From the Executive Director
Although he is missed, I am very pleased to report that Vedat Diker, has taken a faculty position at the University of Maryland in the College of Information Studies. The Society hired Vedat in the summer of 2000 as a graduate assistant, but he has been a volunteer here since 1998. Vedat was instrumental creating the web-based conference paper review and submission systems. He also worked closely with Jack Pugh, Webmaster, assisting him to maintain the Society webpage. Vedat will continue to volunteer his efforts; he is currently working with Bob Eberlein to improve the conference paper submission process.
Among many other responsibilities, Jen Rowe, Assistant Director, continues to manage our membership, working hand-in-hand with John Wiley & Sons and MemberClicks, our online membership directory provider. Membership is growing. Last year in mid-October we had 860 members; this year at the same time we have 927. Our total membership for 2002 was 878. In the office we also have Joan Yanni and Robin Langer who continue to support office operations during peak periods. Jennifer Ferris, Bibliographer, continues to update our bibliography twice per year. There are currently over 7000 entries in the System Dynamics Society bibliography.
A financial summary of 2003 will be presented at the February Policy Council Meeting. Our finances are in better shape this year than last, though we continue to operate at a deficit. The New York City Conference made a profit and membership is up, but product sales are slightly down. For the first time in quite a few years, publications sales (Conference Proceedings and past issues of the System Dynamics Review) are up. As a result of the hard work and tremendous commitment of Jay Forrester and Nan Lux, the Society will soon offer a new product. During the last two years, some 4000 D-memos from the files of the MIT System Dynamics Group have been scanned and compiled onto one DVD disk. A notice will be sent out when the DVD is available for sale.
Each year starting in September we run a Society Sponsorship Campaign. Sponsorship funds help provide membership services such as the on-line directory, bi-annual newsletters, updates to the bibliography, and website maintenance. Sponsorship support truly makes a difference and we are endeavoring to expand our sponsorship base of businesses, universities and individuals. If you know of a potential sponsor or would like to learn more, please feel free to contact me.
Very soon you will be receiving the 2004 Oxford Call for Papers by mail. The Conference Chair, Jonathan Coyle, and Program Co-Chairs Mike Kennedy and Graham Winch are already hard at work! In early January we will be mailing the conference registration brochure. Because of limited space at the conference venue, Keble College, we are strongly suggesting to register early. Refer to the Society website for conference updates.
Please contact the Society office if there is anything this office can do for you.
Best regards, Roberta
Roberta L. Spencer
Nelson Repenning Wins Forrester Award
The Jay Wright Forrester Award recognizes the author of the best contribution to the field of system dynamics in the preceding five years. The winner in 2003 was Nelson P. Repenning for his paper “Understanding Fire Fighting in New Product Development” published in 2001 in the Journal of Product Innovation Management. The citation and the winner’s speech (delivered at the award ceremony in New York City) will be published in full in the System Dynamics Review. Below is a short excerpt from the citation.
Nelson Repenning is Associate Professor of Management at MIT Sloan where he recently received tenure. He obtained his PhD in Management from Sloan in 1996 and subsequently joined the faculty in Operations Management and System Dynamics. In the period 1995-1998 he was co-principal investigator on “Designing Sustainable Improvement Programs,” involving joint applied research with MIT, Ford Motor Company, Harley-Davidson, AT&T, and National Semiconductor. In the period 2000-2002 he was leader of the Implementation Dynamics Initiative in the Center for Innovation in Product Development at MIT.
It is out of such in-depth applied research that Nelson’s award-winning work arises. His overall research centers on understanding why many attractive and beneficial innovations and change programs designed to improve organizational performance often fail to take hold. He is especially interested in the implementation process, for example, the dynamics and side effects of a firm’s attempt to implement quality improvement tools or new product development methodologies.
John Morecroft, Chair of the Awards Committee, and Nelson Repenning
Dana Meadows Student Prize Awarded in New York
This year’s winner of the Dana Meadows Prize for the best student paper presented at the annual conference was David L. Cooke, Haskayne School of Business, University of Calgary, Canada. David’s paper was entitled “Learning from Incidents.”
