi-transparent window, where users can
work with e-learning applications in one room, observed through the window and via screen sharing
by the test leader in the other room. In the prodela learn lab observations, participants were asked to
cope with typical challenges and tasks arising in the work processes of public servants, such as
preparing themselves for a new topic or informing themselves about how to implement a certain
method in the practical work. Before and after the observations, they were interviewed to find out
about the influence of using prodela on their attention, exhaustion, etc.
Figure 5: Learn lab test at Fraunhofer IAO
The learn lab observations were a first possibility to really watch people work with the prodela
platform, some using it already as support in their daily work, some being confronted with it for the
first time. The test users in general could be seperated into three categories:
•
learners that had participated in face-to-face seminars and had used the platform reguarly;
•
learners that participated in the seminars (where there was also an introduction to the
platform), but had not used the platform a lot since then;
•
learners that had neither participated in seminars, nor made regular use of the platform.
During the learn lab observations, they thus worked with the prodela platform for the first
time.
It quickly turned out then that the users' experience with the system had a big influence on how they
used it. Learners that had participated in seminars and had often used the platform worked very
structured and fast with the system and used all its features (e.g. consulted forums and used their
contacts to other prodela learners working on similar topics). They also used other on-demand
support, such as Internet search engines, so that prodela seemed to have been integrated in their
daily work, as supplement to their usual tools and contacts. What is interesting is that these learners
used the e-learning modules they had formerly used to learn about a subject in a step-by-step way
now as an additional work of reference, i.e. not working through them, but browsing them in a fast
and very structured way. The regular users handled most of the tasks without any problems.
The users that had participated in seminars but had not made a lot of use of the platform used the elearning modules in a very sequential way. Contrary to the experienced learners, these modules also
were the main content they focused on, they hardly used any on-demand documents and did not
make use of existing contacts to other learners (although they identified contact details of people
involved in similar topics). After a first start with the system, where they worked rather slow and
unstructured, these learners however also quickly came to use the platform in an efficient and
speedy way. The irregular users had some problems in solving their tasks, which mainly resulted
263
from their low level of experience with the platform (they e.g. were not familiar with some features
of the forums).
For the third category of users, the prodela platform was an entirely new medium; they only got a
short introduction to its features through the test leaders. These users kept working in a rather slow
and step-by-step way, concentrating on e-learning modules. They used the forum to find
information and identified contact details of other learners. When trying to solve their tasks, the
unexperienced learners had to cope with various problems. One of them reported that this kind of
learning was more exhausting for him than traditional seminars; another one had the feeling that it
had taken her too long to find the necessary information.
The main conclusions that could be drawn from the observations of the different learner groups are
the following:
•
There is a qualitative development of the learners' use of the platform, from the
concentration on e-learning contents to an experienced on-demand usage of various media.
•
Being familiar with the platform's functionalities and contents is a basic precondition for its
usage. It is thus crucial to give new learners a good introduction to it.
•
The use of contacts to other learners (community aspect) depends on the users' familiarity
with the community elements of the platform (forums, contact lists), as well as on how long
they have been working with the system.
•
For new learners, it could be useful to get a step-by-step introduction to the system, so that
they e.g. first learn about the basic functionalities of the platform, then participate in a
blended learning course, and then get an introduction to the on-demand usage.
Interviewing the Trainers
The focus of the interviews with the trainers was on their experiences with the courses, as the ondemand concept was a new experience for all of them. The training they usually offered were pure
face-to-face seminars or pure e-learning, so the main feed-back of them accordingly concerned the
blending of the online and offline phases, i.e. how to successfully interlink the contents and means
of the different mediums. The main scientific insights from the trainer interviews were that:
•
there has to be a holistic concept for the tutoring and a profound introduction to the specific
course concept. Trainers usually offering traditional seminars have to be prepared on how to
structure their seminars and their online support to ensure a good blending of the different
phases.
•
learners need fixed learning times during the week, especially for learning the basics of a
topic.
•
the training must be supported by the top level of the administration, i.e. the head and the
HR responsible.
Key Findings
The prodela project revealed some important success factors for on-demand training for public
servants. These factors can be summarised according to the categories contents, course design,
tutoring, technical issues and implementation.
264
Concerning the contents, it is crucial that there is a specific focus on the target group public
administration. The training provided by e-learning contents that are designed for private companies
and/or feature their examples is hard to transfer to the work processes of public servants. This is one
of the reasons why it is crucial to provide a style guide for the authors of the online training
concerning the design, the content and the structure. E-learning modules should e.g. be short and
illustrative and have cross-references to other modules. For the on-demand contents, it is important
to provide templates for the learners as well as the experts as a guideline on how to draft them.
As far as the courses are concerned, it turned out to be of major importance that there is a good
scheduling of the face-to-face seminars, so that there is an optimal time span in between them for
the online phase. In addition, the trainers as well as the learners should be instructed on the ondemand concept. Trainers e.g. have to have some guidelines concerning how to make a good blend
of online and offline phases, what to present in the first present seminar, what in the last, and how to
support learners in the online phase in between and after. Learners alike have to be informed about
how they can make an optimal use of the different phases. Concerning the participants in the
courses, a mixture of learners from different communities in the face-to-face seminars should be
fostered to strengthen cross-organisational community building.
The main problem of the trainers turned out to be the initiation of the online communities of
practice. They found it difficult to initiate collaboration and knowledge exchange in the different
forums and via sharing practice reports. There should thus be a holistic concept on how to activate
the communities of practice, from which tutoring guidelines can be derived. This concept should
not only focus on forum activity, but also pay respect to the correlation of the different mediums,
i.e. how the interplay between forum moderation, content publication and newsletter
announcements can be optimised to get a more active participation of the learners. For the
newsletter itself, a specific guideline is also necessary that specifies – on basis of the tutoring
concept –its whens, hows and whats.
Some of the crucial success factors concerning technical issues were a detailed and regularly
updated user handbook and an automated user administration.
To support the successful implementation of the on-demand training in an administration, it proved
to be important that the there is promotion from the top level. User should have fixed learning
times per week. And there should be a training responsible in each administration.
Business Impacts
The prodela on-demand training presented in this case study provides a useful example of how ondemand support for public servants can be designed and implemented. Moreover, the insights
gained in the project can also stimulate future adaptations of the on-demand approach for other
target groups and further scientific investigations on the successful blending of e-learning
approaches with knowledge management aspects in general.
Conclusions
As promised in the introduction to this chapter, the prodela case study presented has specified the
main characteristics of on-demand training as well as the specific implications for the target group
public administrations. On-demand training can be understood as an ad hoc support of learners in
the work process, which can be realised by a blend of e-learning, traditional seminars and
knowledge management. Aspects of community building and tutoring are especially important in
this respect, as peer and expert support are crucial for the fast handling of upcoming questions and
problems.
265
For public servants, it is especially important to consider their work reality, often marked by
constant contact with the public and/or colleagues, which makes longer, concentrated learning
periods difficult. In addition, the exchange with other employees, also from other communities, is
an important requirement in the modernization process.
When creating on-demand training for public administrations, it is thus indispensable to pay
respect to the appropriateness of the contents (e.g. focused to public servants), the course design
(e.g. optimal timing of online and offline phases), the tutoring (e.g. optimal blending of online and
offline contents) and the technical aspects (e.g. user guide). The success of its implementation then
depends on the support of the administration itself by granting fixed learning times.
The evaluation concept employed in the project is a guidance on how to ensure a holistic
evaluation of on-demand training, beyond pure usability considerations. It points to the importance
of noticing how learners make use of the training, what they really learn and how they apply this in
their practical work. Together with the concept, possible means of evaluation such as interviews,
surveys and observations were presented.
As general advantages of the prodela on-demand training in contrast to e-learning approaches it can
be stressed that:
§ prodela provides a better integration of the learning into the work process;
§ it is a demand-driven approach, as it provides ad hoc contents in the situation they are needed
in;
§ it strengthens community aspects and merges the learner-teacher roles, as everyone can
provide input to the training contents (all learners can also be teachers).
An important emerging issue for further exploration is fast supply with up-to-date contents. In
public administrations, there are many changes in regulations and templates that would need further
instructions beyond the mere information on them. It would thus be very useful to develop a
concept for the fast transfer of this information into short e-learning modules providing a more
holistic information on the expected changes and their implications.
•
Always pay respect to the specifics of the target group when designing ondemand support.
•
On-demand training should provide ad hoc support in the work process, which
not only impacts the design of the contents, but also the support structures
(peer/expert support); community building is essential.
•
It is crucial to provide guidelines for the involved trainers and users, such as
style guides and instructions, to make them acquainted with the concept.
•
A holistic evaluation of the success of an on-demand training initiative should
consider various levels, from the design of the concept to its acceptance to the
actual learning and its transfer into real-life application.
Acknowledgements
We especially want to thank our colleagues Dr. Gerd Gidion and Dipl.-Psych. Kathrin Schnalzer,
who collaborated with us in the prodela project. In addition, specific acknowledgements go to the
other organizations involved in the project, Hyperwave, memoray, Tellur and Wirtschaftsförderung
Region Stuttgart (WRS). We also want to thank all public servants, who participated in interviews,
surveys and work/learning observations and thus helped us to get the necessary feed-back for
optimizing the concept and deriving scientific insights.
266
References
Bundesministerium des Inneren (2002) Der öffentliche Dienst in Deutschland, Bundesministerium
des Inneren, http://www.staat-modern.de/Anlage/original_548856/Der-oeffentliche-Dienst-inDeutschland.pdf, pg. 6, 25 (date accessed: 18 October 2005).
Kommune21 (2005) Modernisierungsstrategien der Kommunen.
http://www.kommune21.de/meldung.php?id=5031n (date accessed: 18 October, 2005).
Mandl, H., Gruber, H., and Renkl, A. (1997). Situiertes Lernen in multimedialen Lernumgebungen.
In: L. J. Issing and P. Klimsa (eds.). Informationen und Lernen mit Multimedia, Beltz, ISBN:
3621274499, pp. 167-177.
Schenkel, P. et al. (2000). Qualitätsbeurteilung multimedialer Lern- und Informationssysteme.
Evaluationsmethoden auf dem Prüfstand, BW Bildung und Wissen Verlag und Software
GmbH, ISBN: 3821470208.
Städtetag Nordrhein-Westfalen (2005) Engere Zusammenarbeit zwischen den Kommunen eröffnet
große
Chancen
für
effektivere
Verwaltung,
http://www.staedtetagnrw.de/schlagz/archiv/2003/s_20030604.htm (date accessed: 1 November, 2005).
Liza Wohlfart studied Economics, English and French at the University of Stuttgart,
Germany, where she graduated as an M.A. in 2002. Since her degree, she has
been working at the IAT, University of Stuttgart, in several industrial and research
projects. She has strong experience in leading and collaborating in national and
EC-funded projects in the area of business development, corporate learning and
knowledge management.
Dipl. Wirt.-Ing. Dorothee Frielingsdorf is a senior researcher and consultant. She
has eight years of experience in research and industrial consultancy in national and
international projects ranging from the implementation of e-learning and business
TV approaches in companies to innovation management and new product
development.
267
15
Building KM @ Patni
Shashi Kadapa
Building KM @Patni
Shashi Kadapa, Patni Computer Systems Ltd. ([email protected] & [email protected])
Abstract
The Knowledge Management Initiative (KMI) at Patni is a show case
example of how to build and sustain a focused KM solution to meet
the challenge of generating knowledge value for a highly mobile global
10,000+ workforce, serving Fortune 1000 customers. With increased
high end work the KMI solution needed to be highly agile and allow
Patni workforce to leverage legacy learning’
s. This case study brings
into focus the salient issues and features of the Patni KMI solution. It
explains in details how business drivers were harnessed to form the
knowledge strategy and how community learning’
s is fostered to bring
mass movement that aligns itself with the organizations goals.
Keywords: knowledge management, CoP, contributions, architecture, KMI brand building
Background
Patni Computer Systems Ltd. (Patni) is a global IT services provider with revenues of US$ 326.6
million in 2004. The organization was founded in 1976 by its US based CEO, Naren Patni, an MIT
and Sloans School of Management alumni.
Committed to quality, Patni adds value to its client's businesses through well-established and
structured methodologies, tools and techniques. Patni is an ISO 9001: 2000 certified and SEICMMi Level 5 organization, assessed enterprise wide at P-CMM Level 3. In keeping with its focus
on continuous process improvements, Patni adopts Six Sigma practices as an integral part of its
quality and process frameworks.
This high level of interaction generates a large amount of information and data. The Patni KMI
helps to turn this information and latent data into Knowledge that can be reused and applied to
reduce through put time and shorten the learning curve.
Industrial Context
Our Customers
Patni has about three decades of IT presence serving blue chip organizations that operate in diverse
industry segments. The portfolio covers more than 180 clients including leading names like General
Electric, Hitachi, NBC, Royal Bank Scotland, Metlife, Southern California Edison, Electrolux, Le
Meridian, Schindler, BaaN. Over 10,000 professionals service clients from 24 international offices
across the Americas, Europe and Asia-Pacific, and eight offshore development centers. Patni has
registered a compounded annual growth of 34% in the last four years and is ranked among the “top
10 best companies to work for in India”.
271
Building KM @ Patni
Our Competitors
Patni is one of the leading IT services providers and along with others, provides software services to
companies across the globe. The market is quite large and companies have to prove themselves on
fronts like technology, service and customer satisfaction. When bids are floated by the clients,
interested companies apply and the companies that have consistently excelled and those who have
proven expertise, bag the order. It is a question of how fast resources can be deployed and how
effectively existing knowledge can be reused. All the top service providers have a knowledge center
that helps to accelerate deliveries.
The worldwide software services are estimated at a few billion dollars and there is ample scope for
all to grow. The market witnesses healthy competition and each company operates in its own space.
How We are Organized
Patni is organized into various Strategic Business Units (SBU) and Centers of Excellence (CoEs). It
caters to its clients through its industry-focused practices, including insurance, manufacturing,
financial services, telecommunications, and its technology-focused practices like Product
Engineering Services and Independent Software Vendors. The service offerings include application
development, application maintenance and support, packaged software implementation,
infrastructure management services, product engineering services, business process outsourcing and
quality assurance services.
The CoEs at Patni keep pace with the latest development in the IT world and provide service for
practices like business intelligence and data warehousing, IT Governance, Verification &
Validation.
In addition, Patni has a number of essential functional units like Quality and Delivery Innovation
(QDI), Knowledge Management (KM), Six Sigma, Corporate Management Information Systems
(CORMIS). These units provide metrics framework, productivity improvement, competency
development and roll out of best practices & processes across business units.
Problem
Patni has thousands of man-years learning’s covering almost the whole IT spectrum. The skills and
learning’s were accrued by development and marketing teams from different SBUs, spread across
multiple cities in India and regions like US, Europe and Asia Oacific.
