System Dynamics Society Logo

System Dynamics

Volume 22 – Number 4        October 2009

Page 1, 2, 3, 4, 5, 6, 7

Session Reports - Monday

MON 9:00 AM Plenary - Opening Challenges

The Monday morning plenary session entitled “Opening Challenges” was chaired by Andrew Ford. “Opening Challenges” consisted of two presentations: 1) Four Grand Challenges for System Dynamics by Alan Graham, and 2) What Models Are Required To Understand Global Limits to Growth In The Next Two Decades? by Dennis Meadows.

Alan Graham’s talk entitled Four Grand Challenges for System Dynamics, presented four challenges for the future of system dynamics. Before discussing these challenges, he presented three criteria that were used to evaluate potential problems in the future. These criteria were patterned after the Defense Advanced Research Projects Agency’s (DARPA) criteria for future research. These criteria were; a) impact on lives or livelihood of millions or billions of people, b) currently being governed by mental models seriously in error, and c) being addressable by system dynamics and seemingly by few other disciplines. This approach resulted in problems that seemed to cluster into four areas.

These four areas represent the Four Grand Challenges for system dynamics are: 1) Implementing controls for global warming, 2) Global financial stability, 3) Insurgency, governance and political stability, and 4) Harmonious Chinese growth. Rather than go deeper into these four challenges with the time available, Alan focused on just two of these by presenting the apparent mental models governing action and then provided a closer to the truth mental model look. For example, for the Insurgency and Political Stability Challenge, he presented the following mental models held by many that are seriously flawed a) corruption is a crime and if we just passed stricter laws and enforced them more vigorously it would stop, and b) insurgents need to be stamped out. Alan offered that given certain levels of oppression or hardship, corruption is self sustaining and more closer to the truth and will likely not be eliminated.

Alan concluded by presenting two actions that can be taken by practitioners of system dynamics that can help with these challenges; a) follow a research agenda that aligns with a grand challenge, and b) write papers that build knowledge about the grand challenges. Which grand challenge will you choose?

Dennis Meadows presented the final talk of the opening plenary session entitled What Models Are Required To Understand Global Limits To Growth In The Next Two Decades? The industrialized democracies will experience more change in the next two decades, than they did in the entire 20th century. The change will be driven in part by the exploding effects of physical limits to growth such as water scarcity, declining availability of fossil fuels, climate change, and soil erosion. Growth stops when the forces that oppose it equal the forces that sustain it. The real strain to society is not after the peaks to growth, but pre-peak. How do we go about our work to influence the next two decades?

Currently most system dynamics work is done for corporations. In the seventies many systems were below their carrying capacity, but over the succeeding years approached or exceeded their carrying capacity. Once one goes above the system’s carrying capacity it typically deteriorates, e.g. our personal bank account. If we let it go unchecked it will organize on its own. System dynamics can help. There are two main classes of problems ahead, 1) Global Problems; affects everyone but requires international efforts, 2) Universal Problems; affect everyone and can be solved by local efforts. As individuals or groups of individuals, Dennis proposes that we can have a greater impact on focusing on universal problems to reap benefits sooner; e.g. soil erosion, water pollution, loss of market share, and energy scarcity. We can use one of two types of polices; a) Preventive policy, which avoids future problems, but requires action now in anticipation of future problems, or b) Adaptive policy; which reduces the harmful consequences of the problem already being experienced. System dynamics can assist by helping prepare the people to react. We can focus our work in system dynamics on adaptive policies that help solve universal problems; like water pollution, energy use and demand, improving our climate, preparing for interruptions in our food supply and reducing energy cost of collection and disposal of our system waste.

David Melton      

MON 11:00 AM Parallel - Intellectual Property, Knowledge and Sustainability

audienceSynopsis: Chaired by Deborah Anderson, this session included three mentally enriching papers. The session was well attended.