Honorable mentions went to (in alphabetical order):
Jan Jaap Bezemer, Eindhoven University of Technology, The Netherlands, and co-author Henk Akkermans, for “Understanding Delays in Semiconductor Supply Chains”
Özge Pala, Nijmegen Business School, The Netherlands, and co-author Jac A.M. Vennix, for “A Causal Look at the Occurrence of Biases in Strategic Change”
Klaus Vogstad, Norwegian University of Science and Technology, Norway, and co-authors Ingrid Kristensen and Ove Wolfgang, for “Tradable Green Certificates: The Dynamics of Coupled Electricity Markets”
David Cooke, Klaus Vogstad, Jan Jaap Bezemer, and Özge Pala
In Memoriam – Lee Frost-Kumpf
Lee Frost-Kumpf passed away on August 15, 2003. In recent years, beginning at the conference in Atlanta, Lee was enthusiastically and actively involved in the efforts to establish a “policy dynamics” strand of activity in the Society. Indeed, Lee chaired a session at the New York conference in one of his last public, professional activities. The Society has lost a member whose potential was only just beginning to be appreciated.
Nan Lux Retires from MIT
After some 16 years working with Jay Forrester at the System Dynamics Group at MIT, Nan Lux retired at the end of August, 2003. During that time she made many and continuous contributions to system dynamics. For several years, Nan was Vice President for Chapters of the System Dynamics Society. She played a major role in guiding a rotating group of MIT undergraduates who created the Road Maps series of exercises and the Guided Study in System Dynamics Program to create self-study material for learning the basic principles of the field. During the last two years, Nan Lux scanned some 4000 D-memos from the files of the System Dynamics Group. This archive, along with many theses and other papers, will become available from the Society on a DVD disk.
Jay Forrester, Nan Lux, and John Sterman
Alexander Ryzhenkov, Math. and Instrumental Methods of Economic Theory,. a second dissertation approved by Russia’s Education Ministry, July 2003
Hakan Yasarcan, Feedback, Delays and Non-Linearities in Decision Structures, Bogaziçi University, July 2003
New York City Conference Report
Photos from the New York City Conference are posted on the Society website at http://www.systemdynamics.org/conf2003/pictures/ –please take a look if you have not already done so. One of the best is that of Nan Lux, at the surprise tribute for her retirement. We wish her much success and happiness as she moves on to new adventures! Although there are no photos, an unforgettable session was “Celebrating the Diversity of System Dynamics” honoring Barry Richmond. We thank the many members of Barry’s family for attending.
You will find below articles written by volunteer conference session reporters. While not all sessions are covered, we hope these give a feel for the conference. There was an interesting mixture of work presented at the conference. The Public Policy Thread included issues such as health care, security, urban dynamics and military applications. It accounted for almost one quarter of all the conference presentations. Another sizeable thread was Methodology, again including a range of issues and topics. The third largest thread related to the conference theme, Economic Dynamics. The New York City conference had the highest registrations to date, totaling 460 from 45 countries. Extra copies of the proceedings are available through the Society office.
From the 2003 Conference Organizing Team, thanks!
Some members of the conference team: Pål Davidsen, Bob Eberlein, Mike Radzicki, Roberta Spencer, and Allen Boorstein
Conference volunteers Ignacio Martinez, Silvia Ulli-Beer, and Vedat Diker
Located in the heart of all Conference activities–on the Mezzanine Level–the participants encountered twice a day–in the morning and in the afternoon–a refreshment break for both flesh and soul. While coffee, tea, and some pastries helped to keep the energy level high, the poster market generated stimulating discussions in the process of exchanging ideas. This forum broke the mainly one-way communication of the plenary and parallel sessions–a welcomed change to keep the mind awake.
Nearly a hundred projects were presented in numerous theme clusters. An overall impression of these sessions was stamped by diversity: in the design of the posters, in the topics, and in the state of the art of the work.
Nearly all posters attracted interest. The authors enjoyed direct contact with their audience and they were quite busy explaining their work. Mostly, they found themselves involved in deepening discussions and exchange of experiences. Nearly always they received constructive feedback. Authors and participants made important contacts and, in the process, discovered peers that are working (or have interests) in related topics.
A poster session is an attractive alternative to an oral presentation. Many claimed that poster presentations had a higher return on investment than oral presentations. This indicates that presenting work in a poster session is also a great opportunity for networking, exchange and inspiration.
In retrospect, it is fair to say that it was not only the quality of the posters but also the overall setting of these sessions that was critical for their success. The informal atmosphere was refreshing and facilitated the exchange of ideas. The central location brought the audience in contact with the presenter. Given these highlights the crowding in the hallway could not cast any deep shadows on the poster sessions.