At Patni, passion to share Knowledge was always there and without a formal initiative there were a
number of active isolated repositories maintained by different SBUs. While valuable documents
were available in the repositories, they were static and used by a few employees.
Hundreds of applications and technologies like Java, .Net, Mainframes and specialized applications
like SAP, Embedded Systems and many more are used. While the tools and application used across
many projects are the same, the solutions delivered differ in requirements, functionality and intent.
Teams that had managed to handle a technical issue had no means to share the learning throughout
the organization or could only share it in a closed loop. Other teams from different SBUs, handling
similar problems would in all probability spend time trying to solve the same problem. ‘Reinventing
the wheel’did happen a few times.
The SBUs operate within their expertise areas and interaction and exchange of knowledge was
occasional. Besides there is the issue of IPR and copyrights and core information, software
components could not be shared. This lead to more estimated efforts and lower productivity.
Drafting proposals is another area where knowledge sharing is crucial. Capabilities, expertise and
projects completed in a specific area are illustrated through case studies. While responding to RFP,
272
it helps if case studies that illustrate past experience in handling similar requirements are used as
supporting documents. Collating case studies and other proposal collaterals was an issue since the
required documents resided in isolated pools. Consequently the won rate of proposals needed to be
improved.
A mechanism was needed to make these barriers permeable and allow knowledge to flow smoothly
while respecting the IPR issues. The Knowledge Center has addressed this issue very satisfactorily.
This case study will provide information on:
•
Getting people to contribute to KMI
•
Harnessing various drivers like market, people, technology and others to make
KMI a success
•
Desired features of KMI
•
Process flow for information
•
How to involve other employees to create a knowledge army
•
How to create dynamic knowledge generation mechanisms
•
How to build a taxonomy
•
How to build the KMI brand in an organization
Approach
At Patni, passion to share Knowledge was always there and without a formal initiative there were a
number of active isolated repositories maintained by different SBUs. While valuable documents
were available in the repositories, they could not be harnessed effectively to address the new market
demands.
Aggressive growth plans mean substantial investment in improving Sales and Delivery capabilities.
Starting from a small base of employees based at its Mumbai development centre, Patni now has
development centers in major Indian cities like Chennai, Gandhinagar, Pune, Bangalore,
Hyderabad, Navi Mumbai and NOIDA (near Delhi). Earlier there were about a thousand employees
and now there are now more than 11, 000.
The projects executed encompass a variety of technologies and domains. Software applications
developed for the customers are typically an amalgamation of multiple technologies. As a result,
projects using similar technologies and tools like .NET, mainframes, Oracle, among others, are
found across all SBUs. Additionally, a customer focused ODC would be interested in adding value
by replicating processes followed by the customer, whereas an SBU with a vertical focus (say,
Insurance Business Unit) would be interested in developing and retaining domain and
implementation expertise within its ranks.
The SBU structure and customer-specific ODCs have need for having their own Intranets for IPR
protection and security reasons. But these intranets tend to become “islands of information”, over a
period of time. The key KMI inducted persons in such intranets; upload what can be shared across
Patni, thereby bridging these islands of information.
273
Building KM @ Patni
People Drivers
Patni workforce is the most important asset. There are two kinds of people, those who have put in
some years with Patni and new recruits. People who have been working with Patni for some time
have a good understanding of their expectations and duties. The need is to keep them updated about
the latest developments in their areas of interest, give information related to metrics, QMS, new
projects and other business development.
Patni had a 37% increase in the head count over the past year and this includes experienced
professionals and trainees with no experience. These new members of the Patni family need
accelerated learning and adapt to the Patni culture quickly. Fresh software trainees are given
through practical training in different software languages through the famous ‘Bootcamp’. Senior
level people need information related to customer, business, methods and RoI models. They need to
be given qualified information that speaks about how technology is applied in Patni and how
customers are serviced.
Patni KMI sends an automated welcome mail to all new employees that explains succinctly the
KMI role and mission and what they can find in the knowledge center and more important, how
they can apply this knowledge in their projects. The mail includes a link that opens the KMI portal
invariably people click the link to see what this is all about. Once they are there a user friendly and
uncluttered interface allows them to visit different sections and see the knowledge that is available.
Some document types like case studies, proposals and others are restricted to people with senior
designations.
Market and Technology Drivers
Patni marketing teams work with clients in US, Europe, Asia Pacific and have a good understanding
of the market trends, developments and opportunities. The technology cells analyze these
opportunities and a techno-commercial scenario is forecast.
Through a diligent learning process, the market drivers are worked on, competencies forged and the
drivers are metamorphosed into capabilities that can be offered as services. KMI interacts with the
technology cells and a sublime documentation set of learning’s called kits is created and a special
space for them is provided in the KMI portal.
The kit includes case studies, approach notes, estimation and requirement analysis studies, proof-ofconcept documents, details of other technologies used, configuration and deployment methods,
contact details of the key personnel, etc. Future projects that need to operate with these technologies
now have a ready base of ‘qualified’information that they can use to shorten their learning curve.
This epitomizes the KMI slogan ‘Building Knowledge, Creating Value’. It also endorses the main
theme ‘build once, reuse many times’.
It takes dedication and grit but the KMI team does it with élan. Kits have been prepared for
technologies like Embedded, SAP, RFID, V&V and many more. The reaction time to identify an
opportunity, address it through proposals, service it through projects and finally deploy solutions is
considerably reduced by these kits.
274
Case Analysis
KMI Features
This section provides details about the KMI architecture and the KMI features.
KMI Features
The core features of the KMI solution:
•
One Stop Source – The main goal is to make the Patni KMI a single stop source for the
collective Patni learning’s that run into thousands of person years. KMI is used by
individuals to train themselves, by projects to shorten the delivery time and by the BU think
tank to frame the three year growth plan
•
Knowledge for individuals - To make learned decisions by offering them an optimum
knowledge base. It may be as complex as using a .jar or a .ini file for coding to something as
simple as giving a good orientation to employees who are going on onsite assignments
•
Pincer Knowledge Generation Mechanism - To keep knowledge fresh, it needs to be
captured, applied, organized and most important disbursed to the right people at the right
time. The KM team has come up with different mechanisms to capture current knowledge
and disburse it effectively. Mechanisms like case studies, project summary and completion
reports, relationships with different clients and others, capture distilled project knowledge in
all phases of the project. The discussion forum provides invaluable tips and tricks that glean
and store in KMI.
•
Blending with the organization’s culture – Patni’s highly mobile work force keeps
moving from projects and locations, taking their latent learning’s with them. Effective
mechanisms are in place that encourage and reward team members to document their
learning’s and share them on the KMI portal. Mechanisms like knowledge sharing sessions
in which team members who have developed competencies on different topics, give
informal presentations through the Community of Practice (CoP) in which like minded
people with interests in specific technologies come together and discuss current issues,
relevant to Patni
•
Focussed marketing - KMI offers a ready bouquet of qualified case studies, project
summary reports, learning’s in different technologies, proposals and many more, arranged
with keywords and as per the KMI taxonomy. When developers need tips to overcome a
technical problem, they use the KMI search engine. When proposal owners need to submit a
proposal, they can browse through the past proposals, pick up the supporting documents and
quickly submit the proposal. Focussed marketing ensures that each employee knows about
the latest contributions
•
IPR - A very crucial aspect is protecting the customers’IP rights. As a matter of policy,
Patni does not disclose or share even internally IP artefacts like source code, reusable
components, confidential or proprietary data that the client has given, among different
SBUs. This does hinder the process of knowledge sharing but protecting clients’IP is very
important.
•
Cutting Edge Paradigms –CoP –KMI has initiated a number of communities of practices
like Java, .Net and Mainframes where people who use these technologies and applications
meet and share best practices. Patni stalwarts give presentations on various aspects of a
technology. People at other locations join in through web casts.
275
Building KM @ Patni
Figure 1. KMI Features
A central core KM team mentored by the Patni senior thought leaders nurture the KMI solution.
KMI performance and strategy for growth is continuously analyzed, measured, and improved.
The KMI solution is not just the KMI portal and the core team but blends the mindset of the entire Patni
delivery engine. It includes all the processes and work drivers, HR and administration departments, sales
and marketing teams, front office personnel and the top officers of Patni.
The KMI offering
The KMI solution at Patni is:
•
Embraced and endorsed by all SBUs who consider it as another arm
•
Scalable to quickly absorb new technologies, as they came up in the IT market
•
Mirrored throughout all worldwide Patni locations through remote cache servers
•
Blends with the company culture and moral fabric without inciting feelings of resentment
•
Driven by a robust search engine with refined and advanced search features that uses
keywords to search in the metadata and content
•
Encourages knowledge generation, capture and reuse as a part of the KM process fabric
276
Figure 2. KMI Offering
Building the KMI Solution
Getting the Right Fit
To frame the requirements for the Patni KMI solution, the following issues were addressed in depth:
•
Participants – There was an intense internal debate and polling on whether to make the
KMI portal accessible to all employees or only a selected few leaders from each SBU. After
weighing the options, it was decided to make the KMI a mass movement and involve all
personnel. As a matter of prudence, sensitive documents like business proposals,
relationship documents with clients and others were restricted to employees of a certain
grade. Technology related documents like coding tips, sample code snippets, installation
guidelines, etc. were available to all employees. Considering the industry turnover rates of
young software engineers, this was thought of as a prudent move
•
Central versus decentralized model. The past experience with a decentralized model was
not pleasant. So Patni adopted the central model. The KMI solution is a designated Patni
internal project and is administered by a central core team of developers and content
management experts.
•
Knowledge artefacts. Content is the essence of the KMI portal. People visit KMI to gain
knowledge by reading documents. The repository has documents like case studies, project
summary reports, DD/FS documents, approach notes, informal presentations that are
generated in the knowledge sharing sessions, fact files, proof of concept documents, etc.
Suitable document templates were created to ensure a minimum common structure for the
documents. Documents have a validity date and authors are sent an automated mail that asks
them to send updated versions. Old documents are deleted and purged from the repository
277
Building KM @ Patni
KMI Rollout Plan
For executing external KMI projects, Patni uses a structured roll out plan for the hardware and
content related issues. A time and milestone tagged approach is used. These are illustrated in the
following image.
Figure 3 KMI Rollout Plan
KMI Taxonomy
Taxonomy is the basis on which content is organized. SBU stalwarts, technology and functional
experts, project engineers and leaders, SBU heads and other roles were polled through structured
questionnaires and face-to-face meets and firm requirements of the portal were framed.
This intense exercise crystallized into the KMI Taxonomy on which the KMI solution is built. The
taxonomy is not static but live, dynamic and scalable and allows more nodes to be added as and
when they come up. It is true to life and represents the organizations activities and business
processes. In effect it brings to life the over three decades of Patni activities into one chart and with
a few clicks, users can delve into legacy technologies and shift to the latest in .Net or embedded
applications. It allows users to reach back into history and move forward into the future.
Care is taken to ensure that each node of taxonomy has relevant documents. KMI submissions from
different SBUs are scanned and if new technologies have been used, then the source is tracked
down, qualified documents are obtained, gaps filled and published in the KMI portal. This allows
KMI to keep up with the latest developments.
278
Process Flow
KMI uses a structured information flow process where documents are qualified by taking it through
a review process. The documents are verified for technical accuracy, relevance, authenticity and
copyright issues. The process is illustrated as below:
Figure 4 KMI Process Flow
The Knowledge Army
KMI at Patni is a cooperative movement. It is enabled by people who remain the key drivers.
Without an unprecedented mass movement, the initiative would be hollow. While the core KM
technical team focuses on the technology front to ensure smooth flow of information across
different internal and external networks, the content management team, lead by a spirited project
manager and a visionary KMI manager, forays into the realms of different technologies. Individuals
are strongly encouraged to contribute the learning’s and insights they have gained to the KMI.
The core KMI team is assisted by more than 50 voluntary experts who spend extra time and efforts
in assimilating and collecting knowledge artifacts and furthering the KMI cause. They are location,
SBU and technology specific. They are also called the Knowledge Champions and help the core
team to reach out to far-flung project teams and also find suitable subject experts.
279
Building KM @ Patni
Figure 5. KMI Assets
A growing band of over 300 subject experts helps the KMI in evaluating and giving suggestions to
improve quality of the contributions. Subject experts are people who have gained sufficient
expertise in different areas like software languages, software tools, design, methodologies, etc.
After they evaluate a submission, they assign K points that contributors can accrue over the quarter
or a year. Prolific authors are awarded a quarterly “Knowledge Star”award. Subject experts and
knowledge champions who have performed excellently are also recognized and awarded.
Contests are organized for white papers, tutorials, use cases, code snippets, creating reusable
components and so on. An expert panel of reviewers selects the best documents and the winners are
given awards.
Knowledge Generation Mechanisms
Information is power but a static information is just figures and facts. But information needs to be
converted to knowledge that can be applied. KM team seeks knowledge actively and does not wait
for it to come. Experts in different domains are actively sought, persistently encouraged to write and
the KM content team members sit with them, record their thoughts and knowledge and transcribe
them into documents.
A number of innovative mechanisms are used and some of them are illustrated in the following
figure and explained in the following sections
280
Figure 6. Knowledge Generation Mechanisms
Formal mechanisms
As a part of the project completion process and the audit systems, projects and SBUs create case
studies, project summary reports, customer relationship documents, proof of concept and other
documents. These are sent to the KMI and published as per their taxonomy and keywords. Access
restrictions are also assigned so that certain documents can be seen by people with specific
designations and roles.
We Listen and Learn
Voice of the Customer (VoC) is a valuable feedback mechanism and is the pulse of the
organization. Both eulogizing bouquets and brutal brickbats are given. VoC has told the KMI what
is lacking, what should be done to improve, information gaps and the new technologies Patni is
using. All suggestions, criticisms, complements are debated and applied when feasible. VoC is
collected meetings with project teams and also through the feedback form, provided in the KMI
portal. Rather than feel righteously hurt and wait for a chance to get back, the KMI team reacts
proactively and immediately hastens to take corrective actions.
Many improvements in the way the KMI database tables are structured, faster display of pages,
secure modes, encryption sequences are organized, page layouts and many others were suggested
by irate users who faced problems.
Discussion Forum
The KMI Discussion Forum is one of the best examples of community learning and cooperation. It
provides an informal platform for people who use specific technologies to pool their learning’s and
knowledge and help each other to resolve technical showstoppers. There are more than 260 forums
on subjects like Java, .NET, COBOL, AS 400, C#, SAP, RFID and many more. Users register to the
forums of their interest and post queries and others post solutions. The turn around time is in
minutes and the community sharing spirit and drive is immeasurable. We have more than 3000
registered users and the number of threads has crossed 7000.