Full Report: Sustainable organizational change –can you make the change happen with large teams alone? by Felicjan Rydaz et al.
This paper provided overall insights on improvements and actions conducted by teams within a manufacturing organization. It compared two types of approaches: the bottom up approach, in which small, empowered, cross-functional teams are included and the top-down approach, in which large teams work together following rigorous problem-solving methodologies. The paper demonstrated the application of system dynamics in investigating the impact of small teams against large teams on organizational system structure. Through results obtained, the paper highlighted the following conclusions: there is a natural tendency of managers to look for order, discipline, and rigor, even during organizational changes; the high probability of large teams’ success might be the reason for favoring structured and planned actions like those proposed by RCM or Six Sigma; engaging the entire organization and development of socio-technical network via small cross-functional teams enables the spreading and sustaining peoples’ attention to defect elimination; and combining a set of interventions as a ‘package’ enables the organization to change and sustain the performance gains achieved.

speakerExploring the dynamics of music piracy by Trond Nilsen et al.
This paper explored the decline of sales of physical music media. The causes of this decline are currently poorly known. Trond presented three models that explored claims made in conversations over piracy and the music industry’s future. The modeled results of the paper indicated that music labels’ mental models are simplistic and do not reflect current research findings; artists can enhance their concert revenues through piracy. The free market and competition with other media explain part of the declining music industry revenues.

Modeling knowledge reuse in technical support operations by Mauricio Uriona et al.
The authors explored the increasing concern on the part of the corporate sector of the importance of harnessing knowledge as the most valuable resource of the corporation. The presentation emphasized the effects of knowledge re-use in service systems using a system dynamics model as a basis for drawing this conclusion. The presentation highlighted that the model aids the designer in evaluating several aspects of the system as well as its performance, including the effects of knowledge re-use, and that, based on simulation results, knowledge management enhances service system performance. Through results obtained from the model, the following conclusions were reached: knowledge bases impact service technical support operations; impacts on service operations (phone and e-mail support) are heterogeneous; and knowledge bases depend intensively on the speed knowledge is distributed among the organization.

Thokozani Simelane

MON 11:00 AM Parallel - Applied Microeconomics

Walking at the end of the aisle, you saw a great gathering of people discussing the economic mystery. Then, you heard the chair, M. Dolores Soto-Torres, call the meeting to order with his clear sound. Here began an enlightening and interactive discussion on applied microeconomics.

Business Dynamics Model for Market Acceptance Considering Individual Adoption Barriers, by Kirsten Matheus (presenter) and Thomas Harbich
The authors illustrated that marketing and technical implementations are often only partially responsible for adoption barriers. Consumers additionally need to overcome personal barriers to adopt an innovation. This article provides a feedback loop between the number of existing adopters and the number of consumers prepared to adopt the innovation. It presented the model that can reproduce tipping points and visually explain slow market developments or unexpected late successes.

speakerCapacity Utilization in Electricity Market Behavior: An Experimental Analysis by Santiago Arango (presenter) and Jaime Andrés Castaneda Acevedo
The presentation detailed an experiment to study the effect of variable capacity utilization over the behavior of an electricity market.Tests show that one market presents well-defined and pronounced cyclical behavior, while another two markets present minor oscillations. Results suggest that varying capacity utilization favors stability rather than cyclical behavior.

Economic Origins of the Mafia and Patronage System in Sicily, by Jason Sardell (presenter), Oleg Pavlov, and Khalid Saeed
This paper described that organized crime satisfies a demand for necessary social services in the absence of effective government institutions. Based on the experiments, the authors found that the most effective policies for reducing mafia power are those that decrease the demand for private protection services by reducing the threat of banditry in society and not those policies that act against the mafia directly.

aud aud

Many questions and responses were raised after the presentations that made participants feel that a great interaction and communication were wrapped in a pleasant atmosphere.

Min-Ren Yan

MON 11:00 AM Parallel - Information and Project Management

Synopsis: Luis Luna-Reyes presented the paper Using Institutional Theory and Dynamic Simulation to Understand Complex E-Government Phenomena, co-authored by J. Ramon Gil-Garcia. Luis said that governments hope to gain important benefits from e-government initiatives; however, about eighty percent of these initiatives fail. Luis discussed how organizational forms (e.g., centralization, formalization, and communication channels) and institutional arrangements (e.g., laws, regulations, and cultural constraints) affect how that organization implements technology. The authors’ model reproduced different technology enactments consistent with content development networks present in an e-Mexico program. When asked about his main recommendations, Luis said that trust was important in building the collaborating relationships needed for technology enactments. He also said that strong leadership and the existence of a previous network were key components in the creation of a project team. A good balance on relationships, results, and process-orientation are also important in the capitalization of the efforts of the team.