Defining and Refining Methodology
If system’s redesign is the raison d’être of system dynamics, what is hampering its successful application as a policy design tool? What challenges are presented to the field as it struggles to improve the world? The two views presented in this plenary session provided different answers to this appraisal. One stated we need a good and reliable compass; the other argued for more velocity.
Henk Akkermans and Georges Romme presented the paper “System Dynamics at the Design-Science Interface.” The authors argued that the academic success of system dynamics has largely been accomplished by positioning it as part of mainstream science, but this may now split the field into two disconnected segments, one for practitioners and the other for academics. According to the authors, these two forces are a product of the tension between the competing goals of design and science: relevance of applications and rigor of analysis. While practitioners are hard pressed to apply the method to enhance organizational effectiveness, academics struggle in a “publish or perish” environment, and must seek out and conform to specific scholarly interests and norms.
How can a healthy coexistence of practice and academia be maintained? The answer: by being determined to stay at the design-science interface! The paper discusses how this can and has been accomplished. It traces the success of system dynamics in its struggle for both relevance and rigor, highlighting each step along the way. This is an interesting argument that underscores the major phases and the most relevant work in both domains of our field. It is an invitation to continue this discussion through modeling. Henk and Georges crafted the initial loops of this dilemma and invited everyone’s contribution.
Jim Hines presented an alternative viewpoint, entitled Modeling at Conversation Speed, asserting that the current slow speed of modeling is what consigns system dynamics to science and not design. This limitation undermines practice even more than academia. Excellent practice in the field (e.g., Pugh Roberts) has been pushed in the direction of large-scale models, carefully calibrated and validated, sacrificing design for prediction and retrodiction. In fact, according to the presenter, it is academia that has been able to focus on design. This is where parsimonious models aimed at policy insight found their place.
How can system dynamics be used for design in the world of practice? The answer: by increasing the speed of modeling by ten to 100 times! Since it is not reasonable to slow down the managerial process, then the alternative is to speed up modeling to keep up with conversation. Jim described how modeling technology has significantly changed since the 1960’s, becoming more efficient and interactive, and how with a few more advances it will be possible to speed up the whole process just enough for modeling to keep pace.
The two arguments presented the following puzzle: Is the pursuit of relevance really at odds with rigor? Where’s the leverage, in the philosophy embraced by the field or in the methods and technology applied? While these two views addressed these questions differently, both shared in the concern that system dynamics should be focused on the design of improved organizational form and policies, just as originally conceived by Jay Forrester. If one has to do with compass, and the other with velocity, then both are useful to go from point A to point B.
This plenary session showcased the power of system dynamics as a tool for tackling such diverse challenges as facilitating chronic health care planning in Washington state, automotive product development strategies at Toyota, and theory-building in the area of risk misperception and erosion of security compliance.
In semi-rural Whatcom County, Washington, Jack Homer and Gary Hirsch have worked with community leaders to improve care for diabetes and heart failure patients. Their model projects costs and benefits and identifies resource planning strategies over a twenty-year period. The modeling process has also enabled community leaders to forge a consensus regarding goals and evaluation for this program, as well as lay the groundwork for a broader chronic care program.
Building on their extensive experience in analyzing Toyota’s automotive product development process, David Ford and Durward Sobek have developed a system dynamics model that reveals how the Toyota strategy works and why it succeeds. Toyota outperforms competitors with an unconventional product development strategy that keeps alternative design options alive much longer than industry standard. The model combines system dynamics and real options theory to demonstrate how a competitive advantage emerges from the Toyota strategy.
Security systems are high priority in many organizations these days, yet cost-effective security is difficult to sustain. Breakdowns occur. Human compliance erodes. At Agder University College in Norway, Jose J. Gonzalez and Agata Sawicka are developing system dynamics models in which the driving mechanism for erosion of compliance is misguided learning due to misperception of risk. The model replicates the oscillating reference pattern that characterizes organizations that drift into complacency, overreact to danger signals, and then resume a drift that reflects unrealistic risk assessments.
Jonathan Coyle chaired this interesting session, which vividly demonstrated the range of issues within the domain of system dynamics and the variety of purposes that system dynamics modeling can serve.
Economics and System Dynamics
This plenary session was well attended, even coming as it did on the second day of the conference when many people started thinking about taking a little time to explore New York City. Khalid Saeed moderated the session and the panel consisted of Mike Radzicki, Glen Atkinson and John Shilling.