281
Building KM @ Patni
There are many examples when problems faced by a developer in US were solved by another
developer working in Mumbai, India. Innumerable practical solutions are given. The person who
raised the query rates the answers. Top contributors who have given the maximum points are given
the quarterly K Star award.
Main points from the threads are extracted along with the query and a put in a document. This
document serves as a FAQ and the knowledge carried there is practical and invaluable.
Inviting External Faculty
Professionals like doctors, photographers, scientists, lawyers, etc. are invited to address multiple
locations through web casts called Mindshare. The professionals speak on topics like managing
stress, avoiding heart attacks, improving photographic skills, etc. These web casts have a
considerable attendance and ensures the all round development of the employees.
Regular technical knowledge sharing sessions like K-Unlimited, are conducted and the gist is
captured and stored in the KMI portal. In such sessions, experts in different technologies share their
tacit knowledge with other employees.
Creating Knowledge Nuggets
Every week, a presentation of general interest on self improvement, sent by KMI users is featured
as a Knowledge Nugget in the KMI home page. The contributor’s name and SBU is given along
with a link for other users to view the document. This initiative is highly appreciated by the
employees as they get instant global recognition and there is long-waiting list of potential
contributors.
Express Learning is a weekly capsule that is mailed to all employees. The capsule gives
information on a variety of topics like management, data warehousing, self-improvement, etc.
These capsules are featured on KMI.
Building Learning Communities
The KMI initiates communities of practice (CoP) where people of like interests join to share their
knowledge, exchange tips and where subject experts are invited to talk about an aspect of
technology. People from other locations join in through the web casts. The learning’s are
documented and published in the KMI portal. Annual meets for the CoP members have been
organized and this is a two day gala affair. Guest speakers from other companies and Patni
employees give presentations on selected topics. The attendance is very impressive and the main
goal of knowledge generation and exchange is achieved.
Extending Brand Building to other SBUs
The Patni KMI has a considerable fan following among the Patni ranks. Other practices use the
KMI portal to announce new initiatives and to gain wide public recognition. The KMI portal is
integrated with the Patni Intranet and also provides links to dedicated and specialized SBU servers
on a restricted access basis.
Evoking Contributions from Employees
A common challenge for implementing the KMI solution is non-willingness to share information.
Any KMI solution needs individuals who contribute voluntarily and with a passion for sharing.
Hoarding information or maintaining silence does not lead anywhere. So how does the KMI
solution get people to share knowledge. This is addressed with the excellent HR policies, the wholehearted support of the top management and recognition of the top contributions. Issues related to
282
employee growth, their hopes, aims in life, desire to excel, rewards and recognition, etc. have been
suitably addressed.
Recognition is one of the best forms of encouraging knowledge flow. Prolific authors of good
documents win the quarterly K Star award and the annual K Super Star award. When young
software engineers get to see their names in the KMI hall of fame, the warm glow of contentment is
immeasurable and the KMI has a passionate convert to its legion, who would continue enriching the
knowledge base in the years to come.
The SBU heads and other senior managers present the awards in an SBU meet or on special
occasions with large gatherings. Senior members of SBU’s access the Dashboard that gives a
snapshot of KMI usage by their teams. This nurtures a healthy competition amongst SBUs.
Building the KMI Brand
Building the KMI brand was indeed a very challenging task. It took a lot of creative ideas and
intense advertisement campaigns and today, the KMI solution is the best known brand in Patni and
people from different Patni locations in various cities in India and across the continent instantly
recognize it. This section illustrates the efforts we took to create the KMI brand. All the
advertisements given in this case study have been created and run by the author:
Challenges
The following tasks illustrate how difficult the challenge was:
•
Patni onsite teams operate on a 24x7 basis across the world. Our development teams are
spread across multiple SBUs in different Indian cities
•
The project pressure is very intense and people do not have the time to indulge in other
activities
•
Contributions have to be voluntary. So the Ad campaigns have to be eye catching enough to
move the people to work extra and contribute
•
The bandwidth available for mailers has restriction due to server space. So, the Ads had to
be small, move people and keep them asking for more. Patni KMI works under a tight
budget and splurging on expensive gifts is frowned up on.
Advertisement Campaigns
To build the KMI brand, we launch periodic contests for web designs, punch line competitions, as
shown in Table 1.
283
Building KM @ Patni
Table 1. KMI Advertisements
Quarterly K Star
Award Winners
A Flash movie with
good animations is
mailed
to
all
employees giving
the winners bio and
picture
giving
winners a lot of
publicity.
Flash Movies Sequence created
for Six Sigma week at KMI
We celebrate the KMI birthday in a big way. A series of
Flash movies, posters, Quiz and other events are run in all
Patni Centers
We used Obelix to launch a KMI module. This is one of
the Flash movies in a series
Results
The Patni KMI solution has over 17,000 knowledge assets that translate to a ratio of 1.7 documents
per employee. It has received over 1.3 million hits since the initiative started about two and half
years ago. The KPI index and contributions per employee shows an increasing trend.
Systems are in place that track how knowledge and artifacts are used and the extent to which efforts
and time to deliver projects have been reduced. KMI has initiated a high level of cross resource
sharing and cooperation between the SBUs, both through informal and formal means.
The KMI solution has consistently received a rating of 4.4 (out of 5) in the internal customer
satisfaction survey. This means that the customers are fully satisfied and that KMI has exceeded
expectations.
The Patni KMI solution was awarded the Best KMI Solutions Award at the Livelink Up
conference, 2003, in Paris. The judges comment read, “Knowledge Management marketed to the
users and not force fed or left to the committed few". Given below is the picture of the award.
284
Figure 7. Inukshuk figurine made of stone that was awarded
Business Impacts
There have been many-sided impacts since KMI was implemented:
•
Technical showstoppers have reduced and lesser time is needed to solve such issues.
•
Proposal creation and submission that used to take days has now been shortened to a few
hours
•
Knowledge transition is much smoother and employees don’t spend hours trying to find out
information. It is on KMI is the common refrain.
•
Number of efforts and rework has reduced. This is reflected in the quarterly SBU metrics
study analysis
•
SBU knowledge barriers have been done away with.
Conclusions
While KMI in Patni has become very popular, it still needs to have increased document hits and
contributions that speak of technology as applied in the projects. Issues of acceptance and making
employees contribute have been successfully tackled.
285
Building KM @ Patni
It is essential to make the KMI an integral part of the organization. Core technical data and
implementation details should be documented in a systematic way by the end users. Quality and
document usability are critical.
•
Individuals need to be motivated enough to spend extra time to contribute
learning’
s and their tacit/implicit knowledge
•
Given a conducive environment, people take pride in sharing their learning’
s
•
The KMI solution should not attempt to change the organizations fabric but
blend with the it
•
Fresh recruits are most susceptible to a warm welcome
•
Feedbacks and suggestions from users are most important. Act on the
suggestions and more important implement nag give due credit where needed
•
Involvement of seniors is very important
•
Usability engineering plays a very important role in customer satisfaction.
People should be able to find what they want quickly and easily
Acknowledgements
The author wishes to express his thanks to Pradeep Waychal, Vice President, Quality and Delivery
Innovation at Patni who mentored this work and also Rajiv Deo, Manager, KMI and Harshita Goyal, Project
Manager, KMI for their help.
Author Biography
Shashi Kadapa nurtures the Patni KMI and helps to drive and sustain it throughout
the organization. He also looks after the brand building activity and has created
many Ad campaigns to make the initiative popular. He has as a Bachelor of
Engineering degree in Manufacturing and an MBA in Marketing. He has worked in
the IT industry for about 7 years and has also worked in manufacturing dept at
Cummins India. Shashi has also worked as a Sales Manager with companies like
Balzers AG, Carborundum Universal and Ashok Iron Works.
A person of many talents, he has a flair for creative writing and has also worked as a
journalist with many reputed Indian print and on line companies. He lives with his
wife and two daughters in Pune, India.
286
16
Enterprise
Gary Cairns
Enterprise
Gary Cairns, AMEC NNC Limited ([email protected])
Abstract
In the UK nuclear industry, and many other sectors where safety is a
major issue, it is very important to provide evidence to the regulator
that staff are suitably qualified and experienced for the work they carry
out. Also, the international standard, ISO 9001:2000 requires that
staff skills and competencies are determined, evaluated and effective
training provided.
Many organisations have a poor understanding of the skills,
experience, and qualifications etc. that are available in their staff.
Some organisations use a database of CV’
s or provide staff profiles
via their intranets. This paper describes how a bespoke enterprise
solution is provided to assist in identification and communication of
skills & competencies using a structured database approach based on
web technologies.
Following a two-year study and implementation programme, a
competency based taxonomy for identifying qualifications, skills and
experience has been successfully created and implemented across
the enterprise with details of over 1000 staff captured. The solution
known as “QuEST” (Qualifications and Experience System Tool) is
regarded as one of the organisation’
s key knowledge management
tools allowing particular expertise to be quickly located.
Keywords: Skills, qualifications, experience, competency, taxonomy, knowledge capture,
knowledge sharing, enterprise solution, identifying knowledge, SQEP, QuEST
Background
Industrial Context
AMEC NNC and other companies that make up the British nuclear industry have been engaged in
the design, build and commissioning of civil nuclear power stations for over 40 years. The UK
nuclear industry provides about 22% of the UK’s electricity needs, employs directly and indirectly
about 55,000 highly qualified people, and earns the UK about £500 million per year from overseas
business.
The UK has 23 operating commercial nuclear power reactors, of three different designs. The
comprise: 8 Magnox (gas-cooled reactors), 14 AGRs (Advanced Gas-cooled Reactors) and one
PWR (Pressurised Water Reactor). Britain’s oldest reactor, and the world’s first commercial nuclear
power station, Calder Hall, a Magnox-type station, has only recently closed, having exceeded 40
years of life. Britain’s newest station, and only PWR, Sizewell B, started commercial operation in
1996. At the present time the UK focus is on safely operating the newer stations and
289
Identifying Key Skills & Competencies Across the Enterprise
decommissioning the older stations but given climate and other issues relating to the use of oil, gas
and coal there is a possibility that a new build programme may be initiated.
AMEC NNC operates in the nuclear and defence sectors where compliance with public and
personal safety issues is regarded as a key ingredient to all activities undertaken from concept
design through to construction, operation and eventual de-commissioning of nuclear facilities. In
the UK nuclear sector the independent government regulator is the Nuclear Installations
Inspectorate (NII) and as part of the licensing conditions for nuclear sites, it is a requirement to
demonstrate that key operating and supporting staff are “SQEP” (Suitably Qualified and
Experienced People). The SQEP requirement has a similar impact on the supply chain supporting
nuclear operations at site and suppliers are also required to demonstrate that its staff are suitably
qualified and experienced. This fundamental requirement from the NII is one of the main drivers to
establish a formal system as described in this paper.
Another more general requirement is for those organisations who are seeking accreditation to the
international standard ISO 9001:2000 (ref 1). Section 6.2 of ref 1 outlines the requirements for
education, training, skills and experience, which require organisational commitment. The AMEC
NNC Qualifications and Experience System Tool (QuEST) was developed specifically to address
both areas.
Problem
This case study describes AMEC-NNC’s approach in finding a system and process for capturing,
assessing and interrogating qualifications, skills and experience of its entire staff across the
enterprise. Such a system should be capable of satisfying the compliance requirements as outlined
above for SQEP staff and also ISO 9001:2000 requirements. Additionally, the system should be a
key knowledge repository for identifying skills and competencies for general use in the business, in
particular as a tool for managers to identify suitable candidates for projects when they arise.
The specific business problems that the case study addresses are listed below:
1. To identify and capture qualifications, skills and experience for all staff in all locations
worldwide to demonstrate staff are competent in support of activities in the nuclear sector.
2. To provide evidence that staff are competent and experienced in the skill areas claimed.
3. To share information about staff skills and competencies for use throughout the organisation
in all locations.
4. To integrate skills information with other related issues such as training and development,
CV’s, academic and professional qualifications.
5. To provide a process and a system that is easy to use by all staff requiring minimal training.
AMEC NNC’s initial strategy was to locate a system “off the shelf”that would do the above but
market research showed that there was no such system available that would meet all the necessary
requirements. A development programme was thus established with a team leader and a small
number of support staff to design and implement a process and system from first principles.
290
The reader of this case study can expect to learn:
•
How to create an enterprise system to identify staff with particular
qualifications, skills & experience.
•
How to create a taxonomy of skills and competencies and build a database for
enterprise sharing of this information.
•
How to demonstrate and record, via assessment, that these competencies are
genuine
•
How to apply change management principles to gain “
buy-in”from all staff
involved in the implementation and roll out.
Approach
The need and business issues relating to the implementation of an enterprise wide qualifications,
skills and experience system tool are described in the previous sections. AMEC NNC’s approach to
the implementation of such a system began in January 2001 when the Company’s technical director
was nominated as sponsor for the project. A project team was established with a structure outlined
below in figure 1.
Project Sponsor
Phase Review Team
Business Rep 1
Business Rep 2
Business Rep 3
Business Rep 4
Project Manager
IT Development Team
Database Developer
Web Developer
Process Development Team
Business process analyst
QA support engineer
Figure 1: Project Team Structure
The roles of the team members were as follows:
•
Project Sponsor – To liase with members of the Executive to ensure adequate funding,
benefit realisation and to co-ordinate communication activities throughout the organisation.
•
Project Manager –To plan, control resource and ensure that the project was run to budget,
quality and time constraints. Also to liase with business leaders to ensure that their needs
were being adequately addressed.
•
Process Development team – To develop the concept and provide the necessary process
amendments to existing procedures. Also when implemented, to audit and gather feedback
on the implementation.
291
•
IT Development team – To develop the database and web front end of the technical
solution.
•
Phase Review Team – To provide an independent assessment of benefits and progress to
ensure correct alignment with the needs of the organisation.
Apart from the Project Manager, the project team were not engaged on the project full time. To
obtain early buy-in and ensure that the work would reap business benefit, members of the main
business units were involved in reviewing the project on a periodic basis (Phase Review Team).
One of the first activities undertaken in July 2001 was to produce a Technical Specification
describing the aims of the system and general requirements of what scope it would cover and what
it needed to achieve for the business. The Technical Specification is aimed at vendors as it was
assumed at this time that an off-the-shelf system would be procured and configured in-house. The
Technical Specification, therefore, contained not only typical database requirements but also details
of user interface, training, reports, data management, response times, system security, system
integrity, administration features etc. By September 2001, it was clear that there was no readymade solution for qualifications, skills and experience management and that a development
programme was needed.
Case Analysis
The implementation of AMEC NNC’s Qualifications and Experience System Tool (QuEST) is a
development and implementation project rather than a research project. The content and layout of
this section therefore reflects this type of project.