Kim van Oorschot presented the paper Dynamics of Agile Software Development, co-authored by Kishore Sengupta and Luk van Wassenhove. Kim began with a discussion of the problems with waterfall development (e.g., long delays, high costs, and low quality).  She said the waterfall approach is not flexible enough to deal with frequent requirement changes. She discussed how agile development claims to overcome these problems. Agile development divides the total development effort into a series of short “design, code, and test” iteration cycles (sprints), each with its own start and end dates. Kim and her team modified an existing software project system dynamics model by Abdel-Hamid and Madnick (1991) to test the effect on project performance of different iteration periods (sprints). The periods varied from 20 to 260 days with each agile scenario having a fixed period for all its sprints. In addition, for each agile scenario, the team looked at two cases, one with no requirements changes during the entire development effort and one with changing requirements. Her team concluded that mid-levels of agility outperform low and high levels of agility in terms of quality, resource costs, and development time. These results are consistent with the average schedule pressure which is lowest for mid-levels of agility. Someone asked about the assumption that thirty-five percent of all tasks in need of completion are unknown at the start of the project. Kim said that Abdel-Hamid and Madnick made the same assumption, and, since it was based on a real project, her team kept it. Someone else asked if she varied the duration of the iteration period for a given agile scenario. In many agile development projects, the early sprints are longer than later ones. She said they had not done that but that it was worth trying.

Hazhir Rahmandad presented the paper Deciding on Software Pricing and Openness under Competition, co-authored by Thanujan Ratnarajah. Hazhir said that if firms want to encourage open source contributors and complementary products, then they need to decide on an openness policy (i.e., whether to share all or part of their code) and a pricing policy (e.g., how much to charge for their software over time). These decisions are based on how much the market values the software, which depends on price, product feature richness, complementary software developed for the product, and installed users. The authors conducted a series of model runs including two firms. They sought a strategic equilibrium between the firms by simulating competition using a game-based approach. The authors concluded that openness is the dominant strategy when reinforcing loops are strong, yet it cuts deeply into profits when firms are in competition. Finding equilibrium led the authors to conclude that it is possible to find an open-loop Stakelberg equilibrium in differential games using typical systems dynamics tools. Someone thought the optimization approach Hazhir and his co-author used was unusual since they sought a dynamic pricing policy, i.e., the best pricing policy at different intervals of time given a constant openness policy.

Full Report

Pascal Gambardella

MON 11:00 AM Parallel - Improving the Modeling Process

Synopsis: This parallel session was devoted to improving the modeling process, validation, and documentation. The session was chaired by Peter Milling. All the talks were well received and invoked a good deal of discussion and follow up.

Full Report: SILVER: Software in Support of the System Dynamics Modeling Process, by Nathaniel Osgood. The first presentation introduced an open source software tool being developed at the University of Saskatchewan for supporting model validation and version control. The tool is designed to JAVA to support multiple operating systems and is being designed also to support multiple simulation platforms.

A validation Methodology for System Dynamics Modeling,by Stefan Groesser The talk defined a formal method for validating models. Stefan introduced a validation hierarchy that outlined a series of formal validation steps. The methodology would help standardize how system dynamic models are validated. He also included a discussion on validating to the level needed. The more validation required, the higher the cost and the lower the probability of incorrect analysis. Each project should determine beforehand what level they need can afford to attain.

Methodological Changes Needed to Meet the World’s Grand Challenges, by Alan Graham
Graham posed some grand problems for the system dynamics community to pursue. He also proposed that, to answer some of the more complex problems, a combination of system dynamics and other modeling techniques such as agent-based modeling would be necessary. Alan also believes that the system dynamics community needs to publish more of our work in journals outside system dynamics. 