The session was aimed at examining the divergence and overlap between system dynamics and economics. This was one of a number of sessions that was devoted to bringing in folks familiar with system dynamics, but not necessarily practitioners, to discuss the relationships between their fields and system dynamics. John Shilling has spent most of his career working as an economist in academia and as a research economist with the World Bank. Glen Atkinson is a Professor of Economics at the University of Nevada, Reno and Mike Radzicki is a system dynamicist with formal training in economics.
The authors all took the long view and examined the development of economic thought from its beginnings to current philosophies. All the presenters emphasized that economics has shifted from descriptive techniques for explaining economic activity to mathematical techniques. These mathematical techniques now dominate the field of economics and are powerful tools for examining different economic theories. However, the authors pointed out that the price of mathematical rigor is often an oversimplification of the complex problems examined. The authors in this session argued that common ground exists between economics and system dynamics that could lead to fruitful results for both fields.
Teaching System Dynamics
The Teaching System Dynamics parallel session was a combination of papers that encompassed teaching system dynamics at a university, the application of system dynamics to examine problems in an urban school district, and a set of small models used to show policymakers the types of lessons that could be learned from a system dynamics intervention.
The presentation by Hakim Remita and Michel Karsky, “Teaching System Dynamics in a French Engineering School,” provided background on the school and students to whom they taught system dynamics. Remita and Karsky were able to effectively show how their students performed by showing examples of their students’ work. The student projects presented examined topics as diverse as competition between ant colonies and the dynamic issues facing a start-up software company. These examples captured the high quality of student work as well as the diverse interests of the students. Remita and Karsky’s success in teaching system dynamics is achieved by engaging students in projects of their own choosing.
Don Morris (his paper is titled “Peer Influence in Educational Reform: A System Dynamics Approach”) works in an urban school district in Florida and presented a system dynamics model that examined school performance, peer influence and the economic integration of schools. Dr. Morris’ presentation showed how system dynamics could be used to explicitly state a pre-existing theory examined in the education literature.
The presentation by Rod MacDonald, “Lessons from Simple Stock and Flow Models,” highlighted three models the author uses to convince clients in government agencies of the utility of using system dynamics models to analyze public policies. His examples were tied to very small models developed to address specific issues faced by different agencies.
More than 30 system dynamicists–amongst them Jay W. Forrester, the founder of the field–followed WPI’s Mike Radzicki’s call for an Economics Roundtable at this year’s conference. Initially, the discussion concentrated on the question of whether economics activities within system dynamics should be bundled into a formal chapter or a special interest group (SIG) of the Society. It was agreed that such a formalization would greatly enhance the visibility of economics in the field of system dynamics, and also would allow for ease of approaching economists outside system dynamics. Working with economists ignorant of system dynamics and demonstrating what system dynamics can offer regarding formulating and understanding economic issues was identified as a major aim. One advantage of system dynamics named is that system dynamics secures analytical rigor despite not following a neo-classical approach (i.e. focusing on real-world economic problems).
Another idea discussed at the Economics Roundtable was to publish either a book or a special issue of a journal dedicated to economic applications of system dynamics. Also, many attendants spoke in favour of a focused website about economics and system dynamics. With the help of this website, the quality of economic system dynamics models could be raised due to interaction with other members interested in the issue. It could also be used to prepare public relations activities, for instance media reports, teaching, and consulting in the political area. With measures like these, the dominance of the neo-classical school in economics could be weakened in order to allow for more pluralistic and heterogeneous approaches to economic issues.
Around 20 participants attended the second Military Roundtable discussion following from the first such session held at the conference in Palermo last year. Alan McLucas led a discussion covering two main themes: opening a debate on the sharing of knowledge between modelers and improving model ownership by military decision makers.
Both topics were covered with some lively debate. It was suggested that sharing of knowledge could include the development of an annotated bibliography and a model library. Participants agreed that commercial and security issues could limit this although there are many examples of military-related system dynamics work in the open literature. The establishment of model libraries is likely to include these issues and there was a debate on the reusability of such models.
A military model lifecycle can vary from the “dispose after use” model for immediate issues to planning tools developed and used on an ongoing basis. Alan stressed that these latter models are difficult to maintain since the military decision makers have high turnover rates (postings being part of their career pathway). Familiarity with the modeling approach appears to be a factor and Alan outlined the depth of system dynamics awareness that military officers receive through the military training establishments in Australia as a way of addressing this.