QuEST Concept Derivation
As with most development projects of this nature, the main driver for change came from senior
business managers who were struggling to meet customer requirements to demonstrate staff
competency. Initial meetings with managers suggested that a database approach was needed as
typical CV information in the organisation didn’t address the concept of demonstrating competency.
Anecdotal information suggested that competency was best measured via a mix of educational and
practical application experience gained on the job. Evidence for qualifications was relatively
straightforward to collect. Experience, however, needed a different approach requiring evidence
from personal knowledge of performance, customer feedback, written reports etc. The concept of
building a taxonomy to codify these aspects fell out naturally from discussions within the team and
due to a lack of available information elsewhere, this idea was developed over several months from
first principles.
Project Programme
An outline project programme for the implementation of QuEST is given below in table 1 with
major milestones indicated:
292
Table 1: Project Milestones
Date
January 2001
July 2001
September 2001
November 2001
December 2001
May 2002
June 2002
November 2002
December 2002
Milestone
Project agreed and sponsor and project manager assigned
Technical Specification available
Commercial products evaluated
Taxonomy Defined
Database development commenced
Database & user interface complete
User training complete
Data load from users complete
Data attributes validated
The full duration of the project from inception through to roll-out and data validation of user
information (1000 staff) was two years.
Taxonomy Development
Table 2 below defines the main areas where a taxonomy was needed to support the QuEST database
construction.
Table 2: Taxonomy Scope
Area
Technical discipline
Skill
Experience
Level of competence
Academic qualification
Professional Qualification
(Chartered status)
Client awarded qualification
Understanding of regulatory
requirements
Definition
The main technical disciples for staff (e.g. mechanical,
electrical, safety, IT, QA etc.)
Particular areas and sub areas of expertise within the technical
discipline
Industrial sectors (e.g. nuclear, defence, aerospace, automotive
etc.) and for nuclear the types of reactor system and locations
worked.
6 levels of competence for each skill from international expert
to basic knowledge
University and School qualifications for UK awarding bodies
Levels of award (e.g. CEng, fellow, member, associate member
etc. for all recognised professional institutions.
Client specific awards and accreditations.
Licensing regulations and standards experience
Fortunately much of this information was already available within the organisation but a
considerable amount of effort was required to develop the relationships between discipline, skill and
experience.
(The full taxonomy comprises 30 pages of tables and thus cannot be fully reproduced in this
document. However, the full listing can be made available on request by contacting the author.)
293
Web User Profiles & Interface
All inputs, reports, administrative actions etc. required for interfacing with QuEST is carried out
using a standard web browser. Access to the system is available to all users via the company
intranet.
In total there are 9 user interface areas or profiles, each with the following functionality:
1. Administrator – Used to set up user permissions, upload/download data into other
applications, system messages, general database admin etc.
2. Statistics – Used by managers to review incomplete user profiles, status of assessments,
number of logins etc. to help measure full roll out across the enterprise.
3. Attribute Maintenance – Used by the systems manager to set up taxonomies and data
attributes in these taxonomies.
4. Engineering/Technical Manager –Used by senior management to set up and allocate staff
to job roles.
5. Manager – Used by line managers to search entire database for any attribute, including
training & development data.
6. Training & Development Plans – Used to collate training and development plans across
the enterprise and for approval of specific training objectives.
7. Technical Supervisor – Used by line managers and senior staff to oversee the assessment
process
8. Assessments –Used by line managers and senior staff to provide an overview of the status
of the assessment process
9. Member –Main end-user interface with the system.
A typical end user will have restricted access to only his/her skills & competency details, training &
development plans and CV’s. A screen shot showing the layout of these is given below in Figure
(2) with the “experience”attributes expanded.
294
Figure 2: QuEST database –Members Area
Database Structure
The QuEST database is a SQL Server 2000 database that is designed to hold details of its member’s
competences and provide an assessment process that verifies the information entered is accurate.
All business rules are encapsulated in SQL stored procedures. Client applications access the stored
procedures through a thin VB COM layer. The following diagram illustrates the N-Tier architecture
used:
IIS
VB COM
SQL
Figure 3: QuEST N-Tier Architecture
Database tables are structured to allow the uploading of personnel data from other in-house
systems.
A database attribute structure is designed as shown in figure 4, which allows drill down
functionality against each main attribute type.
295
Register
Attribute Type
Academic Qualification
Membership of Professional Body
Skill
Experience
Client Qualification
Certification
Regulatory Requirements
Measure
Members assigned at
Attribute 1st level
Attribute 1st level
Attribute 1st level
Figure 4: Database Attribute Structure
Competency Assessment Workflow
Within AMEC NNC, a process was developed and agreed to ensure validity of the following data
provided by staff:
•
Academic qualifications (evidence from certificates obtained from awarding bodies)
•
Professional Qualifications (evidence from certificates obtained from awarding bodies)
•
Skills (evidence of significance competence from technical papers, personal knowledge,
verbal questioning, witness testimony)
•
Experience (evidence of significance competence from technical papers, personal
knowledge, verbal questioning, witness testimony)
296
In QuEST colour coding is used to distinguish between information that has been accepted (green),
not assessed (blue) or rejected (red).
To ensure that competency attributes can be assessed and validated by the relevant person, the
QuEST system has a built-in workflow to allow managers to delegate assessment where these are
needed. For example a line manager may delegate the assessment of technical skills to a technical
expert who may be better qualified to make the judgment. A permanent record is kept on the
system of the individuals who made the assessment and the assessment criteria used.
Implementation Strategy and Roll Out
To trial the system before full roll out, a small business team was chosen to take part in a pilot
project. This demonstrated the feasibility of the concept and provided valuable information on
expected timescales for full implementation across the whole organisation.
For complete Company roll out it was decided to run the project as a change management initiative
and adopt best practices such as:
•
Effective sponsorship and communication (to ensure that top-level commitment was
obtained and that all stakeholders were aware of developments. Communication was
targeted at several organisation levels with details available in corporate communication
briefs and newsletters.)
•
Education and training (Educating users of why these changes were needed together with
the basic simple training needed to interface with the system.)
•
Adoption of project management and planning techniques (All aspects of this initiative were
carefully planned with resourced project plans and full supporting documentation)
•
Risk management (Risk management techniques were deployed which identified the major
project risks together with planned mitigating actions)
•
Stakeholder buy-in analysis (An analysis was undertaken to look at particular stakeholders
and identify those who needed to be targeted to ensure buy-in. This was part of the above
risk analysis)
•
Process integration (To ensure that the new processes emerging from this initiative were
successfully integrated into the Company’s existing processes and procedures. This
involved updating engineering and HR processes found in the Company’s Quality
Procedures and providing additional guidance documentation describing the new process.)
•
Technical support (Assistance with entering and viewing information via the web interface
and the database)
•
Budget provisions for time and cost allocation (To ensure that costs were correctly
monitored and controlled)
Technical skill areas across the business were targeted first and, after this had been completed, the
system was then implemented for all support staff. In total approximately 1200 staff in various
locations in the UK, Canada, South Africa and Eastern Europe were added to the system and input
data validated.
297
Key Findings
Implementation of the QuEST system was generally regarded as a success story but like with the
introduction of any new system or processes there were a few teething troubles along the way. In
general, the system implementation and software was completed without too many problems. The
main issues centred on user acceptance as described below:
•
Time to input data. Typically 1-2 hours is needed per person to input all the relevant data
relating to qualifications and experience. Although not considered excessive, there have
been numerous occasions where individuals have claimed that they hadn’t enough time to
do this. Strong line management is needed to overcome such objections for success.
•
Non-standard data. The decision was made early in the project to pre-populate the database
with various attributes (e.g. skill types, experience areas, types of qualification etc.) and to
make these accessible to users via drop-down menus. Although relevant to most users,
occasionally there was found to be an ongoing needed to add non-standard data to allow
users to complete their profile. In the early stages of the project this proved to be a
bottleneck but with time proved to be less of a problem.
•
Database search facility. The database search facility, used by predominately by managers,
allows advanced search capability across numerous areas simultaneously. However, it was
later discovered that managers also wanted very simple search criteria such as locating all
engineers in the organisation with a mechanical engineering background. Modifications had
to be made to the search facility to allow this.
•
Taxonomy describing experience. Following extensive use of the system it was decided to
abandon the taxonomy describing experience. Due to the large diversity of experience
within the organisation it was decided to allow users to input their experience in more of a
free format nature but based on an internal guidance note for consistency.
Metrics on user input and validation were built into the administration modules of the system to
show the uptake of new users, new data entry and validation statistics. From these measurements,
we now know that over 95% of staff details have been correctly input into the system and are being
used by others (mainly line management) for knowledge sharing purposes.
Early in the project it was feared by some staff that the competence data could be misused by some
and used for poor remuneration awards, targeting redundancies etc. There were also fears regarding
the Data Protection Act. In the last 12 months however, cultural issues like this have begun to wane
and the next update of the system will allow the search facility to be used by all staff –not just line
managers.
From other feedback comments and staff surveys, it has been decided to increase the functionality
of QuEST to allow:
•
Integration with training and development systems
•
Tracking of training and development objectives at Department and individual level
•
CV information to be input and linked to skills
•
Easier searching
These improvement are planned to take place in the last quarter of 2005 and early 2006.
298
Business Impacts
The main business impact of this work was the launch of a new process and system that for the first
time captured all the qualifications, skills and experience within an organisation. Internally, it
meant that line managers and those responsible for resourcing projects could readily identify
relevant skills and with confidence allocate the right person to the right job. For the individual, it
meant that skill strengths and weaknesses were evident that allowed a structured approach to career
planning, training and development in line with the organisation’s HR policies and processes.
Externally, the main business impact was the ability to maintain business operations in the nuclear
sector. Without QuEST, or an equivalent system, it would have been impossible to satisfy the
regulatory needs of our customers and consequently a substantial amount of business would have
been lost (difficult to quantify exactly, but this could have been as high as €50m/year) resulting in a
near business collapse. Following the implementation of QuEST, our major clients had increased
confidence in our technical abilities and this has helped business growth, however, it is very
difficult to isolate the benefits against many other factors. A further spin-off from the QuEST
development work is the interest shown from clients and other organisations in purchasing the
system for their own use. AMEC NNC are currently now actively involved in marketing and
promoting QuEST as a product.
Conclusions
The introduction of a qualifications and experience database into an organisation is a key first step
in identifying “knowledge” that exists in the workforce and the level of competence that is
associated with the application of that knowledge. In most organisations, systems exist that identify
staff skills using techniques such as expert yellow pages, staff profiles or CV’s on-line but with
these systems there is no evidence of an individual’s level of competence. AMEC NNC’s QuEST
system combines the functionality of traditional approaches and provides evidence of competency.
This approach has been successful in that our clients have full confidence in the abilities of our staff
and independent assessors have confirmed our compliance with the international standard ISO
9001:2000.
In implementing a process and a system around a staff competency framework, AMEC NNC has
met stringent requirements set by the nuclear regulator and currently leads the nuclear sector in
developments in this area.
By measuring system deployment, user interaction and listening to feedback from all stakeholders
AMEC NNC is confident that QuEST is here to stay and has already begun implementing plans for
its future development.
299
•
Identifying and capturing qualifications, skills and experience of staff is an
important first step in any knowledge management strategy.
•
Validating the competency of staff and making this visible is a valuable enabler
to the business.
•
The success of running projects in the knowledge management arena is
invariably dependent on the people and the processes involved rather than the
technology.
•
Run all major knowledge management initiatives along side change
management best practice.
•
Taxonomies describing skills and competencies are important concepts in
knowledge management.
•
Always seek feedback from all stakeholders on new system deployments.
•
For this type of project, success is measured better by perception rather than
financial ROI.
Acknowledgements
Many thanks to Graham Kelly of AMEC NNC, who provided the technical development focus for the QuEST
system.
References
International Standard EN ISO 9001: 2000 –Quality Management Systems Requirements
Author Biography
Gary Cairns is employed as Head of Management Consultancy within AMEC -NNC
Limited specialising in knowledge management and change management activities. In
his current role, Gary provides management consultancy support to the nuclear,
defence and allied industries. Previously, Gary was Head of IT within AMEC-NNC and
Head of Management Systems and has been involved with knowledge management
concepts and implementation aspects for over 10 years.
300
17
Open Source CMS for a
Sports Centre
Arthur Visser
Open Source Content Management
System for a Sports Centre
Arthur Visser, I.C.E. Europe ([email protected])
Abstract
Implementation of a Content Management System (CMS) is no longer
the exclusive domain of large companies. Using a CMS may bring
considerable benefits to any organization using an interactive dynamic
website as an additional communication channel to add value to the
services provided to their customers.
Like many other Small and Medium sized Enterprises(SME), sport
centres often have no or limited access to the technical resources
required to study and analyze the potential benefits of a CMS to their
organization. Outsourcing this job is often considered (too) costly and
license fees for proprietary systems may be barriers to even consider
using a CMS.
In this case study we have analyzed the business processes in a
typical sports centre in The Netherlands. Based on this analysis,
conclusions are drawn about the usefulness of a CMS for the sports
centre, the potential benefits, feasibility and cost. As one of the criteria
was to find a 'low cost solution', the focus has been on Open Source
CMS solutions. . We build a prototype website in OpenCms 5.0 to
evaluate our findings.
Keywords: Open Source, Sports Centre, OpenCms 5.0, Evaluation Criteria CMS, Benefits CMS
for SME, Cost of Ownership
Background
Industrial Context
Content Management Systems have been around for many years and the rapid growth of the
Internet in all levels of society has made 'content management' an important element for
commercial enterprises, governmental organizations and other organizations interacting with
customers, stakeholders or others using the Internet as a communication channel. In the nineties, the
use of CMS has been almost exclusively the domain of large, often multi-national, companies and
governmental organizations. This has changed and CMS are now used by Small and Medium Sized
Enterprises (SMEs) to deliver 'content' to their customers and/or employees. SMEs often have a
limited capacity to allocate the necessary resources (money & people) to conduct the required
analysis of their business processes, evaluate CMS software and develop and implement
applications. This has opened the door for vendors to offer 'standard' CMS solutions usually based
on a limited selection of templates. By adding the company logo and colours these 'out-of-the-box'
solutions offer SMEs alternatives to ‘expensive’customized developments.
The other solution for SMEs may be found in using Open Source CMS. Using Open Source
software requires no license fees and allows customized development so that the company can use
its creativity to build a system in line with its image, philosophy and mission, without the
limitations of fixed templates. In addition it also allows companies with existing websites to
303
develop a new dynamic website using Open Source CMS with the same look and feel as their
existing site and thus remain consistent in the communication to their customers.
The sector of ‘Sports Centres’ in The Netherlands has initially been slow to adopt the new
technology made available for content delivery. This is not surprising as this sector is driven by
mostly non-technical people and the services they provide to their customers do not (did not)
require the use of technology other than purpose build exercise machines and a database for
membership management and accounting purposes. In addition many privately held sports centres
did not or do not have the resources to start with software development.
