Decision Analysis Today January 2016

Transcription

!"#$%&'()*'+$%,&-'(*'./0$/-1'2345'
!
'
)
"#$%$&'(
')*+%$%(
!"#$%$&'()'*+,%$%(-&.*,(
,
&!)+(
-./0(123('.0(13(4567589(:;<='
89&'0&:;#&<<&-'"='<9&'>+?@ABC'D&EF;F"0'G0/#1;F;'C"EF&<1'
Inside:
President’s Letter-----------------------1
Letter from the Editors----------------3
Upcoming Conferences----------------4
Book Announcement ------------------5
Call for Papers--------------------------6
Call for Applications--------------------7
2015 INFORMS DAS Awards --------9
Decision Analysis Journal-------------10
DA Around the World-----------------14
DA Practice------------------------------17
Society for Decision Professionals---21
Research----------------------------------23
Ask DAS---------------------------------25
Editorial Team--------------------------28
DAS Officers----------------------------29
_
President’s Letter
Bickel, J. Eric
Dear DAS Members,
I hope you had wonderful holiday season! As we begin 2016, I would like to
thank the many people that I worked with over the last year:
• Past-President Jeff Keisler has provided valuable guidance and will serve
as the Chair of the Ramsey Medal Award Committee in 2016.
• Yael Grushka-Cockayne, our Secretary/Treasurer, keeps our society
organized and on track! I greatly appreciate her help and look forward to working
with her again this year.
• Jason Merrick, our President-Elect, represented the DAS on the
INFORMS Subdivisions Council. This is an important role, which ensures that
the needs of DAS are communicated to INFORMS.
• John Butler and Casey Lichtendahl have started making plans for the 2017
Advances in Decision Analysis conference to be held in Austin!
6/7&'4'
Volume34,Number3,January2016
DecisionAnalysisToday
•
The 2015 DAS Council (Phil Beccue, Victor Jose, Frank Koch, Jun Zhuang, Joe Hahn, and
Debarun Bhattacharjya) worked very hard. This included the development of new guidelines
governing DAS awards. In addition, they also worked on several DAS initiatives, which I highlight
below:
o Phil and Joe worked to improve our membership renewal process.
o Victor and Bob Nau worked with INFORMS to develop the new DAS logo!
o Frank chaired the DAS Practice Award.
o Debarun, Melissa Kenney, and Mazen Skaf organized the DAS Track at the 2015
INFORMS Annual Meeting.
o Jun and Heather Rosoff co-edited the DAS newsletter. This is a very important and time consuming job. I thank them both very much for their efforts! In addition, I would like to
thank Jing Zhang for helping Jun and Heather.
• In addition to the individuals acknowledged above, I would like to thank the following colleagues
for their support and service to the DAS:
o Ron Howard chaired the 2015 DAS Ramsey Medal.
o Kevin McCardle chaired the 2015 DAS Publications Award.
o Canan Ulu and Robert Hammond chaired the 2015 DAS Student Paper Award.
o Jay Simon maintains our website and manages our online presence.
Several of our friends have completed their service to the DAS or will be transitioning to new roles:
• Phil Beccue and Victor Jose have completed their terms on the DAS Council and were recognized
for their service at the INFORMS 2015 meeting. Please join me one more time in thanking them
for their service!
• Jun Zhuang and Heather Rosoff are stepping down, after this current issue, as editors of DAS
Today. Their work over the last several years has been extraordinary and I am very grateful for
their service!
As announced at the INFORMS meeting, our new officers and committee members are:
• Melissa Kenney and Emanuele Borgonovo were elected to the DAS Council.
• Joe Hahn and Jun Zhuang agreed to co-chair the Membership Committee.
• Melissa Kenney and Andrea Cadenbach will co-chair the DAS track at INFORMS 2016.
• Matt Fitch will chair the DA track at the Analytics Conference in April 2016.
• Vicki Bier and John Celona will co-chair the DAS track at the 2016 INFORMS International
Meeting in Hawaii.
• Debarun Bhattacharjya and Cameron Mackenzie will co-edit DAS Today. Mavis Chen will join
DAS Today as a column editor.
• Jay Simon will continue to guide our social network presence.
• Jeff Keisler will chair the Ramsey Medal Award, Casey Lichtendahl will chair the Publications
Award, and Robert Hammond and Emanuele Borgonovo will co-chair the Student Paper Award.
We are still working through the structure of the 2016 DAS Practice Award. Once that is finalized, a chair
will be named. The DAS leadership has been active in formulating our 2016 initiatives, which I will begin
to address in the next issue of DAS Today. I wish you and DAS much happiness and success in 2016!
My warmest regards,
Eric
Page2
!"#$%&'()*'+$%,&-'(*'./0$/-1'2345'
'
!"#$%$&'()'*+,%$%(-&.*,(
Letter from the Editors
Heather Rosoff, Jun Zhuang, and Jing Zhang
Hello everyone,
We hope that you had a great holiday and we wish you a happy
New Year! With the New Year comes the first issue of the
2016 newsletter.
89-&&'&HF<"-;'IJ'<"'AK'.$0*'.F07'/0H'L&/<9&-MN'
For starters, we open with a brief photo summary of 2015 INFORMS DAS Awards. This is followed by a
summary of the 2015 December and March issues of Decision Analysis provided by Rakesh Sarin and
Kelly Kophazi. Next we travel around the world where Matthias Seifert introduces two exciting
opportunities in Brazil and Spain: the first is an announcement of the 2016 summer school in multicriteria decision aid / multiple criteria decision making in Recife, Brazil in July 2016; and the second
refers to a tenure-track faculty opening in Decision Sciences at IE Business School in Madrid, Spain.
Returning back to the United States, in the DA Practice column, Larry Neal has invited Dave Macway,
Greg Parnell, Terry Bresnick and Ralph Keeney to offer their takes on valuing metrics. Next SDP asks for
us to save the date for the DAAG 2016 Conference, scheduled for April 6-8 in Banff, Canada, and bring
your attention to a DAS and SDP jointly sponsored a webinar featuring Dr. Robert Hammond of Chevron
held on December 16th; a link to the recording is provided. This is followed by the Research Column
where Debarun Bhattacharjya gives a summary of Research Column articles from 2012-2015. This is
Debarun’s last issue as research column editor as he will be taking over as co-editor of DA Today with
Cameron Mackenzie will co-edit DAS Today. We close this issue with the Ask DAS column, where
Florian Federspiel shares the insights he obtained from a conversation with Andrea Vermehren, Lead
Social Protection Specialist at the World Bank, about the bourgeoning application of decision analytic
insights to applied problems in developmental economics.
This is the last issue that we (Heather, Jun, and Jing) will be serving as co-editors of the newsletter. We
are grateful for the support of the amazing column editors over the years. We wouldn’t have been able to
pull the newsletter together without you. We leave you in the very capable hands of Debarun and
Cameron and look forward to seeing everyone at future INFORMS/DAS events.
Enjoy the read,
Heather, Jun, and Jing
6/7&'('
Upcoming Conferences
March 8- March 11, 2016
The 12th International Conference on Operations
Research
Riviera Hotel
Havana, Cuba
http://faculty.sites.uci.edu/lrkeller/cubanoperations-research-conference-2016/
Instituto Tecnológico Autónomo de México
(ITAM) in México City
http://ifors.org/icord2016/
June 12-June 15, 2016
2016 INFORMS International Conference,
Hilton Waikoloa Village
Waikoloa Village,Hawaii, USA.
www.informs.org
April 10 - April 12, 2016
2016 INFORMS Conference on Business
Analytics & Operations Research
Hyatt Regency Grand Cypress
Orlando , Florida USA
https://www.informs.org/Attend-aConference/Conference-Calendar/2016INFORMS-Conference-on-Business-AnalyticsOperations-Research
July 3-July 6, 2016
28th European Conference on Operational
Research
Poznan University of Technology
Poznan, Poland
http://www.euro2016.poznan.pl/
November 13-November 16, 2016
INFORMS Annual Meeting 2016
Music City Center & Omni Nashville
Nashville, Tennessee, USA
http://meetings2.informs.org/wordpress/nashville