Jake Jacobson

MON 11:00 AM Parallel - Social Welfare & Public Policy

Synopsis: The first study examined the reference modes behind racial disparities in the United States and talked about how disparities are able to persist and endure, even in the absence of racist intent, due to feedback loops. Eric made the case for why this was a particularly important social issue, and why system dynamics is a key methodology for both building understanding around structural racism, but also being able to anticipate and visualize policy resistance and look for interventions that are both high-leverage and sustainable. He used the example of racial wealth disparities in the United States and looked at how a racist input into the system over fifty years ago (granting FHA loans based on race in the United States) had become much larger wealth disparities in the present day, even though the racist input had been largely eliminated, simply because of the effects of the reinforcing growth feedback loop that characterizes wealth and investments. Eric also talked about using group model building as a tool for community organizing.

The second presenter was Mr. Lopez from INCAE Business School. The study explored institutional failures in the implementation of deterrence policies using Costa Rica’s dataset to calibrate a system dynamics model. The model suggested that criminals tend to be punished not exclusively on the basis of their behavior, but in terms of other institutional variables. The final simulation model explained the causes of increasing crime rate by demonstrating a vicious cycle that creates direct and indirect costs to society and that, most likely, will not reduce crime rates.

The third paper focused on the Dutch Soft Drugs policy, soon to be under review. Regarding Dutch Soft Drugs policy, the Dutch population and political arena could be divided into three groups: those who do not care, those strongly in favor of legalizing cultivation and use of soft drugs, and those strongly in favor of banning soft drugs. The points of view of the two latter groups are analyzed in this paper using a qualitative system dynamics perspective.

Full Report

Chanyeong Kwak

MON 2:00 PM Parallel - System Dynamics and Agent Based Methods

Synopsis: This session, chaired by Jay Forrest, explored the relationship between system dynamic models which have traditionally been more aggregate in nature with agent-based models.The three talks explored different aspects of this relationship. Osgood’s work took the domain-specific concept of co-morbidities that serve as a significant source of heterogeneity in health and epidemiological modeling and evaluated different aggregate versus individual-based modeling approach for their handling. In general, a more individual-based approach was advantageous for desired modeling characteristics of modularity, flexibility, transparency, and expressiveness. Liu and colleagues’ work looked at the effects that assumptions regarding individual versus global rationality in decision-making may have on model outcomes for one well-known system dynamic model: the beer distribution game. The results showed that there are substantial differences between the game outcomes if individual agents in each sector use globally versus individually rational decision-making strategies. Kortelainen and Lattila’s work moved away from the comparative assessment of different modeling strategies to evaluating the potential for hybrid models for the specific application of strategic technology management. While such models combined the advantages from aggregate and individual models, the processing time proved to be cumbersome. This limited the size of the model as well as the opportunities for sensitivity analysis. The model seemed to validate many of the traditional discussions regarding individual versus aggregate-based models: there is always a trade-off between model complexity and how well it can address the real characteristics of a dynamic system.

Full Report

Katherine Dykes

MON 2:00 PM Parallel - Microeconomics: Price Dynamics

Synopsis: The first presentation in this session was Logical vs Historical Time in A Price Adjustment Mechanism, by Kaoru Yamaguchi. He examined the price adjustments mechanism built on logical time and then applied it to the real market economy, which is on historical time, using system dynamics modeling.

The second presentation was Platinum Supply and the Growth of Fuel Cell Vehicle, by Khalid Saeed. He first described fuel cell production in the US and raised the problem of limited platinum supply. By using system dynamics, he found what the platinum prices and cost of platinum for fuel cells might be in 30 years.

In the last presentation, Estimating Impacts of Water Scarcity Pricing, Jason Hansen described the water scarcity problem in the western US. Since the water is under-priced, over-consumption will lead to shortages. Jason has developed a model that provides optimal water use, proper prices, and resources constraints.

Full Report

Phuong Linh Nguyen

MON 2:00 PM Parallel - Analysis of SD Models and Their Behavior

Synopsis: The three presentations in the Analysis of System Dynamics Models and their Behavior session provided a multi-faceted view of the role of system dynamics modeling in promoting and supporting systems thinking. Delivered to a standing-room-only audience, the speakers highlighted the importance of exploring and understanding the complex interactions within a system dynamics model, between modeled domains, and between modeling methods.