In the last 5-7 years this has been changing as the larger centres, often part of an international group
or franchise, started using the Internet as an interactive tool to communicate with their members and
provide information on health related issues. The use of electronic media has since been growing
fast. Sector related organizations, such as the ‘Nationaal Gezondheidsplan’whose mission is to
promote a healthy lifestyle are developing applications to make people more aware of their health
and the need to exercise. Not in the least thanks to the relative large investments made by the Dutch
government in public health and prevention as % of the health expenditure. In addition, or perhaps
because of this, software vendors have started to target this market and have been developing sector
specific software products. Some software vendors are specializing in software tools for sports
centres to manage e.g. membership administration, fitness programs and other applications. In other
words: content for healthy lifestyles is growing in this sector in The Netherlands and independent
sports centres are becoming aware of the necessity to keep up, communicate with their members
and deliver interactive content via the Internet. Summarized, the need to venture into the world of
content management for sports centres is driven by:
•
To need to keep up with competition and market trends
•
The pressure from own members who want ‘online’content & services
•
The need to run the sports centre, its communication and administration, as efficient as
possible to remain financially healthy
Problem
The sports centre in our case study has been growing consistently since its establishment in 1996.
A growing number of club members, staff, and activities triggered the need for better and efficient
communication between all these parties to improve overall efficiency.
As the organization became more complex, accurate, timely and correct information (exchange)
became more challenging and the existing communication channels proved to become insufficient
to do the job.
The sports centre we used in this case study had a primitive static website. This website was
considered 'very poor' by the club members which became clear in a communication survey that
was conducted in 2004 to investigate the quality of communication in general. The sports centre did
not have any in house resources or the financial means to start any major new development so the
problem it faced could best be described as:
“How to create a website that meets both the needs of the club members and of the management of
the sports centre, use it as a professional communication channel, at low cost, and, if possible,
without the need for technically skilled personnel to maintain the site.”
The key issue was therefore to come up with a solution to deal with the growing information flow
and communication requirements that could be used as an additional, mature and valuable
communication channel.
304
Our proposal was to investigate the possibility of building a new, dynamic website, based on CMS
technology. This solution could prove be capable of handling the information flow in a dynamic
interactive way and improve overall communication efficiency.
•
CMS for sports centres (SME): evaluation criteria
•
CMS for sports centres (SME): benefits and pitfalls
•
Interpreting differences between Open Source CMS and Proprietary CMS
•
Evaluating cost of ownership when considering the use of (Open Source) CMS
Approach
In order to fully understand the problem described in the previous section, and the business
processes that are related to this problem we have systematically analysed the company’s business
processes and used feedback and input from the company management, staff and club members to
get a thorough understanding of the communication requirements. The steps we have taken are
listed below:
•
Business analysis and familiarization with the help of the company management
•
Business Process analysis and definition
•
Market scan: Analysis of websites of other sports centres & related businesses
•
On line survey for club members using the (new) static website
•
Interviews with club members
•
Definition of the content requirements
•
CMS pre-selection (short list)
•
CMS evaluation & selection
•
Development & Implementation of prototype
In the next section we will summarize our actions and findings for each of these steps. For ease of
understanding, throughout this case study, some symbols are used as follows to provide/indicate:
Intermediate conclusions;
Practical tips or lessons; (+) Benefit; (-) Possible Pitfall
Case Analysis
Business Analysis
The sports centre is located in the centre of The Netherlands. It has about 1200 club members and
20 largely part-time staff. The company is in full expansion after its creation in 1996.
The sports centre offers about 18 different scheduled activities for adults and an additional 8
scheduled activities for children. To take part in these scheduled activities club membership is
required. Besides the scheduled activities the sports centre takes part in or organizes various events
and also runs short term courses. An example is a 6 weeks outdoor inline skating course in the
305
summer or a 10 weeks training in self-defence for girls. To take part in such a short term course
club membership is not required.
In addition the sports centre organized scheduled sports activities for schools, companies and
communal organizations.
Finally the sports centre invites individuals and groups/companies for one-off activities such as
parties, team building weekends and sportive social activities.
Its communication channels consisted of a primitive static website; a paper based Fitness newsletter
(every 3 months), a notice board (in the centre) and face-to-face communication.
Business Processes
In table 1 we have listed a summary of some business processes that will be affected by our
proposal to implement a dynamic, CMS based, website as communication channel. We included in
the table an indication of the expected return in terms of increased efficiency and elimination of cost
related to inaccurate, incorrect communication.
Table 1 Business processes sports centre
Main
process
Information
Personal
Performance
Membership
Activity
planning
Special
actions
Target
Business processes
Frequency of
Characteristics
change(*)
Login
Complexity
Expected
return
one-way traffic,
general information
about company/
program/ actuality /
lifestyle / health
Low
no
Low
Medium
customers
interactive / twoway traffic, specific
information /
training schedules /
progress / ask
trainer
High
yes
High
High
customers
interactive / twoway traffic
start/stop/change
membership
(details) / pricing
Low
no/yes
Low
Low
Medium
no
Medium
High
Low
no
Medium
High
customers,
stakeholders,
employees
customers,
employees
customers,
employees
interactive / twoway traffic consult
agenda/register for
activities/events
interactive / twoway traffic
participate in
actions &
promotions / activity
on request
(*) High : daily
Medium : weekly
Low : > weekly
Below we provide a short definition of each of the processes in the table:
1. Information: To provide (general) information on the sports centre, its activities, its contact
details, and other relevant company information.
306
2. Personal Performance: Relevant information exchange between the sports centre and an
individual club member about his/her performance and progress
3. Membership: Information about membership (profile/account), which can be changed and/or
updated by the club member
4. Activity Planning: Planning and scheduling of sports activities and events, allowing club
members to register and/or provide feedback and consult the planning agenda
5. Special Actions: Launch of special actions to recruit new members, to promote new
activities and to highlight specific events
These processes are generic for most sports centres. They all have a need for activity
and events planning and they need to communicate these (and changes to the
schedule) to their members in an easy and uniform way. What is interesting is that the
highest expected return comes from the personalized performance follow-up since this
represents a new marketing & communication tool.
Analyse your business processes and check where value can be added before
embarking on a CMS journey.
Market Scan
In order to get a better picture of the market in which our sports centre is operating and the kind of
website development that is going in this market we included in our research:
•
websites of other sports centres in The Netherlands, mostly in the same province;
•
websites of companies active in the health & sports market.
From the websites of other sports centres we have included in our research, most of them in the
same geographical area as they can be considered competitors, we concluded that website
development, especially when using a database driven dynamic architecture, has not yet matured in
this segment. Most websites were ’information’driven and static sites. Interactive, personalized,
communication with the visitors was not included in any of the websites we included in our
research.
The websites of companies active in the health & sports market illustrate that the health market (in
The Netherlands) is quite dynamic and that there is no lack of new initiatives to make people aware
of the importance of a healthy lifestyle. Exercise, food, and general lifestyle aspects are all covered.
These websites tend to be more sophisticated and dynamic. At the same time, software companies
are discovering a niche in providing specific software targeted at this market.
This phenomenon shows that people are interested in advice and personalized
interactive communication about their health and lifestyle, which supports our
initiative for the development of a CM-architecture for ‘local’sports centres.
Analyse your business environment (local competition & market trends) to get a feel
for the technology currently used in the market and the need to develop applications.
Online Survey
The purpose of the online survey was primarily to get feedback from club members on their
browsing experience and their expectations in terms of content requirements for the new dynamic
website we intended to develop.
A secondary target was to increase the number of visitors to the website. Since the existing website
was considered very poor and club members had virtually stopped visiting the site we decided to
307
completely renew the static website before launching the online survey for club members. The new
site offered an improved user interface, more and better information and simple interaction between
club members and the sports centre. The launch of this new site was used as a vehicle to launch the
online survey.
The idea of this intermediate step was to use the same user interface when building a site in a CMS.
This can be an important criterion for other SMEs as well since the communication style, look and
feel, should be kept as consistent as possible and according to the house style.
At the same time the new site offered the possibility to obtain valuable feedback from users about
their browsing habits, which, in this case, would have been impossible (or at least more difficult) if
the existing site was not changed (they stopped visiting).
We will not discuss all results in detail. However we do want to highlight a couple of interesting
findings. We collected 18 responses divided in 15 members and 3 non-members. The number of
members at the age of 10 years and older was at the time 1058. From an earlier research (2004) we
know that about 65% of the members rarely or frequently visited the (old) website. Among
members using the Internet, the response rate was therefore roughly 15/(0,65x1058) = 2,1%.
•
Everybody was using Internet Explorer
•
Most respondents were between 20-49 years old which follows the age distribution in the
sports centre. No respondents in the age group 50-59 years!
•
The frequency visiting the website was low on average (majority visits only once a month or
less!)
•
Most people visited the website to see the latest news or get information on the activity
schedule
•
Hardly anybody at the time of the survey visited the website to register for events / courses
•
The interest for a personalized section is clearly present for one third of the respondents
while 50% hesitate (depending on the content). Only 3 respondents had no interest at all.
•
Most respondents are interested in their personal results, personalized advice and
downloading of their (personalized) fitness programs. In addition they like to be able to
(electronically) communicate with the trainer and change their account information.
Since building a personalized section is one of the main design objectives for our CM
architecture with the highest expected return (see table 1), the interest for a
personalized section is an important result and it justified to continue development!
The overall conclusion from this survey was that it supported us in the continuation of
this project and the selection of a CM-architecture to develop a dynamic website.
Although your ideas to develop new applications on your website for members look
great on paper, do not omit to check them with the potential users before starting to
implement them. This will determine the scope of your project.
Face-to-Face Interviews
Since the development of the personalized section was at the core of the project and the choice of
our CM-architecture, we decided to follow up on the survey by inviting some of the respondents
who were clearly interested in the personalized section for an interview in order to clarify in more
detail what exactly they expected from this feature in terms of content.
308
These interviews showed that in general the user interest is focused on the online
follow up of personal progress and results as illustrated in table 2 (content
requirements).
The scope of the project has been determined by combining the survey with face-toface interviews.
Content Requirements
Based on the business process analysis, the online survey, the interviews and input from the
management we made a short list (table 2) of the content/functionality requirements for the
personalized section of the website.
Table 2 Simplified list of content/functionality requirements
Selected functions
Check Weight & Body Mass Index
Check Fat Percentage
Check Condition
Check (max) Heart Beat
Check Blood Pressure
Get Personal Advice
Download fitness programs
E-mail to trainer
Get Health report
Remark
These functions should be
included in the first release of
the website. They were also
used to test the CM-architecture
and set up some programming
examples in the prototype
The architecture should support
this functionality but
implementation was not
planned for the first but
eventually a second release of
the website
For the functions that were to be integrated in the first release of the software (see table 2) we have
defined very simple Use Cases as shown in table 3.
Table 3 Simplified Use Cases
Simplified Use Cases
Login / Logout
View progress (history)
Enter target value (objective)
Enter current value (today’s value)
Clear values (delete history)
The content & functionality requirements in table 2 refer to the personalized section of the website.
From table 1 we learn that the highest return was expected from this new development. In fact, it
would be a competitive advantage to provide club members with this functionality, which we
learned form the market scan.
However, the content requirements are of course not limited to the ones in table 2. The website
should also cover the following content and functional requirements:
•
provide (general) information
•
allow membership/account consultation and changes
•
provide schedules of activities
•
allow registration as new member (and end membership)
309
•
allow registration for events and short term courses
•
ask questions (faq section) and provide feedback
•
provide general information (electronic newsletters) on food, weight, health etc.
•
offer information on other facilities in the sports centre
•
offer information on and allow requests for ‘customized’group activities
The ultimate goal therefore is to build a CMS based website which can be linked to
the club members administration (database) so that the membership details in this
central database can be shared and used to be completed with personalized
performance data for those who are using this option.
Make sure when starting to work with databases and (sensitive) membership
information that (data) security and integrity issues need to be taken into account.
Personal information is protected by law.
CMS Pre-selection
Since we are about to select a CMS to build our website, it makes sense to first try and define what
a Content management System actually is. There are many definitions, but we’ll use one from
Wikipedia (a free online encyclopaedia, created in a CMS itself!):
“A content management system (often shortened to CMS) is a system used to organize and
facilitate collaborative content creation.”
A typical way to represent a CMS is illustrated in Figure 1.
Publishing
Content Creation
Content Management
Presentation
Figure 1 Basic representation of a CMS
The CMS facilitates content creation, which can be managed within the CMS and, after approval,
published for presentation.
When evaluating our CMS we need to consider:
•
Content –what goes into the system and what should come out!
•
People –who is using/managing the system and what do they expect/is expected of them!
•
Processes –how do we manage and maintain the system (incl. security issues, procedures)
•
Architecture –how to integrate it into the current system design
•
Future of the organization –plans, strategy, future requirements
Based on these evaluation criteria, we summarized the CMS requirements for our sports centre as
follows:
•
310
Simple workflow management: Selected staff members should be able to create content
which can be published after approval by the management/owner. Currently only the owner
provides content.
•
Integration in the existing IT infrastructure: Many sports centres use (often dedicated)
software for member administration and have a relative simple IT infrastructure. Our sports
centre only has a few PC’s, an Internet connection and no internal network.
•
Easy to use: As we are dealing with non-technical staff the CMS should be easy to use
(intuitive handling).
•
Language support: In our case English can be used (for the CMS, not the content itself!)
although support for the Dutch language would be a plus.
•
Customizing: The CMS should allow us to create our own templates, look & feel and user
interfaces. We want to be able to use the same interface as is used for the static website.
•
Security: Since the website will be database driven special attention goes out to the security
features that can be implemented to protect data and data integrity as well as authentication.
We will use login procedures and protected databases. The database will run on a different
server than the web application.
•
Others: We prefer support for multi-users, version control & archiving, content re-use
(RSS), Metadata creation, no license restrictions.
We now reached the point where we had to test various CMS and make our final. Since there are so
many CMS available these days we had to make a pre-selection first of CMS we want to examine
closer. The selected CMS were then installed to get a better understanding of the functionality and
easy of use. We made a pre-selection based on the criteria in table 4.
Table 4 Pre-selection criteria
Criterion
Low budget
Well established
IT integration
Customizable
Description
Our sports centres cannot
afford to spend a lot of money
on a CMS.
We are looking for a CMS that
is already established in the
market, can produce reference
installations and has an active
support
and
development
community.
We are in principle looking for
an OS independent CMS that
fits into existing and future
infrastructures.
Since we started with an
existing website and want to
retain a similar (but improved!)
look and feel. Customization is
a must.
Conclusion
We will try and find a solution
in the Open Source community.