2016/
May 21-May 24, 2016
2016 Industrial and Systems Engineering
Research Conference (ISERC),
Disneyland Hotel
1150 West Magic Way
Anaheim, CA 92802
http://www.iienet2.org/Annual2/default.aspx
June 9-June 10, 2016
International Conference on OR for development,
ICORD 2016,
December 11-December 15, 2016
SRA 2016 Annual Meeting
Sheraton San Diego
San Diego, California, USA
http://www.sra.org/events/sra-2016-annualmeeting
Page4
Book Announcement
Value of Information in the Earth Sciences:
Integrating Spatial Modeling and Decision Analysis
By Jo Eidsvik, Tapan Mukerji and Debarun Bhattacharjya
Hardcover: 396 pages
Publisher: Cambridge University Press
Language: English
ISBN: 9781107040267
Web site: www.cambridge.org/9781107040267
Availability: Available from January 2016
Contains: 137 b/w illus. 22 tables
Gathering the right kind and the right amount of information is crucial for any
decision-making process. Already commonly used in medicine, economics and finance, value of
information (VOI) is becoming increasingly popular with Earth scientists.
This book presents a unified framework for assessing the value of potential data gathering schemes by
integrating spatial modeling and decision analysis, with a focus on the Earth sciences. The authors discuss
and compare the value of imperfect versus perfect information, and the value of total versus partial
information, where only subsets of the data are acquired. Concepts are illustrated using a suite of
quantitative tools from decision analysis, such as decision trees and influence diagrams, as well as models
for continuous and discrete dependent spatial variables, including Bayesian networks, Markov random
fields, Gaussian processes and multiple-point geostatistics. Numerous examples are used to illustrate the
applicability of VOI to topics such as energy, geophysics, geology, mining, and environmental science.
Real datasets and Matlab codes are also provided as online supplementary material.
Unique in its scope, this book is of interest to students, researchers, and industry professionals in the Earth
and Environmental Sciences who use applied statistics and decision analysis techniques, and particularly
to those working in petroleum, mining, and environmental geoscience.
Page5
Call for Papers: Risk Analysis Special
Issue on ‘Spatial Risk Analysis’
Risk and decision analysts, as well as policy makers, are often faced with problems having spatial
characteristics. Where to bury nuclear waste? Where to build up flood defenses? Where to prioritize
cleanup operations after disasters? How to assess the chance of finding a missing aircraft over a wide
area? How best to present geographical uncertainty to emergency managers who need to make urgent
decisions? Spatial variability is increasingly recognized as a key component in health, safety, and
environmental exposure and risk assessment. In addition, there is a broad literature on spatial decision
support in the environmental domain. A subset of this growing research field deals with risk, but,
arguably, without a proper risk-analytic framework. Presentation of spatial and geographical uncertainty is
also of key importance, with many open challenges.
The time is thus ripe to explore how to conceptualize spatial risk analysis frameworks to better support
assessment of both vulnerabilities and multidimensional impacts. We therefore invite papers for a special
issue of Risk Analysis to provide impetus in this direction. Submissions should emphasize current cuttingedge research and serve as point of reference for risk analysis researchers, as well as practitioners in fields
such as Health, Environmental Protection, Emergency Management, and Logistics.
Key areas of focus for the Special Issue include:
1.
Models for understanding and improving decisions involving spatially distributed risks
(probability distributions of multidimensional outcome distributions over a certain area and over time).
Utility (or Loss) functions for spatial outcomes.
2.
Effective communication and presentation of spatial and geographical uncertainties about
exposures, vulnerabilities, risks, and impacts, including behavioral aspects, cartographic conventions and
the use of interactive displays.
3.
Expert judgment probability elicitation and validation for spatial risk analysis.
4.
Aggregation of multidimensional risk impacts for a certain geography.
5.
Appraisal of spatial heterogeneity and heterogeneous vulnerability assessment frameworks for
spatial risk analysis.
6.
Spatiotemporal logics for interrogating GIS to support risk analyses and underpin decision support
systems
Key Dates
Page6
Submission Deadline: June 30, 2016
First-Round Reviews (target): October 1, 2016
Special Issue Published: Early 2017
We encourage you to submit your papers. The full call for papers, which includes links to detailed
instructions on house style and submission, is at: www.sra.org/sites/default/files/pdf/...
Please contact the editors if you require any further advice.
SI Editors: Nikolaos Argyris (Loughborough University, [email protected]), Valentina Ferretti
(London School of Economics, [email protected]), Simon French (University of
Warwick, [email protected]), Gilberto Montibeller (Loughborough
University, [email protected]).
Call for Applications to the Behavioural
OR Summer School
May 15-22, 2016, Finland bor.aalto.fi/summerschool.html
********Please note that the application deadline is February 20, 2016*********
We invite applications from doctoral students and advanced masters students as well as post docs
interested in this new and exciting field within OR . We also welcome people from other non-OR
backgrounds such as behavioural, social, computational and organizational sciences. Knowledge of OR is
not a prerequisite - interdisciplinarity is key in BOR work.
WHAT IS BEHAVIOURAL OPERATIONAL RESEARCH:
Behavioural operational research (BOR) is a new very rapidly developing area in operational research. It
is interested in the behavioural aspects related to model-supported problem solving and decision making,
as well as in using models to describe human behaviour in problem solving and complex settings.
BOR SUMMER SCHOOL
This international summer school is the first in this new area. Our distinquished lecturers are pioneering
professors in the field including Alberto Franco, Paul Goodwin, Konstantinos Katsikopoulos, Jeffrey
Keisler, Gilberto Montibeller, Stewart Robinson, Esa Saarinen, Ahti Salo and Jyrki Wallenius. This will
Page7
be a unique opportunity to get an early start in this emerging new field in OR. The presentations will
cover different areas of operational research including problem structuring, multicriteria decision making,
heuristics and biases, simulation, systems thinking and forecasting. There will also be talks by brain
researchers and demonstrations of new measurement techniques, including brain imaging, which are used
in studying human emotions and behaviour. Students attending the summer school will have the
possibility to describe their research projects and present posters. Students will also have the opportunity
to discuss their research ideas and collaborate with our distinguished speakers .
The summer school is organized by the Systems Analysis Laboratory in Aalto University and the BOR
community.
It will take place on the Otaniemi campus of the Aalto University near Helsinki.
General info: bor.aalto.fi/summerschool.html
How to apply: bor.aalto.fi/applications.html
BOR web site: http://bor.aalto.fi
Welcome to the BOR Summer School in Finland!
Raimo P. Hämäläinen
Professor
Page8
2015 INFORMS DAS Awards
Frank P. Ramsey Medal Award
Prof. Robin Keller (Paul Merage School of Business,
University of California, Irvine) has been named as the 2015
Ramsey Medalist!! Congratulations Robin! The 2015 selection
committee was Ron Howard (chair), Jim Dyer, Jim Matheson,
Greg Parnell, and Larry Phillips.
Picture: Ramsey Medal Award winner: L Robin Keller (R)
with DAS President Eric Bickel (L)
DA Publication Award