The meeting proved a useful opportunity to share experiences and there will be a third session in Oxford next year.
A trio of papers about the application of system dynamics to fisheries was presented in a parallel session. Topics discussed were a model of the Thai shrimp aquaculture fishery including politics and pollution (“Boom and Bust Shrimp Aquaculture: A Feebate Policy for Sustainability,” by Steven Arquitt, Honggang Xu and Ron Johnston); a biological model of the Northwest Pacific rockfish fishery (“A System Dynamics Model of the Pacific Coast Rockfish Fishery,” by Wayne Wakeland, Olgay Cangur, Guillermo Rueda and Astrid Scholtz); and a general discussion of system dynamics applied to complexity in fishery management (“A Basis for Understanding Fishery Management Complexities,” by Richard Dudley).
Arquitt used system dynamics to test a taxation policy to improve the Thai shrimp aquaculture industry: Namely, could the shrimp fishery adopt an economic incentive, a “feebate” or tax, to induce Thai shrimp farmers to locate their aquaculture facilities away from coastal regions instead of along the coast? The Thai shrimp fishery had been established along coastal regions near mangrove forests that provided nutrition and ecosystem cleansing benefits. The farms expanded along the coastline when the earlier coastal locations became unproductive, when the surrounding ecosystems became polluted, and world shrimp prices rose dramatically. Since these coastal locations were “commons” property there was no responsible authority to manage them.
In an effort to encourage farmers to move the shrimp farms further inland onto land actually owned by the farmers (thus avoiding the “commons” problem) a market incentive in the form of a tax on shrimp products was imposed on all growers in coastal regions and in inland regions: The plan was that later this tax would be refunded only to the inland farmer to serve as a financial incentive to encourage the inland location. System dynamics simulations indicated that this tax could encourage the use of inland locations and smooth out the severe economic oscillations in the Thai shrimp market price.
Wakeland presented an analysis of the localized yellow rockfish fishery from the Northwest US, concentrating particularly upon the balance of forces between the carrying capacity of the biological system and the fishing effort and technology. The analysis was prompted by a decrease in groundfish revenues of 70-80% in the Pacific Northwest.
A variety of factors were tested for sensitivity analysis as to their effect upon the total gross rockfish revenue: the natural mortality rate, bycatch rate, average vessel capacity, spawner rate, normal fishing rate, maturation time constant, and others. Of these, the critical controlling variables were the spawner rate and the maturation time constant (from 2 to 6 years) for the fish stocks. A maximum harvest of 40% of the stock per year was found to be the required level for long term stock sustainability. Simulations were tested successfully against data using biological and fishing factors. The next step is to include feedback from the economic and revenue sectors.
Richard Dudley used system dynamics to sort through the broad questions and complexities of fisheries management. The record of world fisheries shows rampant over-fishing: 33% of commercial stocks and 80 species threatened with extinction from North American waters–even with a regulatory system based upon considerable scientific expertise. Nevertheless, over-fishing is linked to the complexity of the process: feedback loops that are not taken into consideration in the decision making process, violation of regulations, time delays for stocks to come to equilibrium after a management action, the appearance of excessive over-fishing before over-fishing capacity becomes apparent. Management perceptions of stock may be too optimistic and failure can follow good intentions. A holistic approach encompassing biological, political, social and economic conditions is required, including delays in recruitment, effects on stock size, addition of vessels and new gear and efficiencies.
Success is viewed by management as maintenance of high stock levels; by the fishermen as whether they can get an adequate catch per unit of gear and effort; and by the politicians as whether this year’s catch is as good as last year’s. To understand the behavior of the overall biological fishery system and the impacts of management policy, simulations were made with variations in fishing vessel numbers, gear modification, monitoring the stock, changes in fish growth rates, and recruitment rates. Simulations derived from these inputs to the model are made to demonstrate the relative importance of these factors.
The opportunity to hear these three papers in a single session was most instructive since they each approached the fisheries issue using different stocks, oceans and approaches. Similar to the story of the blind men and the elephant, a deeper understanding of complexity is achieved through the application of several points of view.
The telecommunications session I attended consisted of three good papers, characterized by a business/industry-oriented approach.