Not all Open Source CMS are
in production phase. We focus
on high end solutions with
market presence only.
The CMS has to fit in the
existing
IT
infrastructure
(Windows, no network)
We are looking for a CMS that
allows easy customization to
create a similar user interface.
Since the market is flooded with CMS it is important to make a pre-selection of 3-4
products you want to evaluate in more detail, in order not to loose yourself in
information and details.
CMS Evaluation & Selection
Based on these criteria in table 4, we made a pre-selection of four different Open Source CMS.
Three software packages were downloaded and installed for testing purposes. The fourth (Midgard)
311
was not installed as it did not support a Windows environment and this was one of our requirements
(it was short listed since it is a well known and established CMS and could be used as benchmark
for the other packages). Please find an overview of our findings in table 5.
Table 5 Overview of evaluation results Open Source CMS
FEATURES
Midgard project 1.6
Plone 2.0.5.
OpenCMS 5.0.1.
Web server
Apache
Zope
Tomcat, Apache
Application server
Midgard framework PHP
All Linux/ BSD/ Unix-like
OS - no Windows!
Zope
Tomcat
Tiki CMS 1.7.5.
Apache or other PHP
enabled
PHP4.1.2+
OS independent
OS independent
OS independent
OS
Database
Core language CMS
Support community
Professional support in Benelux
Ease of set-up & installation
(1=very difficult - 5=very easy)
> remarks
installation
Dutch language support
Can be extended (own
development)
Easy to import existing
site/templates (1= very difficult 5= very easy)
> remarks about
import
Features (1=very poor ; 5=very
good)
License
User interface
Development status : production
Security issues (e.g. SSL )
Complexity for first time user
(1=very easy - 5=very complex)
MySQL
most relational
database systems
C, PHP
Good
no
Python, Zope
Very good
yes
MySQL, other SQL
db with JDBC
connector
Java
Good
yes
n/a
4
3
2
fine tune TomCat
conflict existing mySQL
MySQL; PostgreSQL
JavaScript, PHP4.2+
Very good
no
no
no
no
97% done
yes
yes
yes
yes
n/a
2
4
3
n/a
write import scripts
integrate into jsp
pages
convert to Tiki format
3
5
GNU Library or
Lesser GPL
Web based
stable
GNU Library or Lesser
GPL
Web based
stable
limited
good
2
4
GNU Library or Lesser
GPL
Web based
stable
good
n/a
GNU GPL
Web based
stable
limited
(free add-ons)
3
As we can see in table 5, some criteria (in red) were bottlenecks for our particular project. Others
(in blue) were considered a plus. This can naturally be completely different for other projects and
applies only to our specific case! The Midgard project did not support a Windows environment
which made it less useful for us. Plone was written in Python / Zope which meant a new language
had to be mastered to use it. This left us with OpenCms and Tiki.
Although its standard functionality was more limited than e.g. Tiki CMS, OpenCms
was fairly easy to use for a first time user and importing existing templates was not
too difficult which made us decide to select OpenCms as our platform for further
development.
Final selection of a CMS must depend on the individual project requirements. No two
projects are the same and therefore you need to define your own criteria. You can also
opt to build a more complex evaluation system with weighing factors, but beware for
too much detail, which makes this process unnecessary complex and time consuming.
312
Development & Implementation
The final step was to develop a prototype of the new website, including the database, and set up a
test environment. As OpenCms supports de Model-View-Controller architecture (figure 2), we used
this architecture to build our prototype.
Sqlyog.lnk
Apache Tomcat/5.0.18
Figure 2 Model-View-Controller architecture
Our experience during the development of the website illustrated that, although ‘on paper’
everything looks straight forward, the reality is often not ideal. Trying to find solutions for
programming problems in Open Source software can often be time consuming.
In this case we encountered some problems with the implementation of Java Beans and Applets. In
addition, not all JSP commands were supported (e.g. <jsp:forward>) and the implementation of the
Lucene Search Module required some changes in the setup of our Tomcat application server which
was not documented.
In other words, the fact that full compatibility with JSP is not guaranteed in OpenCms
5.0 and that not everything is documented resulted in a development phase that took
more time than initially anticipated!
In addition we discovered that the website build in version 5 of OpenCms (the version we used)
could not simply be imported in version 6 which was due for beta release at the time. This could
mean that support for version 5 from the community would slowly disappear in favour of version 6.
OpenCms also supports SSL connections. Security is a very important issue and should not be
underestimated for any website development project as we already mentioned earlier. Especially for
projects that involve a database with sensitive information. We learned through the user forums that
some of the problems that were encountered in Version 5.0 with SSL are supposedly solved in
Version 6.0 but have not verified this.
313
Developing a website in OpenCms can be time consuming due to (undocumented)
restrictions or incompatibilities and the reliance on support forums and development
communities.
Key Findings: Benefits & Pitfalls
OpenCms Software: The Product
(+)
OpenCms 5.0 is a mature and stable Open Source software. The OpenCms platform
allowed us to develop a prototype of a new dynamic website according to our
requirements.
(-)
The fact that there was no full compatibility with e.g. JSP and that not everything was
documented resulted in time consuming development.
OpenCms projects are, as Open Source software, by definition driven by the community and in that
context virtually anybody can contribute to the development. Contributions were made in the form
of core system development , such as the development of Version 6.0, or the development of
modules, such s the LDAP module or bug fixing , writing documentation and “how to ‘s” or
localize the workplace (for example in the Dutch language).
(-)
One of the major pitfalls we encountered during our development stage is that actual
development took much more time than anticipated due to a lack of documentation
and some incompatibilities (e.g. with JSP).
(-)
Another pitfall was that this project could not have been realized without a thorough
technical knowledge and programming skills, despite the fact we choose a CMS that
was, at first sight, easy to use. So, while the software license is free of charge for
OpenCms, professional support (if required) is provided by commercial companies
such as Alkacon and has to be paid for.
(-)
We also discovered that our development in Version 5.0 could not be migrated to
version 6.0. In other words there is no upwards compatibility.
In the case of OpenCms de development was driven mainly by Alkacon, although many other
companies are involved. Alkacon also provides professional support for development.
(-)
The bottom line remains however that for future development and improvement of
OpenCms depends on the ‘goodwill’of the community and as a users, you get no
guarantees on compatibility, integration or future releases
Development: The Customer
A survey among the club members which was conducted in 2004 showed that about 65% of the
members had at least once visited the website (the old one at the time).
(+/-)
For our sports centre the website is an additional communication channel to the
existing channels in the communication strategy of the company.
(-)
At this point in time we are still working with the new, but static website. The proposal
for the development of a dynamic website using OpenCms has been put to the
management/owner of the sports centre, but no final decision has been taken on the
continuation of the project.
314
Why not? Two main reasons:
1. Lack of time to study the proposal and its implications in more detail
2. For the project to be successful it is necessary that the Internet/website is
used by management and staff as an integrated part of the business.
Currently we see that the website is still very much an ‘add-on’and used on
an ad-hoc basis.
Only by providing a positive browsing experience and continuous updating the website will
increase the number of returning visitors, and thus increases the value of the channel and return on
investment. We have to conduct a second (online) survey based on the existing but completely
renew website. A third survey will be required to evaluate the dynamic website using CMS once
launched.
Development: The OpenCms Community
(+)
OpenCms has an active user & developer’s community. This was one of our selection
criteria. Our experience with the user forums (also those from the other CMS we
tested) is that they can be very helpful and in active communities you regularly get
some kind of answer fairly quickly.
(-)
Our experience in this case study is it wasn’t always easy to solve a programming
problem using the forum. It was hard to get anyone to answer a second question on
the same subject (i.e. a thread), if the first hint wasn’t helpful or properly understood.
(-)
In addition, we were always depending on the goodwill of the others and were never
sure when someone would answer. In this sense the user forum was for us a time
consuming way to find an answer to certain problems.
(-)
The community working with OpenCms was nevertheless very active and in general
helpful. However, during the last month or so, they seemed mainly occupied with the
beta release of Version 6.0 and we got the impression that Version 5.0 questions
became less of a priority.
Development: Maintenance & Hosting
Before we launched into the development of our own website based on Open Source software we
checked the availability of professional support in the Benelux (see table 5). What we omitted was
to thoroughly check the hosting options for OpenCms. It turned out that the possibilities to host our
OpenCms project are quite limited.
(-)
Since our sports centre does not have the IT infrastructure to host their own website a
third party had to be found. We found a possible hosting partner for our OpenCms
product in the Netherlands: Proteon in Delft.
OpenCms: Cost of Ownership
One of the benefits of OpenCms is the fact that the licenses are free of charge and the software can
be freely used, modified and expanded. We have not found literature in which the real cost of
ownership is calculated for OpenCms projects compared to e.g. proprietary CMS.
What we do find in literature about commercial CMS projects is that the software usually only
represents 10% of the total project cost and that development and maintenance takes up 75%!
315
(-)
In this context the bonus of not having to pay any license fees for OpenCms may in
fact look already far less attractive. The flipside is namely that there is typically no
product guarantees, support is either on a goodwill basis or has to be paid for (as in a
commercial project) and the total development time may increase due to unexpected
‘behaviour’ of the products, bugs or other functions that are not or not well
documented.
(+)
No license fees have to be paid and the software can freely be used.
We have no experience paying for professional support in this particular case study and can
therefore not quantify the (potential) impact.
Business Impact
The development of our prototype proved the feasibility to create a CMS based solution for the
sport centre using OpenCms (see above). Using the new features offered by a CMS, especially the
‘personal performance’functionality, adds value for the club members and thus to the company.
However, as we have not yet gone beyond the prototype development and continuation is pending a
decision from the management/owner of the sports centre, we cannot yet provide feedback on its
use, the way it is perceived by the customers/members and whether it helped to further increase the
number of visits to the website.
What we do know is that the renewed static website, made as an intermediate step, already receives
much more visitors than the old one. We cannot put numbers on it yet, but base this conclusion on
the feedback that the management receives and the increased number of e-mails. Members and staff
alike use the electronic forms provided and this communication channel is generally appreciated.
The FAQ section is however hardly used at all. The site is frequently used to consult schedules and
activities. As mentioned earlier, a second online survey is due to measure the impact of this site
compared to the first survey we did. The current site provides the following functionality:
•
up to date and accurate activity schedules
o using the website for publishing new schedules and planning information
•
up to date and accurate membership information
o allowing members via electronic forms to register changes in address, phone number,
e-mail etc. Also offers the option to end membership.
•
an extra channel to communicate special events and allow visitors to register
o electronic form to register for short term courses
•
an extra communication channel to promote customized group activities
•
an extra communication channel to publish (electronic) newsletters
o not only the fitness newsletter but also a new newsletter about eating habits is now
launched online.
•
Visitors can calculate their BMI (Body mass Index) online
•
The site is used to promote facilities (e.g. sauna, child care during training) and other
activities (e.g. sports massage)
Currently, as the CMS is not yet installed, we use a blog to allow the management to write and
publish ‘news’items by themselves. The introduction of OpenCms should make this blog obsolete.
316
Once the green light is given to continue development of the CMS in OpenCms Version 5.0 we can
add the following important functionality:
•
a personalized browsing experience for club members
•
content creation by the management and possibly staff members
Since we manage most other content requirements using the static website these two items will have
the biggest business impact after the introduction of OpenCms.
A personalized section should give the visitor/member a unique browsing experience.
The ability to create content without having to ask this to third parties (technical support), and thus
handle changes in schedules, activities, photos etc. independently will be a major benefit for the
sports centre.
As mentioned in the previous section, we should however remain aware that the website must
become an integrated part of the communication strategy! It cannot survive on its own and omitting
all other forms of communication in favour of the website is not the right way either.
For this case study we developed a prototype of the website using OpenCms. The business impact
in terms of club member satisfaction, reduced cost through improved process control and increased
revenues by additional online marketing is therefore not yet calculated. These are variables that can
be measured over time mainly by monitoring derived variables such as operational cost, increase in
memberships, revenues from events and other special activities and bottom line results. Club
member satisfaction can be measured by organizing surveys on a regular basis.
Conclusions
OpenCms offers a framework to build a CM-architecture that meets all the requirements we defined
sport centre. Professional support is available if this is required and also for hosting the site we
found a solution. The OpenCms community is active and new products (versions) are being
released.
However, we also concluded that developing the application in OpenCms has taken a considerable
time and required significant technical resources.
Although the end result is an application in which the content creation can be organized without the
need for any technical or html skills, the development and maintenance definitively requires those
skills. We defined earlier in this report that it would be desirable to find a solution where no
technical skills were required to maintain the site but this criterion can not be met with our solution.
The business impact looks promising in a positive sense but could not be measured yet since we
have not yet implemented the final solution.
Nevertheless, from our research and development work up until the prototype we can conclude that:
Conclusion 1:
Building this dynamic website, using OpenCms, will bring a significant added value to the sport
centre and will give it a leading edge compared to local competitors when it contains a
personalized section for members where they can monitor and update their personal progress and
results.
One of the major obstacles we identified is the lack of technical skills in our sports centre to
maintain the site which means that external professional supports remains required even though
we’ve build our application in a ‘license free’Open Source software. Add to that the fact that
further development cannot be guaranteed yet since there is no real ‘project owner’in the sports
317
centre (the management/owner is too busy with other priorities) and a successful future of the
dynamic website becomes uncertain.
Assuming the decision is made to develop a CMS application based on our first conclusion, it
leaves us to decide between the use of OpenCms (or another Open Source CMS) or a proprietary
solution. Taking into consideration that we lack technical skills in the sports centre and have no
clear project owner combined with the time consuming development phase and the fact that a
hosting partner was not easy to find, there are many reasons to opt for an ‘off-the-shelf’solution.
Conclusion 2:
Sports centres (SME) are probably better off to select a proprietary solution and outsource the
complete project. We suggest using OpenCms (or Open Source alternatives) when the proprietary
solutions cannot offer the required functionality and/or when there are specific reasons why the
sport centre does not want to rely on one particular vendor. The total cost of ownership will most
likely not be much lower by using OpenCms, despite the free license.
318
•
Analyse your business and business processes: which processes can be
optimized and made more efficient by using a dynamic database driven, CMS
based website? Determine the added value to your organisation.
•
Check what your local competitors are doing by visiting their websites. Follow
online trends in the health market
•
Verify the ‘
needs’of your members/users for new applications before starting to
develop them (such as personal progress reports). This will help you to define
the project scope
•
When developing and application involving personal information (of members),
take into consideration that this information is protected by law. You need to
take measure to secure the data and guarantee data integrity. Does your
solution covers these issues in a sufficient way?
•
Create a short list of 3-4 CMS systems to evaluate based on 4-7 key criteria,
before making a final choice.
•
Final selection of a CMS depends on individual project requirements. Build your
own evaluation matrix and use weighing factors if required.