The winners of this year’s DA Publication Award is David
Brown and Jim Smith, "Optimal Sequential Exploration:
Bandits, Clairvoyants, and Wildcats" Operations Research 61,
3 (May-June 2013). The award selection committee was Kevin
McCardle (chair), Jim Dyer, Gordon Hazen, and Karen Jenni.
Picture: DA Publication Award winner: Jim Smith (R) with
Jim Dyer (L).
DAS Practice Award
The award selection committee Chair of the DAS 2015
Practice Award for this year was Frank Koch (left in the
picture). It is my pleasure to congratulate the winner of this
year’s Practice Award: Michael C. Runge (right in the picture),
Kirk E. Lagory, & Kendra Russell.
Student Paper Award
The winners of this year’s student paper award is: "Eliciting
and aggregating forecasts when information is shared" by Asa
Palley (Middle in the picture, Fuqua School of Business, Duke
University), coauthored with Prof. Jack Soll. The student paper
award committee included Robert Hammond (Co-Chair, Left
in the picture), Canan Ulu (Co-Chair, Right in the picture),
Sam Aflaki, Bill Klimack, Casey Lichtendahl, and Matthias
Seifert.
Page9
The Decision Analysis March 2016 Issue
http://da.journal.informs.org/content/13/1.toc
Probabilistic Warnings in National Security Crises: Pearl Harbor Revisited
David M. Blum and M. Elisabeth Paté-Cornell
Abstract: Imagine a situation where a group of adversaries is preparing an attack on the United States or
U.S. interests. An intelligence analyst has observed some signals, but the situation is rapidly changing.
The analyst faces the decision to alert a principal decision maker that an attack is imminent, or to wait
until more is known about the situation. This warning decision is based on the analyst’s observation and
evaluation of signals, independent or correlated, and on her updating of the prior probabilities of possible
scenarios and their outcomes. The warning decision also depends on the analyst’s assessment of the crisis’
dynamics and perception of the preferences of the principal decision maker, as well as the lead time
needed for an appropriate response. This article presents a model to support this analyst’s dynamic
warning decision. As with most problems involving warning, the key is to manage the tradeoffs between
false positives and false negatives given the probabilities and the consequences of intelligence failures of
both types. The model is illustrated by revisiting the case of the attack on Pearl Harbor in December 1941.
It shows that the radio silence of the Japanese fleet carried considerable information (Sir Arthur Conan
Doyle’s “dog in the night” problem), which was misinterpreted at the time. Even though the probabilities
of different attacks were relatively low, their consequences were such that the Bayesian dynamic
reasoning described here may have provided valuable information to key decision makers.
For more: http://dx.doi.org/10.1287/deca.2015.0321
Partition Dependence and Carryover Biases in Subjective Probability Assessment
Surveys for Continuous Variables: Model-Based Estimation and Correction
Venkata R. Prava, Robert T. Clemen, Benjamin F. Hobbs, and Melissa A. Kenney
Abstract: As probability elicitation becomes widely used, methods other than one-on-one interviews are
being used to elicit expert probabilities. This paper considers biases that may arise when probabilities are
elicited in an online or workbook setting. We develop a prescriptive model in which the elicited
probability is a convex combination of the expert’s underlying probability with elements of partition
dependence and two anchors arising from responses to previous questions (“carryover” bias). Our model,
applied to two data sets, allows us to estimate the amount of the various biases in a set of elicited
probabilities from experts. We find that both the format of the questions—whether they appear on the
same or separate pages/screens—and the ordering of the questions can affect the amount of bias. Our
research addresses biases in the presence of multiple anchors and provides guidance on manipulating the
availability of anchors. The results demonstrate the persistence of anchoring even with careful
questionnaire design; thus, the proposed model-based methods are useful to suggest corrections for the
resulting biases.
Page10
Identifying and Structuring the Objectives of the Islamic State of Iraq and the Levant
(ISIL) and Its Followers
Johannes Siebert, Detlof von Winterfeldt, and Richard S. John
Abstract: This study addresses three questions: 1. What are the objectives of the leaders of ISIL?,
2. What are the objectives of the followers of ISIL?, and 3. How are the two sets of objectives related?
To answer these questions, we analyzed the transcripts of interviews and presentations of 59 subject
matter experts (SMEs) and conducted a separate analysis of speeches of ISIL leaders and selected Internet
sources. In both efforts we identified and structured the strategic, fundamental, and means objectives of
ISIL and its followers. The results indicate that ISIL’s leaders pursue four strategic objectives: (1)
Establish a Caliphate in Iraq and the Levant, (2) Control and Govern the Caliphate, (3) Expand Islam and
Sharia Law Worldwide, and (4) Recreate the Power and Glory of (Sunni) Islam. The followers’ objectives
can be partitioned into three strategic objectives: Humanitarian Fulfillment, Religious Fulfillment, and
Personal Fulfillment. The objectives identified from the SME interviews were similar to those identified
from ISIL leaders’ statements and the Internet. However, the Internet search revealed many more personal
objectives of ISIL followers. The results further indicate that ISIL’s leadership objectives are closely
aligned with those of its followers. There also is a sharp contrast between the objectives of ISIL and those
of Al Qaeda, particularly ISIL’s emphasis on occupying and controlling territories in Iraq and Syria versus
Al Qaeda’s focus on worldwide jihad.
The authors give an overview of their research in the premiere appearance of the DECA blog, Decision
Analysis Review, found here: https://www.informs.org/IOL-Home/Blogs/DECA-Blogs/DECA-Review
Sequential Shortest Path Interdiction with Incomplete Information
Juan S. Borrero, Oleg A. Prokopyev, and Denis Sauré
Abstract: We study sequential interdiction when the interdictor has incomplete initial information about
the network and the evader has complete knowledge of the network, including its structure and arc costs.
In each time period, the interdictor blocks at most k arcs from the network observed up to that period, after
which the evader travels along a shortest path between two (fixed) nodes in the interdicted network. By
observing the evader’s actions, the interdictor learns about the network structure and arc costs and adjusts
its actions to maximize the cumulative cost incurred by the evader. A salient feature of our work is that the
feedback in each period is deterministic and adversarial. In addition to studying the regret minimization
problem, we also discuss time stability of a policy, which is the number of time periods until the
interdictor’s actions match those of an oracle interdictor with prior knowledge of the network. We propose
a class of simple interdiction policies that have a finite regret and detect when the instantaneous regret
reaches zero in real time. More importantly, we establish that this class of policies belongs to the set of
efficient policies.
Page11
The Decision Analysis December 2015 Issue
http://da.journal.informs.org/content/12/4.toc
“Gaming the System”: Decision Making by Interdependent Critical Infrastructure
Allison C. Reilly, Andrew Samuel, and Seth D. Guikema
Abstract: Supporting strong, resilient, and integrated critical infrastructure is vital to upholding the U.S.
economy and its national security. Arguably, no sector of the economy could exist without the reliable and
predictable networks on which it depends. To date, the research on interdependent infrastructure has
concentrated on describing the sources of the dependencies and developing models for predicting