The first one, presented by Rutger Mooy and co-authored by G. Valk (TNO Telecom, NL) was titled “Why Customers Choose Your Product: A System Dynamics Approach to Customer Choice Modeling.” The paper focused on modeling customer behavior in telecom markets and was based on a real case-study from the Dutch telecommunication industry.
In order to explain the churn phenomenon, a model was built by the authors to focus on what they call the “price cascade.” This is an interesting concept to start a dialogue with managers who are not accustomed to the system dynamics method. The “price cascade” describes how customers deal with prices and related factors and does not contain any loops. The cascade is made up of different steps, which are set up in such a way that they can be examined qualitatively and estimated quantitatively and individually. Such a concept is particularly useful to explore customers’ perceptions of different product/service factors, so it is helpful to give a weight to the different factors in the model. Examples are provided and discussed on how to implement such a concept in system dynamics modeling.
The second paper, presented by Douglas Franco (CANTV, Interconexion, Venezuela), was titled “Modeling the Telecommunication Market.”
The author was very clear in his explanation and some relationships (e.g. churn modeling) with a number of issues modeled by the first presenter were found and discussed during the session. In particular, the author explained how, in order to understand the telecom industry’s complexity, a system dynamics model of the telecommunication market was constructed. In the model it is shown how customers choose among alternatives, and mobile and fixed operators grow, merge, offer services and set prices. Regulators control prices for call termination and fixed telephony. The model generates profits, levels of consumption and surprises. A Calling Party Pays billing practice is assumed, but it is easy to adapt the model to RPP (Receiving Party Pays) billing. The model has been implemented in Vensim with data gathering in Microsoft Excel linked to corporate databases.
The third paper, presented by Oleg Pavlov (Worcester Polytechnic Institute), was titled “Using System Dynamics to Assess Economic Feasibility of Satellite-Augmented Cellular Networks”. Also this paper raised the interest of the audience. It was well presented by the author, who focused his analysis on a counterintuitive behavior that was experienced in real company contexts.
Computer experiments reveal that in the early stages of market development by augmenting its network with satellite capacity a cellular operator may improve its performance in terms of revenue, subscriber growth rates, profit and other business parameters. As more cellular capacity is deployed, the advantages of integrated systems disappear. The model was analyzed in its different sectors, including cellular infrastructure, satellite, financials, subscriptions, competition, infractructure cost pricing, and mobile service pricing.
Three speakers spoke in the Psychology Dynamics session. In Harold Kurstedt’s talk, he spoke about a free form model in which two individuals have a negative professional interaction due to a reinforcing loop in which misinterpretations lead to negative behaviors that encourage negative reactions. In this case, one individual must consider alternative interpretations and change behaviors in order to turn a vicious cycle into a virtuous one.
The second talk, by Rohita Singh, was an entertaining exploration of one’s own genius. In this talk, that foreshadowed a later workshop, we were asked to consider our pre-conceived notions of genius and investigate our mental filters.
The third talk, by Ralph Levine, involved a system dynamics model of a psychological theory of attitude change presented in the early years of psychology. This theory predicted a particular phenomenon in which the relative location of one’s prior attitude and the message presented about an object dictated whether change would be toward the message or away. Ralph was able to replicate findings and discover intriguing new ones in the application of his model.
Three speakers and one discussant spoke in the Psychology Symposium. John Flach focused on the cues used to compare current state and desired state, in order to trigger action. His example was the collision of a baseball and a bat. The result of his extensive empirical work was that a batter pays attention to the change in visual angle to the ball, rather than to its absolute distance from home plate or its speed coming toward it. From plotting these data, he was able to show that it is more difficult to hit a slow ball after being accustomed to fast ones, than to hit a fast ball after being accustomed to slow ones.
Alex Kirlik critiqued the exclusive focus by psychology on cognition without action, and revived an early model of Tolman and Brunswik in which cognitions and actions are integrated. In the course of describing ecological psychological approaches from Brunswik and Gibson, he showed his own model, and described empirical research on how airplane pilots can use simple heuristics to taxi around O’Hare to the appropriate gate, even in fog.
Jim Townsend spoke about a matrix model with feedback, the Field Theory of Dynamic Decision Making, for predicting decision processes over time. He showed how odd empirical findings from decision-making could be accounted for by a single model with a kind of co-flow structure in which alternative decision preferences compete over time.