•
Include proprietary ‘
off-the shelf’systems in your evaluation
•
Developing an OpenCms application can be time consuming due to
(undocumented) restrictions or incompatibilities with existing standards in the
market combined with a reliance on support forums and development/user
communities (goodwill). Although no license fees need to be paid, Open Source
CMS applications are not free of charge! Can you afford a relative long
development phase?
•
Think about hosting alternatives before developing an application and include
this in your selection criteria
•
Check for the availability of professional support if no in-house resources are
available as one of your selection criteria. Do not focus solely on the product
functionality itself
•
Check the type of license that comes with your Open Source CMS before you
start an application and include this in your selection criteria
•
OpenCms 5.0 is not upwards compatible with OpenCms 6.0 (migration is
possible but takes time!), so check compatibility issues where relevant.
•
Is there an active community, are there reference sites and is continuous
development visible (bug fixing, upgrades)?
•
User forums are helpful but no guarantee for solving your problem. Is there any
other source available for support?
•
Which OS and hardware is required for your solution?
•
Check other tools needed for your application (database, application server,...)
Acknowledgements
Mr. J. Van Hoolandt –KPMG
Mr. W. Steurs –KPMG
319
References
Horstmann, Cay S. & Cornell, Gary (2002) Core JAVA 2 Volume II: Advanced Features, Sun
Microsystems Press , ISBN: 0-13-092738-4
Kroenke, David M. (2004) Database Processing 9: Fundamentals, Design, and Implementation,
Pearson Education International –ISBN: 0-13-120971-X
Butcher, Matt (2004) Building Websites with OpenCMS, Packt Publishing, ISBN: 1-904811-04-3
Zeldman, Jeffrey (2003) Designing with web standards, New Riders, ISBN: 0-7357-1201-8
Veen, Jeffrey (2001) The Art & Science of Web design, New Riders, ISBN: 0-7897-2370-0
Harold, Elliotte Rusty & Means, W. Scott (2002) XML in a nutshell, O’Reilly,
ISBN: 0-596-00292-0
Internet references of interest:
Evaluated CMS Systems (table 5):
http://www.opencms.org/opencms/en/
http://www.midgard-project.org/
http://plone.org/
http://tikiwiki.org/
Open Source License information (one of the evaluation criteria):
http://www.fsf.org/
Open Source MySQL database used in or project (figure 2) :
http://www.mysql.com/
http://dev.mysql.com/doc/mysql/en/secure-connections.html (SSL secure connections)
User Forum OpenCms (used during development):
http://synyx.de/board/
http://www.opencms.org/opencms/en/development/mailinglist.html
Technical support/Hosting OpenCms:
http://www.alkacon.com/alkacon/en/
http://www.proteon.nl/proteon/opencms/nl/home/index.html
Author Biography
Arthur Visser started his career in 1987 as Product Engineer at Océ Corporate
Headquarters, Venlo, the Netherlands. Arthur has held positions in Product and
Marketing Management, and executive sales at OMRON Electronics and as General
Manager at HARTING from 1998-2003. In 2003, Arthur chose to become an
independent management consultant and started his own company I.C.E. Europe in
Brussels, Belgium. Arthur obtained a Masters degree in E-media Engineering at Groep
T Technology College in Leuven, Belgium in 2005. Arthur has in addition a B.S. degree
in airplane engineering and degrees in marketing and finance. He speaks Dutch,
English, French, and German. I.C.E. Europe focuses on Corporate Performance
Management, IT integration, Process Optimization and Electronic Media Solutions.
320
18
Virtual collaboration in
the Brazilian Mould and
Die Making Industry
Rolando Vargas Vallejos
Patricia Wolf
Virtual Collaboration in the
Brazilian Mould and Die Making
Industry
Rolando Vargas Vallejos, Universidade de Caxias do Sul ([email protected])
Patricia Wolf, Hochschule für Wirtschaft (HSW) Luzern ([email protected])
Abstract
In Brazil, several SMEs in the mould and die making industry
recognised that they could learn trough collaboration with competitors
applying the Virtual Enterprise concept. Competitors decided to join
efforts and competencies and tried to establish stable relationships
through the creation of a Virtual Organisation Breeding Environment.
One of the most important reasons for these efforts was the insight
that the collaborative dynamic of co-operation networks generates
valuable innovation. This paper describes the journey of a successful
establishment of a Brazilian mould and die Virtual Organisation
Breeding Environment.
Keywords: Virtual Enterprise, inter-company collaboration, mould and die making industry.
Background
Caxias do Sul Context
The region of Caxias do Sul, located in the South of Brazil, is considered as one of the main
Brazilian industrial clusters. The main characteristic of this cluster is that there are concentrated big,
medium and small automotive manufacturers, auto-parts suppliers, mould and die makers, and
plastic processing industries. The mould and die sector, actually, is considered as a key sector and
enloses more than two hundred enterprises, employing directly more than thousand and eight
hundred persons. Most of the mould and die enterprises of this sector are classified as small and
medium enterprises (SMEs) (Pioneiro Newspaper, Informe Comercial, 2005).
Mould and die manufacturing occupies a key position in the industrial value-added chain. The
effectiveness of this sector results from a considerable influence on the competitiveness of
production companies. According to Eversheim and Weber (2000), mould and die external
boundary conditions are:
•
high pressure of time and costs as well as
•
high quality standards caused by fierce competition, new technological developments and
lack of qualified personnel.
Internal boundary conditions are a complex production system of “one-of-a-kind”tools for a high
product spectrum that are disturbed by a high percentage of alteration orders, repair orders and rush
orders. Because of these work environment conditions, the mould and die industry companies are
always acting within a turbulent, dynamic and uncertain environment.
323
Virtual Collaboration in the Brazilian Mould and Die Making Industry
To minimise the uncertainty of this turbulent environment, specific collaboration concepts,
philosophies, techniques, methods and tools are used. One of the concepts applied successfully in
Brazil is the Virtual Enterprise concept addressed in this paper. The concept represents a new
approach concerning management and chain production processes (Mitrovic et al., 1999). It is based
on the obvious trend for manufacturing processes towards no longer being carried out only by a
single enterprise but by a virtual network. Here, each enterprise is considered to be a node that adds
some value in the production cycle of the whole network. According to the concept, a Virtual
Enterprise is formed by a number of pre-existing enterprises with some common goals come
together. These companies form an interoperable network that acts as a single organisation
(Camarinha-Matos and Afsarmanesh, 1999).
Problem
The world economy has forced changes in business processes. Organisations are adopting new
forms of work, in order to become more competitive in terms of price, quality, and time-to-market.
In this new paradigm, enterprises are establishing different collaborative relationships with
customers, suppliers and even competitors. Such relationships are based on collaboration, sharing of
resources, competencies and knowledge. Here, the companies are aiming at complementing the
resources of all organisations involved in order to make the whole network more competitive. This
kind of collaboration can be that intensive and continuous, that the companies involved form a
Virtual Organisation where certain rules for the cooperation are applied - a more stable and serious
version of a simple network.
This scenario is not different in Brazil. In the last decade Brazil has been identified as an attractive
market for the automotive manufacturers. As a result, foreign automobile industries are establishing
modern plants in different places of the Brazilian territory, resulting in the creation or consolidation
of automotive clusters.
In Brazil, the mould and die industry is going through a critical moment, presenting some
particularities. The mould and die sector requires specialised personnel as most of the mould and
die designs and manufacturing processes are based on previous experiences, sometimes without
technological innovations (Weingaertner et al., 1998). Actually, most of the mould and die used is
still imported, at least those parts considered as big (size and weight) and complex (design and
geometry), representing the goods with a more aggregated value for the production companies.
Even though, the Brazilian mould and die sector is increasing in infrastructure, technology and
knowledge in order to attend the necessities of automotive manufactures. The customer usually
decides to purchase a mould or die product considering three main factors: quality, cost and delivery
time. Besides that, another factor is emerging in importance, which is the agility in the business
negotiation process.
This work describes a research project experience related to the creation of a Virtual Organisation
Breeding Environment in the Brazilian mould and die sector. Twelve enterprises decided in 1999 to
apply the theory of emergent Collaborative Network Organisations in order to become more
competitive. The question was: How can SMEs, being competitors in the mould and die sector,
become more competitive in terms of price, quality, and time-to-market in the actual market trough
collaboration?
324
•
From this case, the reader gains insights into a real collaboration project
between competitors in the Brazilian mould and die industry.
•
The case presents major success factors as well as pitfalls for collaboration in
mostly virtual collaborative networks.
•
The description serves as a success story for similar activities even in other
industry sectors.
Approach
Nowadays, working alone means to miss opportunities. The company that does not participate in
collaborations with other companies is predestined to lose its competitiveness in the near future. For
this reason, organisations are establishing alliances, coalitions, joint ventures, and other emergent
Collaborative Network Organisations (CNO). One of these emergent CNO is the Virtual Enterprise
(VE) concept that has the potential to be applied with SMEs.
As the VE concept is relatively new, there are some open questions to be answered for its practical
application. In this paper, it is assumed that a “virtual enterprise is a temporary alliance of
enterprises that come together to share skills or core competencies and resources in order to better
respond to business opportunities, and whose cooperation is supported by computer networks”
(Camarinha-Matos and Afsarmanesh, 1999). According with the authors, networking and
cooperation are the keywords that characterize the VE concept. Despite the fact that cooperation
between enterprises is not new, the use of Information and Communication Technologies (ICT) to
support agile communication added a new value for these CNO. In the recent years, the VE
paradigm has been moved from the research domain to the business application, supported by the
recent developments in the ICT.
The VE life cycle considers four interrelated and iterative phases:
•
Creation, where the VE is configured (partners’selection, contracts negotiation, definition
of access rights and sharing level, etc.);
•
Operation, where the VE is performing its business processes in order to achieve its
common goals (requires secure data exchange mechanisms);
•
Evolution, might be necessary during the operation of a VE when it is necessary to add
and/or replace a partner, and
•
Dissolution, when the VE finishes its business processes. During these four phases the
members’interaction, exchanging information and knowledge is intensive.
In the VE life cycle, information and knowledge are used and created, new procedures are defined,
trust between people grows - in other words, value assets (tangible and intangible) are created. To
make the management of these value assets most effective and efficient - especially the intangible
ones, it is proposed to apply the theory of Knowledge Management in a Virtual Organisation
Breeding Environment (VBE). A VBE represents an association or pool of organisations that have
the potential and the will to cooperate on the base of a long-term cooperation agreement. A VBE,
being a long-term networked, presents the adequate environment for the establishment of
cooperation agreements, common infrastructures, common ontologies and especially, the
development of trust between their members, which is the necessary precondition for creating
successful VEs (Camarinha-Matos and Afsarmanesh, 2004).
When a VE is created, a Distributed Business Process (DBP) is formed that is split into Business
Processes (BPs), where each BP corresponds to an individual part of the process or of the product.
The information, procedures and knowledge created in this BP rarely will be reused along the DBP
325
again if this process is not facilitated. However, the reuse of information, procedures and
knowledge in other future VEs seems to be a good mechanism. The authors assume that by forming
a VBE and creating and using an efficient Knowledge Management system, people, companies and
future VEs will become more competitive. This becomes even more crucial as the number of
enterprises in a VBE is not limited like in a VE.
This paper focuses on a particular case study of an alliance between SMEs that are competitors.
Although the alliances with competitors seem to be risky, the value of them can be high. The
cooperating companies have the same types of products, purchasing activities, market interests,
operations, processes and technologies. Those similarities create more opportunities for the
collaboration than those among companies of different sectors. The establishment of a collaborative
network among competitors can bring more benefits than disagreements, since the objectives,
interests and limits of this cooperation are clearly defined.
Case Analysis
VIRFEBRAS Case Study
VIRFEBRAS is a VBE that resulted of a research project coordinated by the University of Caxias
do Sul (UCS), and the partnership between twelve mould and die makers, a Brazilian agency for
supporting SMEs (SEBRAE-RS), and the State of Rio Grande do Sul government. These twelve
enterprises, that are competitors, decided to take part in a research project with the purpose to learn
how to build a collaborative environment using ICT (GALELLI et al., 2001).
The main objectives of this VBE are:
•
Qualify the mould and die industries from Caxias do Sul region;
•
Learn how to work in a collaborative way, applying the VE concept;
•
Become more competitive in the national and international market.
Initially, VIRFEBRAS members adopted the VE concept proposed by Camarinha-Matos and
Afsarmanesh (1999) to establish a new collaborative form of partnership. As the concepts of VE
were new, one of the major challenges was to set up this particular VE without previous knowledge
in this area.
The collaboration model of VIRFEBRAS VBE is designed as follows: Whenever an order is
submitted to the VIRFEBRAS VBE, a VE is created, with one of the companies being the
coordinator (VE-C), and other companies being the members (VE-Ms). The VE-C takes the
responsibility on technical and legal aspects of the order. When the mould(s) and/or die(s) are
delivered to the customer and there are no more issues to deal with that order, the VE is dissolved.
This way, within the VBE several VEs may exist at the same time, with one specific company being
coordinator of one or more VEs, and member of others. It is worth noticing that every company
keeps its identity, and is also allowed to do business alone. In figure 1 are represented two VEs in
action at the same time. VE1 is formed by four enterprises, where CJN is the VE-C and Coprima,
Elite and NTC are VE-Ms and in the VE2, NTC is the VE-C and Gama and Sadel are the VE-Ms.
326
VE1
VE2
Figure 1: of the VIRFEBRAS VBE with two VE in operation
The creation of VIRFEBRAS went through several phases, namely training and education, (VBE
harmonisation in both cultural and technological aspects), technology set up, market strategy,
benchmarking, identification of shareable resources, organisational structure, and operational
issues (GALELLI et al., 2001). These phases are described in detail in following sections.
Education and Training
In the early stages, four courses by the academic consortium partners had been offered for
entrepreneurs. The first two courses (Virtual Enterprise: Creation and Dissolution, and Virtual
Enterprise: Operation) targeted general concepts on VE. The third course was about mould and die
budget. The fourth course was about Production Planning and Control (PPC). It was decided to
offer courses on budget and PPC because for the mould and die industry those topics are quite
complex and it was necessary to establish patterns for the collaborative work.
Since the beginning of the project, participants have been alerted that behavioural issues are
fundamental for the success of implementation of VEs. In order to foster development of skills
related to teamwork, trust, cooperation, and participation, a two-day course on Outdoor Training
Experience was carried out.
During the research project, a course on Technology for Mould and Die Manufacturing had been
specifically designed and held for employees of the enterprises that take part in the project.