performance and cascading disruptions after a hazardous event. However, these models fail to capture the
perspective of the operators of these networks and how competing, independent objectives lead to
suboptimal investment decisions and hence suboptimal network performance. Rather than focus on how
interdependent infrastructures operate and possibly fail, we take the perspective of their operators and ask
why they make the decisions that they do. The goal of this paper is to demonstrate how strategic
interdependencies may impact performance of coupled systems by shifting investments away from what is
collectively best toward decisions that are more myopic and optimal from the perspective of a single
infrastructure. Through our model, we make inferences on the level of investments networks make, relate
this to performance, and provide policy recommendations on how to promote reliable infrastructure.
Modeling Public-Private Partnerships in Disaster Management via Centralized and
Decentralized Models
Peiqiu Guan and Jun Zhuang
Abstract: The objective of this paper is to help both public and private sectors make better decisions in
defensive resource allocation through public and private partnerships (PPPs). In this paper, efficient PPPs
are studied with regard to disaster preparedness using a decentralized model (sequential game where the
public sector is the first mover) and a centralized model. This paper identifies the best public investment
policies by evaluating the effectiveness of incentive provisions based on the various private strategic
responses. This paper also provides insights into understanding (a) how to construct optimal public and
private partnerships and (b) whether, when, and to what extent public and private investments in disaster
preparedness could form better PPPs. We study the conditions of the private and public sectors’ allocation
strategies when they are strategic complements or substitutes. We find that the private sector that has a
higher target valuation or lives in more risky areas invests more and has higher potential to partner with
the public sector. We also compare the decentralized model results with the results of the centralized
model to study the efficiency of the PPPs and find that the results are similar when the target valuation or
the probability of disasters is small.
Equal Tails: A Simple Method to Elicit Utility Under Violations of Expected Utility
Page12
Manel Baucells and Antonio Villasís
Abstract: Classical methods to elicit utility are biased because most individuals do not treat probabilities
linearly. We propose a simple modification of the classical methods that equates, for all prospects being
compared, the range of outcomes. We argue that the modification should work in theory, and test the
modification experimentally. Our first experiment confirms that the modified certainty equivalent method
reduces the curvature of the S-shaped value function. The second experiment is a novel design that
compares the trade-off method with the three classical methods in their original and modified forms. Our
equal-tails modification of both the certainty equivalent and the lottery equivalent method produces results
consistent with the trade-off method. The lottery equivalent modification is particularly useful to elicit
utility points when outcomes are nonquantifiable.
Theory of Generalized Risk Attitudes
Janne Gustafsson
Abstract: This paper develops a theory of risk attitudes that can be applied in a broad array of settings,
including those in which the decision maker (DM) abides by a preference model other than the expected
utility model and in which decisions are being made over multiattribute alternatives. The theory is based
on (i) a set of plausible axioms in which the DM’s preferences over consequences and lotteries are defined
separately and (ii) the premise that a risk neutral DM is indifferent between a lottery and the average (in
terms of preference) of the outcomes obtained from infinite repetition of the lottery. We show that, under
these assumptions, a risk neutral DM seeks to maximize the expectation of classic cardinal utility (i.e.,
measurable value). This means, in particular, that the DM’s risk attitude in expected utility theory is
related to the transformation function between the classic cardinal utility function and the von NeumannMorgenstern utility function. The results also suggest that the applicability of the conventional definitions
of risk attitudes may be limited to settings in which the DM’s classic cardinal utility function is linear and
that a more generalized treatment of risk attitudes is required for settings in which this is not the case.
Attention INFORMS Decision Analysis Society Members!
By special arrangement with the Decision Analysis Society Council,
dues-paying regular members of the DAS receive a subscription to the journal as part
of their membership dues.
The DAS is a subdivision of INFORMS.
For information on DAS: https://www.informs.org/Community/DAS .
Decision Analysis is a quarterly journal dedicated to advancing the theory, application, and teaching of all aspects of decision
analysis. The primary focus of the journal is to develop and study operational decision-making methods, drawing on all aspects
of decision theory and decision analysis, with the ultimate objective of providing practical guidance for decision makers. As
such, the journal aims to bridge the theory and practice of decision analysis, facilitating communication and the exchange of
knowledge among decision analysts in academia, business, industry, and government. Decision Analysis is published in March,
June, September, and December by the Institute for Operations Research and the Management Sciences (INFORMS) at 5521
Research Park Drive, Suite 200, Catonsville, Maryland 21228. Please visit our website at
http://pubsonline.informs.org/journal/deca.
Page13
!"#$%&'()*'+$%,&-'(*'./0$/-1'2345'
'
!"#$%$&'()'*+,%$%(-&.*,(
DA Around the World
Column Editors: Matthias Seifert
In this column we introduce Decision Analysis (DA) communities around the world with the purpose of
promoting their visibility and strengthening the ties between DA researchers and practitioners across
borders. I would like to draw your attention to two exciting opportunities in Brazil and Spain. The first
one relates to an announcement of the 2016 summer school in multi-criteria decision aid / multiple criteria
decision making, which will take place in Recife, Brazil in July 2016 and is sponsored by our colleagues
at Universidade Federal de Pernambuco. For further information, please feel free to get in touch with the
organizing co-chairs Danielle Costa Morais ([email protected]) and Luciana Hazin Alencar
([email protected]). The second opportunity refers to a tenure-track faculty opening in Decision Sciences at
IE Business School in Madrid, Spain, where the school is currently in the process of growing a group of
DA researchers within their Operations & Technology Department. For further information please write to
[email protected].
ANNOUNCEMENT OF MCDA/M SUMMER SCHOOL 2016
The MCDA/M Summer School is a two-week event, taking place July 18th through 29th, 2016 in
Recife/Brazil. Details about the program, the application process and the venue are available at
http://cdsid.org.br/mc-summer-school2016/ The aim of this school is to give graduate students/young
researchers a state-of-the-art presentation of multiple criteria methods, applications and software and to
stimulate a networking of young researchers in MCDA/M.
The scientific program of the summer school consists of invited lectures and teams of participants working
on case studies.
Distinguished professors who have already confirmed their participation are:
•
•
•
•
•
•
•
Ehrgott, Matthias;
Figueira, José Rui;
Geiger, Martin J.;
Greco, Salvatore;
Keeney, Ralph L.;
Köksalan, Murat;
S!owi"ski, Roman.
The program will include the following topics:
6/7&'4)'
•
•
•
•
•
•
•
•
•
•
•
•
•
•
Introduction to MCDA/MCDM;