Finally, John Sterman spoke about how psychology works in smaller time spans than typical system dynamics models, but system dynamics practitioners should put more attention to the functioning of the comparator process (current vs. desired state) and to the link between cognition and action.
Group Model Building
The three papers in this parallel session explored the ability of group model building to change mental models and behavior of participants. Approximately 50 people attended the session. Two papers discussed how best to elicit mental models from group model building participants and one paper proposed a framework to evaluate its effectiveness for changing mental models and behavior.
Jeroen Struben presented the paper entitled “The ‘Standard Method’: Scripts for a Group Model Building Intervention,” co-authored with Peter Otto. Struben and Otto described an application of Jim Hines’ “standard method” for examining possible consequences of establishing a fish processing facility for the Gloucester fishery. They argued that the standard method provides a guided but flexible framework that allows insights to emerge in the process of working with clients. In particular, they found the scripts used in the initial stage of the process, combining “hope and fear” scenarios with dynamic hypotheses, made the system structure and behavior clearer than a reference mode. Aldo Zagonel, in “Using Group Model Building to Inform Welfare Reform Policy-making in New York State: A Critical Look,” described a very large, multi-group, multi-level GMB application to understanding the consequences of welfare reform in New York State. He also stressed that the process was an emergent one and raised the question of how to best structure the process to help insights emerge. Zagonel made several observations from the study: the structure of the model emerges from alignment of group mental models; it is important that parameterization of the model draw on expert judgement; participant learning comes from live policy experimentation; checking model behavior against actual behavior helped build participant confidence; and it was important that discussions emphasize implementation.
Finally, Etiënne Rouwette addressed the lack of evidence for the effectiveness of group model building in a paper co-authored with Jac Vennix (“Process and Outcomes of Modeling: An Attempt at Formulating a Conceptual Framework”). Rouwette and Vennix proposed a conceptual model for evaluating effectiveness of group model building interventions and applied it to five case studies. As Rouwette noted, Jay Forrester’s goal for system dynamics is to change system functioning by changing mental models and ultimately behavior of the people exposed to system dynamics. Rouwette and Vennix, along with the other authors in this session, believe that group model building has the potential to change mental models of participants through the process of model building. Based on their evaluation framework, they concluded that such interventions do support the mental model and behavioral change that Forrester and other system dynamics practitioners are hoping modeling will achieve.
During the question period, one audience member asked whether it would not be better to start by having individuals in the group build their own models, because the literature on problem-solving groups says the outcome of a group decision will not be as good as a given individual view. The response, that individual models would get people embedded in their own views, exemplifies the theme of the session: a key value of group model building is its ability to bring participants to a shared view through the process and the model generated is often less important. Developing consensus around a group mental model is an emergent process, building awareness and confidence in participants.
Letters from Conference Participants
As a first-time attendee I learned a lot at the conference. Hopefully in the near future we can begin an occupational safety group in the Society.
I've enjoyed it so much, and the city itself was simply exciting!
The NYC conference was spectacular. The quality of the program speakers, papers and posters was universally higher than I ever remember. I also am very excited that the Society is reaching out to other disciplines, like the psychology track, that truly broaden and enrich our field. I met more new people doing more new applications than I had before. I believe this is very good for the Society. Moreever, I learned new ideas from new people and deepened my understanding of the areas I am familiar with by listening to the senior statesmen and the next generations. I look forward to continuing my work to contribute and expand the field with my efforts. The bar has been raised yet again for next year!!
All of us learned a great deal from this conference, and had the opportunity to meet many good people.
Several people came up to me after the Business Roundtable and said how much they got out of the event. Each contribution was valuable and the audience undoubtedly benefited. We heard from several quarters how difficult it is for untrained people to grasp stock-and-flow thinking and how strenuously people can resist insight from system dynamics work. The keys to making system dynamics more accessible are embedded in the experience of successful practitioners such as the panel members. While it’s important to hear advances in research, it’s perhaps equally or even more important that system dynamicists learn how such researches are put to use. So I think the Business Roundtable should get a higher billing in future conferences.
I have really enjoyed my first experience in participating in this conference and meeting nice and caring people. I have benefited from attending the different sessions and workshops, specially the modeling assistance workshop. I am sure I will try my best to attend and participate in next year’s conference in Oxford.
We all enjoyed ourselves, whether it was academic discourse or socializing. Great conference in a great city.
[ Return to contents ] [ Previous ] [ Next ]