Technology Set-up
As technology is a critical issue in the mould and die industry, entrepreneurs have a constant need
to have access to up-to-date information in this field. In order to gain insights into alternatives for
technology access, it was decided to visit mould and die shows and to pay for specific applied
327
research in areas hat where critical to all VBE partners. In November 1999, VIRFEBRAS members
travelled to Europe to visit the Euromold Show (in Frankfurt) and research institutes in Germany
and Spain. In September 2000, the group travelled to North America to visit IMTS-2000
(International Manufacturing Technology Show) in Chicago and industries in USA and Canada. In
October 2000, a one-year research project was contracted with IPT-Fraunhofer (Germany) having
financial support from SEBRAE-RS. The research objectives were defined according to common
problems for the mould and die industries.
Market Strategy
VIRFEBRAS as a VBE can offer a larger spectrum of products and services than an individual
industry can and for that reason it becomes more attractive for customers. Due to this differential,
VIRFEBRAS managed to reach important customers like the automobile industry. In September
2000, VIRFEBRAS members visited a General Motors plant located in São Paulo. In December
2000, VIRFEBRAS representatives took part in a conference held in Salvador, Bahia, which
addressed the automobile complex where Ford Motor Company installed a new plant. Up to now,
VBE representatives have been always invited to participate in that kind of activities.
Benchmarking
At the beginning of the research project, it was possible to observe that the participating companies
were used to hide information from the other competitors. As time passed by and mutual trust and
engagement into the common work has been developed, participants realised that this behaviour
should be replaced by a new one, more suitable for a collaborative environment. They started
sharing information and they learned that they usually have similar problems. The companies
involved into the project understood that by sharing solutions, every enterprise involved could
benefit from that exchange process.
After the end of the research project in 2002, the VIRFEBRAS members decided to intensify the
information exchange and sharing experiences in a systemic way, for that reason a specific
Benchmarking methodology has been developed. The VIRFEBRAS Benchmarking methodology is
divided in five phases and was developed to compare/evaluate technological resources and
processes, identifying “best and worst practices” through specific technological performance
parameters (Figure 2). Through the application of this Benchmarking methodology, it was possible
to identify the core competencies and shareable resources of each enterprise involved. Since the
application of the Benchmarking methodology, entrepreneurs and employees changed their
behaviour from the initial posture of hiding information dramatically towards a frank exposure of
even confidential technological issues (Vallejos and Gomes, 2004).
PLANNING
VIRFEBRAS
database
CREATION.
INTEGRATION
ACTION
CHECKING
Worst
Benchmark
Figure 2: Representation of the VIRFEBRAS Benchmarking methodology
328
With the Benchmarking methodology, more than 150 benchmarking analyses were obtained. The
result of this work is reflected within the organisational and technological strategies for the
VIRFEBRAS enterprises. Another important result is the behavioural change that entrepreneurs
went thorough. They noticed that they usually have, if not the same problems, at least similar ones.
Once they started sharing solutions, they got the evidence that every enterprise could benefit from
this exchange process. Nowadays, entrepreneurs usually share information; and they regularly
invite each other to visit the different companies to present its resources and internal processes.
Identification of Shareable Resources
One of the major VIRFEBRAS advantages is the fact that its members can exchange and share their
resources to the full extent. Entrepreneurs got used to share information about manufacturing
resources, which allows every enterprise to accept a customer orders knowing in advance whether
one or more partners will have the necessary machine capacity available.
During the research project, one of the main issues discussed in the group was the development of
an infrastructure to operate the VBE and procedures to handle customer orders and to share
resources. As these procedures require quick and precise communication to meet the customer
demands in terms of time and quality, the partners developed a web-based network structure called
VIRFEBRAS Information System - VIS. There are two areas that can be accessed in the VIS: The
first area is open for everybody and focuses on marketing information. The second area is only
allowed to the VIRFEBRAS partners and dedicated to the exchange of the necessary information to
run the VEs in operation. The shareable resources of each enterprise were well identified after the
implementation of the VIRFEBRAS Benchmarking methodology. During the application of these
procedures and while using the VIS, entrepreneurs noticed that a good management of information
and knowledge would be the key issue for the VBE success. For that reason VIRFEBRAS, is now
developing a Knowledge Management Program in order to speed-up the knowledge dissemination
and reuse and to increase the enterprises competitiveness.
Organizational Structure
One challenging objective of the research project was to discuss and agree on how VIRFEBRAS
should work. It is remarkable that the entrepreneurs decided to form a VBE. Creating a VBE
through the establishment of a long-term cooperation agreement makes it possible to identify and
plan actions to improve the competitiveness of enterprises and future VEs. In that sense, also
specific research or consulting projects can be developed based on a common need. Those projects
can be conducted in cooperation with different external partners like research centres, universities,
consultancies companies, etc.. Each project has an assigned leader, who has the responsibility to
guarantee the project success. The VBE members decided for that kind of hierarchical structure in
order to guarantee efficient VBE operation. The overall VIRFEBRAS VBE organisational structure
shows a directory board composed by a president, a vice-president and a financial responsible.
One important VBE strategy is the search of business opportunities. This search is usually done by
a Broker. The Broker has also the responsibility to support the VE creation and the search of
potential future partners. For that reason, the broker needs to know precise information, abilities and
competencies of the VBE members. The Broker will identify the necessity of additional
competencies in order to serve potential future customers. In the VIRFEBRAS context, the Broker
as a person doesn’t exist. The functionality of the Broker is taken over by each entrepreneur when a
business opportunity appears.
The VIRFEBRAS VBE developed a statute and ethical rules. The VBE statute covers entrance
criteria for companies, members responsibilities, etc.. In the ethical rules are defined the
behavioural aspects within the network, members and employees’sanctions, etc.. Those documents
are reviewed periodically.
329
The VIRFEBRAS Information System (VIS) is considered as a part of the VIRFEBRAS VBE
structure (Figure 3). Each member uses a particular information infrastructure that contains a
database and the appropriate software tools to manage the information about product model, mould
and die design, production capacity, available resources, administrative and business-related issues,
etc.. Each member defines the information that will be stored in the VIS and shared with the VBE.
The VIS is accessible via an ordinary web browser or through the Information System Manager.
The first option offers to the clients some marketing information and the current status of a given
order (using access rights previously assigned to each client). The second option is used only by the
VBE members. They can have access to the shared resources within the VBE and change the
information related to its own production capacity available.
Virfebras Information System
Customer_1
Customer_1
Customer_2
Customer_2
Virfebras
Site
Information
System
Manager
Virfebras
Common
Information
DB
Partner 1
Partner 1
Information
BD
PPC
Applications
Other
Applications
Partner 9
Customer_n
Customer_n
INTERNET
Partner 9
Information
BD
PPC
Applications
Other
Applications
Figure 3: General representation of VIRFEBRAS Information System configuration
Operational Issues
One of the main issues discussed with the entrepreneurs group has been the development of a series
of operational procedures to handle customer orders. The basic principle of the procedures is that a
customer order is shared among the companies and the best offer is selected to fulfil customer’s
requirements. This may result in the creation of several VEs at the same time in the VIRFEBRAS
VBE. These procedures require quick and precise communication to meet the customer demands in
terms of time and quality. For instance, one of the defined procedures for VE-C selection is when a
business opportunity appears: A meeting is organised for the budgets presentation elaborated by
each interested enterprise. The enterprise that presents the best proposal considering price and
delivery time will be the VE-C of that service. Another procedure is used when the customer wants
to take part into the decision how VIRFEBRAS VBE will assist him. Sometimes the customer
330
wants to decide which enterprises will participate in the VE, besides which enterprise should be the
VE-C.
Key Findings
VIRFEBRAS is today a “best practice” example for a successfully established VBE. The
VIRFEBRAS VBE will be documented constantly by the partners involved. Up to now, the VBE
shows like every healthy enterprise network constant improvement and evolution.
In an interview published by the Pioneiro Newspaper in March 2002, VIRFEBRAS members
affirmed that the EV theory applied to SME competitors is valid, and that through joining forces it
was possible to have access to new technologies, to share information and knowledge, to qualify
employees, to contract research projects and consultancy work. The VIRFEBRAS members are
convinced that a VBE is a collaborative network that must be in constant evolution, where the first
paradigm to be broken is trust (Pioneiro Newspaper, 2002).
Business Impacts
It is possible to conclude that through joining core competencies, SMEs, even being competitors,
can become more competitive offering a larger spectrum of products, developing new markets and
serving additional potential customers. All VIRFEBRAS VBE members grew in infrastructure,
technology, knowledge and hired more employees. During the development phase of the present
work, entrepreneurs felt the need to systematise their information and knowledge exchange.
VIRFEBRAS VBE entrepreneurs are convinced that an effective way to be more efficient and
competitive is through a systematic exchange of information and knowledge.
Conclusions
The VIRFEBRAS case shows very clearly that virtual collaboration happens in a paradox
environment: VBE member organisations need both autonomy and a feeling of belonging and
certainty within a dynamic collaborative working environment. This case study can certainly serve
as a best practice case for a big number of companies aiming at the establishment of a similar
collaboration structure.
331
•
The start up phase of a collaboration network –be it virtual or regional –
between competitors needs specific efforts for generating trust among the
network members. In the described case study an especially clever method has
been used to generate this trust: Common travels to events that have been
thematically relevant for all members. This method is definitely creating a
feeling of belonging to a group.
•
The members of a network can not be forced to share information, especially
not confidential one. They must detect and articulate the need for knowledge
exchange themselves. Only than they can develop common processes for
knowledge and information sharing and develop tools that serve their own
needs.
•
The written formulation of ethic rules and of operational procedures seems to
be one of the key success factors in networks between competitors. These
written documents support the growing of trust into the operational system
(this is another level of trust than the one between single persons based on
personal relation).
•
The organisational structure in the above described network is hierarchical and
clear on the one hand side and case based, democratic and fair on the other
hand side. The ability of the network to establish that kind of flexible but very
well- defined structures can bee seen as one of the major success factors of the
collaboration.
References
Camarinha-Matos, L. M.; Afsarmanesh, H. The virtual enterprise concept. In: Infrastructures for
Virtual Enterprises –Networking Industrial Enterprises. Kluwer Academic Publishers, 1999.
Camarinha-Matos, L. M.; Afsarmanesh, H. Collaborative Networked Organizations – A research
agenda for emerging business models. Kluwer Academic Publishers, 2004.
Eversheim, W.; Weber, P. The right strategy to success – Strategic orientation in die and mould
manufacturing. Presentation in the colloquium “Werkzeugbau mit Zukunft”. Aachen: 22.-28.
September, 2000
Pioneiro Newspaper. Matrizarias unidas crescem 30 %. Matéria publicada. Jornal Pioneiro, Caxias
do Sul, 12 mar. 2002. p. 2.
Pioneiro Newspaper. Informe Comercial – Setor de Ferramentarias. Matéria publicada. Jornal
Pioneiro, Caxias do Sul, 20 abr. 2005.
Mitrovic, D.; Hunter, I.; Male, S. Characteristics of networked enterprise in global construction. 5 th
International Conference on Concurrent Enterprising. The Hague. The Netherlands, 15-17
March, 1999.
Vallejos, R. V.; Gomes, J. de O. Applying a Benchmarking methodology to empower a Virtual
Organization. In: Camarinha-Matos, L. M. Emerging Solutions for Future Manufacturing
Systems. Springer, 2004.
Weingaertner, W. L.; Vallejos, R. V.; Gomes, J. O. Uma reflexão sobre as ferramentarias nacionais.
Revista do Plástico Industrial, São Paulo - Brasil, v. 1, n. 3, 1998.
332
Dr. Rolando Vargas Vallejos obtained his PhD in Industrial Engineering in 2005 at
the Universidade Federal de Santa Catarina (Brazil). He obtained his MSc in
Mechanical Engineering in 1992 at the Universidade Federal de Santa Catarina,
too. Since 1992, Rolando is working as a lecturer, researcher and consultant at the
Universidade de Caxias do Sul (Brazil). Currently, he is the Industrial Engineering
Course Coordinator. He worked on numerous research and consulting projects in
the areas of Mould and Die, Collaborative Networked Organizations, specially in
Virtual Enterprises.
Dr. Patricia Wolf obtained her PhD in Business Administration in 2003 at the
University of Witten-Herdecke (Germany). Whilst writing her doctor thesis on
‘
Success Measurement of Communities of Practice’
, she worked from 2000 till 2002
on a Knowledge Management Implementation Project at the DaimlerChrysler
Passenger Car Development. Currently, Patricia is working as lecturer, researcher
and consultant at the Hochschule für Wirtschaft (HSW) Luzern in Switzerland. In
the meantime, she writes her habilitation on a European- Russian collaboration
project at the ETH Zürich, Switzerland. Her research interests include Communities
of Practice and the transferability of management concepts between different
cultures. Prior to this, Patricia worked for three years as Researcher and Consultant
at the Institute of Human Factors and Technology Management at the University of
Stuttgart (Germany) worked on numerous consulting and research projects in the
areas of Knowledge Management and Innovation Management. Patricia is leading
the Zone ‘
Knowledge Management in Central Eastern Europe’ at the
KnowledgeBoard.
333
Real-Life
Edited by:
Abdul Samad Kazi & Patricia Wolf
Real-Life Knowledge Management: Lessons from the Field is about the
art and practice of knowledge management in real life settings.
Knowledge is both created and shared within a context. Based on the
hands-on experiences of the authors’where the knowledge was created
and valuable lessons learned, this book provides the context for the
sharing of this knowledge through a set of case studies spanning
several domains and areas of practice.
Through the eighteen case studies presented in this book, we learn
about practical experiences in consultancies, financial institutions, fire
fighting, global project management groups, public administrations,
research and development organisations, small and medium
enterprises, sports centres, the aerospace sector, the automotive
industry, the construction industry, the IT sector, the maritime
industry, the mould and die making industry, the nuclear industry, the
oil and gas industry, and the water management sector. The case
studies draw from a pool of experience spanning Austria, Belgium,
Brazil, Finland, Germany, India, Israel, Italy, Norway, South Africa, the
Netherlands, and the United Kingdom.
Structured around industrial experiences and communities of practice,
industry analysis and business modelling, and the design and
implementation of knowledge management solutions, each case study
features sections on industrial context, problem definition, learning
objectives, approach, case analysis, key findings, business impacts, a
set of conclusions, and most importantly, offers a valuable set of
practical tips and lessons learned.
Illustrations by Arye & Ron Dvir
ISBN 952-5004-72-4

Real-Life Knowledge Management

Transcription

Similar documents

INVITATION Phast and Safeti Seminar and Workshop

safeti advanced modelling

safeti advanced explosion modelling

PHAST™ - DNV GL

Geological models, heterogeneities and scale relations

some examples of energetic material modelling with lsdyna

Taklimat Pengukuhan Pelan Strategik Organisasi

Kingfisher Investor Event Presentation 2003 (Castorama)

Training-catalogue-Phast-Safeti-Maros-Taro

50 ideas and illustrations on the new dynamic of