Applications in real world problems;
Problem Structuring; Preference Modelling;
Outranking Methods ;
MAVT/MAUT (Multi-Attribute Value/Utility Theory);
Robust Ordinal Regression;
Decision Rule Approach;
MCDM Group Decision ;
Interactive Methods of Multiobjective Optimization (IMMO);
Multiobjective Combinatorial Optimization (MOCO);
Evolutionary Multiobjective Optimization (EMO);
Decision Deck;
Scientific writing and publication strategy;
MCDM community and History.
We intend to limit the number of participants to roughly 50 graduate students. The application process is
already open and the deadline for applying is December 30th, 2015. In order to apply, please send by email
to [email protected]:
•
•
•
•
your detailed curriculum vitae
an abstract describing your graduate work
a letter from your supervisor specifying in clear terms your motivations to attend the summer school
the Application Form filled (download from http://cdsid.org.br/mc-summer-school2016/application/).
The result of selection process will be communicated before January 31st, 2016. If accepted, you will
receive an e-mail with instructions for registration. The early registration deadline is March 15th, 2016.
Registration with accommodation includes attendance to the summer school, educational material, lunches,
dinners, and social activity and is highly recommended to facilitate the highest level of interaction among
students. The venue will be a hotel in Boa Viagem Beach, Recife, capital of Pernambuco, that is
strategically located in northeastern Brazil. Pernambuco has a wide variety of natural, cultural and
historical attractions. It has as a tropical climate which means it is bathed on sunshine throughout the year.
Recife has a prime location, not only in regional terms within Brazil but also in relation to Europe (a 7hour flight from Lisbon) and to North America (a 7h 30min flight from Miami), and of course it is easily
accessible from other Brazilian cities and elsewhere in South America. The International airport of
Recife/Guararapes has direct flights to Lisbon (via TAP), Miami (via American Airlines), Panama (via
Copa Airlines) and Frankfurt (Via Condor Airlines). Other national hubs with daily connections to Recife
include airports such as Guarulhos in São Paulo and Tom Jobim in Rio de Janeiro.
We look forward to seeing you in MCDA/M Summer School in Recife/Brazil, July 2016.
Don’t miss this opportunity!!
--------------------------------------------------------------------
Page15
!"#$%&'()*'+$%,&-'(*'./0$/-1'2345'
'
!"#$%$&'()'*+,%$%(-&.*,(
TENURE TRACK FACULTY POSITION IN DECISION SCIENCES
Operations & Technology Area
The Operations & Technology Area at IE Business School (www.ie.edu) invites qualified applicants for a
tenure track position in Decision Sciences beginning in September 2016.
The successful candidate will be expected to publish in top-tier peer-reviewed international journals, to
teach high level MBA and executive courses in the fields of decision sciences, business analytics and
statistics, to interact and work closely with faculty from other disciplines, and to provide intellectual
leadership in his/her area. Applicants will have earned a Ph.D. in Decision Sciences, Management Science,
Operations Management or related areas from a recognized school by the date of the appointment (entry
level candidates can be close to completion). Preference will be given to candidates with a strong
analytical/econometric background and whose research can be easily applied to the domain of behavioral
operations.
Proficiency in Spanish is a plus, but not required. Salary will be commensurate with qualifications and
experience. IE Business School is an Equal Opportunity Employer and strongly encourages applications
from individuals who will expand the ethnic and gender diversity of our faculty. Please submit your
application in electronic form to:
Sara Flores
Recruitment Coordinator
Email: [email protected]
Tel.:+ 34 91 5689781
Applications should include a cover letter and vita, a one-page research statement, a statement of teaching
philosophy, copies of recent publications, contact information for three references, and evidence of
teaching excellence (e.g., teaching evaluations). Applications received by February 1, 2016 will be given
priority consideration. For informal enquiries about the position, please contact Professor Matthias Seifert
([email protected]).
About the school:
The level of quality of IE Business School’s masters’ and executive education programs has positioned the
school among the best in the world according to international rankings including the Financial Times,
BusinessWeek, Forbes, The Economist, Wall Street Journal and América Economía. IE faculty is
comprised of multi-cultural, exceptional, young and dynamic scholars from all areas of management, who
help our students to develop their full potential from different perspectives. IE Business School is located
in Madrid, Spain, in the heart of the financial district. Madrid is a vibrant, modern and cosmopolitan
capital city of 5+ million that offers an enormous range of professional growth and leisure opportunities.
To learn more about IE Business School and our Faculty, please visit http://www.ie.edu/business-school/.
6/7&'45'
DA Practice
Column Editor: Larry Neal
DA Practice Column
Measuring the Unmeasurable
Hello everyone and Happy New Year. I hope you and your families find yourselves healthy and happy
here in a very young 2016.
This time around, I want to expand on something that can really make a difference in tough analysis
situations when done well, dealing with hard to value metrics.
I am a huge fan of Douglas Hubbard, especially his work How to Measure Anything, Finding the Value of
“Intangibles” in Business, which I have read cover to cover twice. At times however, even Doug’s
pragmatic approaches failed me when things just didn’t want to be measured.
Rather than bore you as usual with my own thoughts, I’ve asked 4 people to offer their takes on this
subject. Each of these individuals has had a significant impact on my thinking, and I hope you will
experience the same.
So without further ado, our guest columnists to kick off 2016;
Dave Macway, a longtime friend and mentor who taught me the real meaning of clarity.
In many organizational decision contexts, there is a need to incorporate difficult to monetize values into
the decision making process. In my experience, it is important to be both careful and practical to help
decision makers reach clarity about the trade-offs they are making. I have found the following useful:
1) Clarify whether the issue is a direct value measure (one that a decision maker is willing to trade-off with
other value measures, say, NPV) or an indirect value measure (one that is a component of a direct value
measure – for example, cost or market share are components of NPV of cash flow). Indirect values should
be integrated into the direct value measures and not “double counted” as a separate decision criterion.
2) Just because you either can’t or are not willing to monetize a value measure (e.g., safety, environmental,
reputation), doesn’t mean that it can’t be quantified. If at all possible (and it is possible way more than
people initially think – be creative), quantify the value measure (e.g., tons of soil remediated, thousands of
cubic feet of gas not flared, etc.). Great advances have even been made in recent years at quantifying brand,
reputation, and other so-called “strategic influences.”
a. Barring a clear metric – even if is a proxy – use a scoring system of some sort, such as traffic lights,
barometers, “low-medium-high”, or some other such qualitative assessment.
3) Calculate, including uncertainty, the direct value measures for each alternative and provide simple, but
insightful, displays to show the trade-offs explicitly. Look for dominate alternatives that “win” on all value
measures and search for hybrid alternatives that can optimize across the decision space.
Page17
Remember, Decision Quality is obtained by stimulating the right conversations with the right people in the
right way. The technique and process described above has worked countless times to enable decision
makers to discuss and gain clarity and alignment around the best course of action when difficult to
monetize values must be considered in the decision making process.
Next up are two very good friends, Greg Parnell and Terry Bresnick. These men taught me how to use
my DA skills to lead. Whether it’s a medical matter or organizational change, we can make a positive
difference in others’ lives.
Working with multiple objectives and value measures that are hard to monetize
Most Government clients don’t think in terms of net present value (NPV) or return on investment (ROI),
so when it comes to their difficult decisions, we almost always are working with multiple objectives and
value measures that are hard to monetize. As private companies consider the Triple Bottom Line (profit,
people, and planet), they may also consider objectives that are hard to monetize. We have found five
practices to be most helpful:
1. Defining the decision problem – The initial problem is never the final problem. We have found
that it is important to spend time talking to decision makers and stakeholders to understand the real
problem and, perhaps, the opportunities. As a wise senior leader once told us, “this is too good of a
problem to waste!” Using his advice, we converted an embarrassing problem to a significant
opportunity to improve the organization’s mission assurance.
2. Assessing values – We show them a simple example of the use of both relative and absolute nonmonetary scales. We often use an example to which most folks can relate such as evaluating
upgrades to a basic computer system. We learned years ago that those who have never done this
are very skeptical about these “soft benefits” but feel very confident about their cost models. By
the time we take them through a sample problem, they believe the benefits and start to question
their cost models!
3. Assessing swing weights – Again here, taking them through a simple example can show them the
negative impacts on the decision that using the importance weights make. We often use a simple
job example with three criteria such as job satisfaction, location, and first year’s salary, and first
ask them to assess importance weights with no alternatives defined. Then we show them
alternatives for which the highest importance weight shows no difference among the alternatives.
We present the results in terms of a swing weight matrix to emphasize the two components of the
swing weight. It becomes obvious that the importance weights push them in the wrong direction.
In one study we worked on, the use of importance weights caused the study results to be rejected
as not credible
4. Visualization of the process – We find that many analysts use a computer to elicit value curves
and swing weights. They do this because it is easy for them. Instead, we need to make the
assessment process easy for the people doing the assessments. Computer screens limit you to one
sight picture at a time, and scrolling is highly disruptive to concentration. Our best practice is to fill
the walls with large wall charts with, for example, value curves where we use colored sticky dots
Page18
to score alternatives. We encourage participants to go to the charts, and physically move dots
around in front of the group to make the models theirs. By having the entire model displayed
around the room simultaneously, it makes it very easy to compare one value curve with another.
Often, we hear, “this value curve should look a lot like that one.” That can save time and effort.
Most importantly, we have full transparency of the model.
5. Communicating the results – We have learned that there is no one best way to communicate the
results of the deterministic and probabilistic analysis. Different decision makers have different
leaning styles and different knowledge of probability. Therefore, we try alternative presentations.
The most popular forms are the value component, cost vs. value, dominance charts, risk profiles,
and football charts for portfolio analyses.
Last but not least is a man that I met later in my career. Ralph Keeney, the man who defined “begin with
the end in mind”, long before Stephen Covey penned The 7 Habits of Highly Effective People. Ralph
rekindled that fire I felt 25 years ago when I first discovered DA and DQ and has helped me find many
new ways to contribute to others’ wellbeing.
Comments on Hard to Value Objectives
For any decision being faced, it is useful to understand the objectives. The objectives state why it is worth
the time and effort to think about a decision. If one is using the objectives to create or thoughtfully
appraise alternatives without analysis, no metric (also referred to as a measure or scale) is needed to
indicate levels of achievement regarding the objective.
If analysis is planned, it is necessary to have a way to measure each objective. For this purpose, metrics
can be categorized into three types: natural, constructed, or proxy. Natural scales, which are in general use
and have a common interpretation, are more useful if they exist. For an objective of “minimize cost”,
dollars is a natural metric. For the objective of “minimize disruption to the public” of a large project, the
number of people disrupted or the total person-weeks of disruption are natural metrics.
The more difficult situations occur when there are no natural metrics. Then the analyst faces the decision
of whether to use a proxy or constructed metric. A proxy is a natural scale, but it does not directly measure
the levels of achievement of the associated objective. If an objective is to “minimize heart attacks from
carbon monoxide”, a proxy is parts per million of carbon monoxide in the air. There is a probabilistic
relationship between carbon monoxide levels and the number of exposed people experiencing heart
attacks. This suggests another proxy, namely the number of people exposed to high carbon monoxide
levels. This requires specifying a high level of carbon monoxide to be used in the proxy. In any case,
eventually constructing utilities for metric levels presumes some understanding of the probability relating
exposure and heart attacks, which is not an easy task. If one can model this relationship, the natural metric
of the number of heart attacks caused can be used.
A constructed metric is a measurement scale that you constructed for the specific decision being faced.
For some objectives, the constructed metric may be best. Consider a possible objective of a major
company to “minimize negative press coverage” on an upcoming project. Such negative coverage may be
accurate or not. For any constructed metric, you want to specify a number of levels ranging from
Page19
something that is possibly worst to something that is possibly best. You also want to specify enough levels
in between to accurately categorize levels that are distinct from each other. Suppose you have a sevenlevel scale from A (the best) to G (the worst). The seven levels are usually defined by written descriptions,
but may be described by pictures for some objectives such as those pertaining to visual impacts. A written
description should describe the consequences of a given level in terms of the achievement with respect to
the objective. There may also be past examples of consequences relevant to the decision makers that can
be included as examples of specific levels of achievement to enhance their meaningfulness.
A common significant mistake in specifying metrics is to oversimplify. If an objective is important, a
metric that essentially categorizes possible consequences into descriptors such as low, medium, and high
with no careful definitions indicating meanings clarifies little to nothing and often distorts thinking. Even
with simple definitions of these meetings, the ranges are often inadequate to characterize important
differences. Both situations suggest that the objective is either unimportant and should be skipped in the
analysis or that inadequate thought and creativity was given to appropriately address it. A second common
mistake is simply describing consequences of alternatives on a scale from 1 to 10 with no levels specified.
With such an inadequate scale, the specific thought and logic given to describe consequences is also
frequently lacking. As a result, the value of any subsequent analysis is greatly limited.
So there you have it. A little longer than normal, but I hope well worth the read.
Next time we’ll tackle a concept that seems to have wings lately, the Triple Bottom Line.
Please feel free to drop me a note at [email protected] with comments or ideas for future columns.
Page20
!"#$%&'()*'+$%,&-'(*'./0$/-1'2345'
'
!"#$%$&'()'*+,%$%(-&.*,(
Society for Decision Professionals
Decision Analysis Affinity Group (DAAG)
2016, April 6-8
and
Joint DAS/SDP Webinar
DAAG Meeting
The Society of Decision Professionals has announced that the 22nd annual DAAG meeting will be held
April 6-8 in beautiful Banff, Canada. The venue is the Banff Park Lodge Resort Hotel & Conference
Centre on the beautiful Bow River. http://www.banffparklodge.com/
DAAG is open to all and is a great opportunity to network with leading decision analysis practitioners
from around the world. Attendees span the gamut of application from pharma, military, personal/family,
and agriculture to oil and gas. It’s a great way to see what works and what’s new in the hands-on
application of decision analysis.
Registration is now open. The early bird rate for the general sessions is $425 until February 15th. The
regular rate after February 15th is $500. Registration includes breakfast and lunch on April 7 & 8.
Here’s a peek at the program;
April 6 is a workshop day. Three workshops will be available;
•
Enriching Your Frame: Use of Scenario Thinking for Strategy Development; a half-day session
featuring Ellen Coopersmith, Kent Burkholder, and Jeremy Walker, $250*
•
Facilitation Training; a half-day session featuring Kathryn Rosback, $250*
(register for both half-day workshops for a discounted $450)
•
Biases in Decision Making; a full-day session featuring Jennifer Meyer and Carl Spetzler, $450*
*includes breakfast and lunch
April 7th and 8th are the general sessions;
The keynote speakers are:
•
Michael Runge, winner of the 2015 DAS Practice Award, will discuss his work on the Glen
Canyon Dam
6/7&'24'
•
Dave Snowden will discuss the Cynefin framework for decision making
The topical sessions include;
•
•
•
•
•
•
•
Corporate and Organizational Decision Making
Medical Decision Making
Decision Analysis for Everyone!
Capital Decision Making Under Resource Constrained Circumstances
Societal and Non-Profit Decision Making
Unsolved Problems and Developing Ideas
The Cutting Edge of Decision Analysis
For additional registration details, please visit
http://www.decisionprofessionals.com/event/daag2016/home.php
Hope to see your there!
Joint DAS/SDP Webinar
On December 16th, the DAS and SDP jointly sponsored a webinar featuring Dr. Robert Hammond of
Chevron with his presentation “Discrete Approximations to Continuous Distributions in Decision
Analysis”. The event was hosted by Dr. Eric Bickel, our DAS President, and was attended by more than
200.
If you missed the opportunity to hear Dr. Hammond, the event was recorded and can be found here.
http://www.decisionprofessionals.com/?scrollTo=http://decisionprofessionals.com/library#library
The DAS/SDP joint webinar effort will continue in 2016.
announcements of future events.
Please watch your email in box for
If you have topics that you would like to see presented at these joint webinars, please contact Hilda
Cherekdjian, SDP Executive Director, at [email protected] with your ideas.
Page22
Research
Column Editor: Debarun Bhattacharjya
Summary of Research Column Articles: 2012 – 2015
Dear Reader,
This will be my last issue as research column editor for DA Today. I wish to thank the co-editors Heather
Rosoff and Jun Zhuang as well as former editor Jeff Keisler for providing me the opportunity to serve the
newsletter and simultaneously learn about a breadth of research in the decision analysis community. Most
of all, I wish to thank the authors who have contributed excellent articles for the column. Below, I list and
briefly summarize the 11 research column articles from my tenure: 2012 – 2015. I hope you have found
the articles informative and inspiring. Best regards, ~Debarun
S. No.
Issue
Author
Title
Summary
1.
2012,
Vol 1
Ross Shachter
(Stanford
University)
Effective Decision
Analysis: A Cause to
Believe In
2.
2012,
Vol 2
Robert Bordley
(Booz Allen
Hamilton)
3.
2012,
Vol 3
4.
2013,
Vol 1
David Banks
(Duke University),
Jesus Rios (IBM
T. J. Watson
Research), David
Rios Insua (Royal
Academy of
Sciences, Spain)
Lav R. Varshney
(IBM T. J. Watson
Research)
Winning the Battle in
the Marketplace:
A Bayesian Alternative
to Chance-Constrained
Programming
Issues in Adversarial
Risk Analysis
Argues that causality is a
fundamental underlying
concept in the foundations of
decision analysis, illustrating
this using influence diagrams.
Discusses a Bayesian
alternative to chance
constrained programming,
making it consistent with
decision analysis principles.
Provides a summary of
literature on adversarial risk
analysis, focusing on security
applications, and highlights
some outstanding problems
for the emerging field.
Surprise in
Computational
Creativity and Machine
Science
Page23
On the use of Bayes’ rule as a
measure of surprise/ novelty
for computational creativity
and machine science.
5.
2013,
Vol 2
6.
2013,
Vol 3
7.
2014,
Vol 1
8.
2014,
Vol 2
9.
2014,
Vol 3
10.
2015,
Vol 1
11.
2015,
Vol 2
Gilberto
Montibeller
(London School of
Economics), L.
Alberto Franco
(Loughborough
University)
Jyrki Wallenius
(Aalto University)
Facilitated Modeling in
Decision Analysis
Presents the facilitated mode
of engagement in decision
analysis, suitable for
organizations that are more
networked and where power
and knowledge are
distributed.
Behavioral Decision
Welcomes the renewed
Research Revisited:
interest in behavioral decision
The Deterministic
making and provides a
MCDM Perspective
perspective on behavioral
decision research in a
deterministic MCDM context.
Gordon B. Hazen Sensitivity Analysis via Introduces information
(Northwestern
Information Density
density as an information
University)
value tool that provides
insights about direction of
concern for sensitivity.
Simon French
Bayesian Decision
Reviews scenario analysis and
(University of
Analysis in Parallel
discusses its potential as a
Warwick)
Small Worlds
powerful tool in a complex
and highly uncertain world.
Léa Deleris (IBM Building Bayesian
Highlights research on
Research –
Networks from a
building graphical models
Ireland)
Variety of Inputs
such as Bayesian networks
using a variety of inputs, such
as information in text form.
Eva Chen
The Contribution
Reviews the contributed
(University of
Weighted Model:
weight model – an approach
Pennsylvania),
Identification and
that identifies expertise and
David V. Budescu Combination of
uses it to combine forecasts
(Fordham
Expertise
effectively.
University)
Samuel E. Bodily Risk Preference
A conversation between
(University of
Lessons from “Deal or student and teacher that
Virginia)
No Deal”
expounds upon the benefits of
utility theory, demonstrated
using a dataset about the “deal
or no deal” game.
Page24
Ask DAS
Column Editor: Florian Federspiel
Developmental Policy at the World Bank Based on Decision Analytic Insights
Implementing decision analytic insights into practice comes with many of its own challenges and rewards.
The interface of policy makers and researchers poses a particularly interesting context of application, as
implementations may affect millions of people within and across different countries. To find out more I
talked to Andrea Vermehren, Lead Social Protection Specialist at the World Bank, about the bourgeoning
application of decision analytic insights to applied problems in developmental economics. One of the
World Bank’s proclaimed goals is to eradicate extreme poverty and a major way of doing so is through the
use of so called Social Safety Nets initiatives. These initiatives, targeted at vulnerable and poor
households to shelter them from economic shocks, consist of various programs from fee waivers for
housing and education to public work programs and conditional or unconditional cash transfers.
With its origins in Latin America in the early 2000s (e.g. in Mexico, Brazil, Jamaica, Colombia,
Nicaragua), many more countries from Pakistan, to China, Nigeria and Madagascar have implemented
cash transfer programs. While some cash transfer programs are bound to conditions linked to the
promotion of human development (such as children's regular school attendance, participation in
educational sessions related to nutrition, etc.), others are not as bounded. The cash transfer in all of these
programs is intended to change the recipients’ behaviors, related to both the human and productive
development of the household. Results, however, have been mixed. While most safety net programs have
shown to smooth consumption and provide some cushioning to incomes in crisis situations, their effects
on key human development indicators, such as nutrition and early childhood development, have been less
conclusive. The disparity in findings thus raises the question of whether there are better ways than
imposing conditions (e.g. school attendance, work requirements) to enhance the effectiveness of social
safety net programs employing cash transfers. That is because the more conditions imposed, the costlier
the implementation, as conditions usually imply the additional element of control.
Page25
!"#$%&'()*'+$%,&-'(*'./0$/-1'2345'
'
!"#$%$&'()'*+,%$%(-&.*,(
In that vein, and because cash transfer programs
are only one of many polices to benefit, on
October 22, the World Bank announced the
launch of the Global Insights Initiative (GINI).
GINI consists of a team of dedicated researchers
with the aim of integrating decision analytic and
related insights from psychology and economics
into the design and implementation of
developmental polices. This came shortly after the
release of the bank’s World Development Report
2015. The report focuses on the themes of Mind,
Society, and Behavior and argues for the need of a
significant re-haul of economic development and
associated policies based on behavioral insights.
The bank further launched a number of
collaborative efforts with outside researchers, -/"(0&1+.(2*'34%(/"*.56*17"1($'(0*%/$'87&'(!9(
think-tanks and practitioners such as UK’s
Behavioral Insights Team and ideas42, a small but influential behavioral consultancy (non-profit) with
academic roots.
One of the major challenges when
implementing
academic
insights
into
developmental policy is the need for simplicity,
efficacy and non-obtrusiveness. While for
instance new cash transfer programs under a
new set of conditions are later evaluated and
rigorously analyzed, programs are almost
always designed with well-established effects
in mind. The focus clearly lies on achieving
the greatest possible – and most certain –
impact, and the difficulty lies more so in
adapting or correctly applying an established
effect in the context of a particular policy,
society or geographic location. For example,
in the case of cash transfer programs, simple
interventions capitalizing on the effect of
mental accounting, such as assigning a label
(e.g. education support) to a cash transfer,
have proven very effective 1 , potentially
removing the costly need of enforcing
conditionality. Interestingly the balance
1
-/"(:1&;7/(&<(9*%/(-1*'%<"1(=1&81*>%'IC"$-E&K'S"-#H'T/0UM(
'''''''''''''''
http://web.stanford.edu/~pdupas/Morocco_Tayssir_LCT.pdf
6/7&'25'
between efficacy and the need for simplicity and non-obtrusiveness always remains to be struck. Notable
academic advisors recently argued for the use of multiple labeled envelopes in the hope of further
compartmentalizing the effective use of the cash, a proposal thus far rejected by the World Bank as it may
cause confusion and would likely come across as too contrived. Much work remains and is currently
underway in the area, as decision analytic insights may further benefit the effectiveness of recipients’
trainings, social service delivery, awareness campaigns, and further activities that accompany the cash
transfer.
While only having provided a snapshot of the recent developments at the World Bank, the message is
clear: The interface between decision analysis and academics from related disciplines with policy makers
(naturally not only relating to developmental policy) will become increasingly interesting and important
over the years to come. Both a thorough understanding of people’s behavior as well as effective,
prescriptive ways of reaching normatively (or societally) desired results are needed, and the field of
Decision Analysis has much to offer in that regard.
Further information about the World Bank’s efforts can be found in the bank’s recent World Development
Report 2015 (http://www.worldbank.org/en/publication/wdr2015), Andrea Vermehren’s blog
(http://blogs.worldbank.org/team/andrea-vermehren), as well as on GINI’s page
(http://www.worldbank.org/en/programs/gini#1).
I would like to thank Andrea Vermehren who offered generous help and insights for this piece. If you have
a response to this piece, or if you have further ideas or questions that you would like me to deal with in
future Ask DAS columns, please do not hesitate to email me ([email protected]).
Page27
Editorial Team
Research:
Dr. Debarun Bhattacharjya
IBM T. J. Watson Research
Center
[email protected]
Ask DAS:
Mr. Florian Federspiel
IE Business School
(Instituto de Empresa)
ffderspiel.phd2014@student.
Ie.edu
Dr. Heather Rosoff
University
of
Southern
California, Sol Price School
of Public Policy and
CREATE
[email protected]
Editor Assistant:
Ms. Jing Zhang
SUNY University at Buffalo
[email protected]
Co-Editor:
Dr. Jun Zhuang
SUNY University at Buffalo
[email protected]
DA Practice:
Larry Neal
Chevron Corporate
[email protected]
DA Around the World:
Dr. Matthias Seifert
IE Business School
[email protected]
Page28
!"#$%&'()*'+$%,&-'(*'./0$/-1'2345'
'
!"#$%$&'()'*+,%$%(-&.*,(
DAS Officers
DAS Council
President:
Eric Bickel
College of Engineering
University of Texas at Austin
[email protected]
Debarun Bhattacharjya
IBM T. J. Watson Research Center
[email protected]
Emanuele Borgonovo
Department of Decision Sciences
Bocconi University
[email protected]
VP/President-Elect:
Jason Merrick
Department of Statistical Sciences &
Operations Research
Virginia Commonwealth University
[email protected]
Joe Hahn
McCombs School of Business
University of Texas at Austin
[email protected]
Past President:
Jeffrey Keisler
College of Management
University of Massachusetts Boston
[email protected]
Melissa A. Kenney
Environmental Decision Analysis and
Indicators
University of Maryland
[email protected]
Secretary-Treasurer:
Yael Grushka-Cockayne
Darden School of Business
University of Virginia
[email protected]
Frank Koch
Brutally Frank Consulting
[email protected]
Jun Zhuang
College of Engineering
Department of Industrial and Systems Engineering
University at Buffalo, SUNY
[email protected]
Social Media Officer/Webmaster:
Jay Simon
Defense Resources Management Institute
Naval Postgraduate School
[email protected]
'
'
6/7&'2R'

Decision Analysis Today January 2016

Transcription

Similar documents

Annonce S1

waiver guidelines jakarta incident

OL396 Gunwharf Quay

Big data Web sites: 40 billion indexed web pages Youtube: 100 hrs

July 10, 2016 - Church of St. Joseph

July-August 2016 - Dallas Fort Worth Metro Golden Retriever Club

LOVE GOD, LOVE OTHERS, SERVE THE WORLD

Press Kit

Dinner Dance Dinner Dance - Franciscan Sisters of Chicago

Spring 2016 Newsletter - Borough of Manhattan Community College

Updates on friends and colleagues in the materials community

Das Haus - Interiors on Stage 2013

management science - Dipartimento di Economia