Full Report: Objective Analysis of Subjective Feedback Structures: The Problem of Consistency in Explaining Model Behavior by Mohammad Mojtahedzadeh
This paper addressed the issue of auxiliary feedback loops that can exponentially scale model complexity, but may not change the dynamics. Use of pathway participation metrics (PPM) was proposed to assess loop sensitivity and identify dominant loops that define the relevant scope of the model. Consistency in interpreting model behavior was a key goal, and an approach similar to renormalization in quantum physics (which assumes the infinite possibilities which cannot be modeled are self-balancing) was used.

Real Estate Cycles: A Theory Based on Stock-Flow Structure of Durable Goods Markets, by Ali Mashayekhi, Soheil Ghili, Arash Pourhabib
The authors addressed a scenario in which the aggregation of values of system components did not yield a true picture of the behavior of the system. Arash Pourhabib presented that the owner-occupied and rental segments of real estate markets cannot simply be aggregated to model the dynamics of the entire system due to differing cycle-producing mechanisms (CPMs), which also interact with one another. To achieve a holistic picture of the system, the interactions between CPMs needed to be identified and modeled, since they create new paths in the system that may not be apparent in its major components.

Using Binomial Decision Trees and Real Options Theory to Evaluate System Dynamics Models of Risky Projects, by Burcu Tan (presenter), Edward Anderson, James Dyer, and Geoffrey Parker
The limitations of the traditional decision analysis modeling methods were discussed, including high mathematical complexity as well as opaque and unrealistic assumptions. System dynamics modeling enables a more comprehensive view of the risk environment and the interactions between drivers. The basic algorithm that proposed was to identify the real options, build the system dynamics model, specify random processes and distributions, run Monte Carlo simulations of the system dynamics model, obtain the cash flow payout rates for each period, and estimate the volatility of the project value by searching for a volatility that best mimics the uncertainty in the original model.

Joseph Glick

MON 2:00 PM Parallel - Organizational Dynamics

Synopsis: The session, chaired by Lascelles Anderson, contained three papers dealing with various aspects of training issues in performance. The topics covered in the session included goal dynamics in product sales performance, post-adoptive enterprise resource planning systems behavior, and training and recruiting in an air traffic control organization.

Francesco Ceresia presented his paper, A Model of Goal Dynamics in Organizations: A Case Study. The author examined a system dynamics model for goal dynamics in organizations with a focus on modeling goal setting, managing by objectives, and training to enhance worker performance in a product sales context.

Mary C. Jones presented the paper Post Adoptive ERP Use Behaviors: A Dynamic Conceptualization, co-authored with Thomas D. Clark, Jr. and Robert W. Zmud. The authors proposed a system dynamics model of enterprise resources planning (ERP) user behavior with a focus on the role of various interventions triggered by a gap between expected and actual system benefits.

Andreas Größler presented the paper Supporting Long-Term Personnel Planning of a Service Provider, co-authored with Alexander Zock. The authors examined the impact of delays arising from training, promotion, and personnel ordering on the recruitment and training processes for air traffic controllers.

Full Report

Mary C. Jones

MON 2:00 PM Parallel - Evaluation of Participatory Methods

Chair: Jac Vennix

Is Group Model Building Worthwhile? by Marleen McCardle-Keurentjes, Etiënne Rouwette, Jac Vennix, and Eric Jacobs

Etiënne Rouwette presented a paper that attempts to evaluate the effectiveness of Group Model Building (GMB). In total, one-hundred-and-seven available field studies describe successful applications of model-driven group decision support methodologies. However, these studies might suffer from a potential bias of contextual factors. In addition, there seems to be no standard set of outcomes for evaluating the effectiveness of group decision support. The purpose of the paper is to determine the effectiveness of GMB on decision-making in a controlled research situation. In an experimental design study, student participants had to solve a complex problem (Hall 1973, The Saturday Evening Post). Thirteen groups were introduced to GMB, and another thirteen groups were introduced to a standard group decision support method. The Competing Values Framework was used for clarification and measurement of effectiveness. The researchers compared the groups’ perceptions of the quality of decision-making and the quality of their decisions. Preliminary results show no winners; at first, no surprising results occurred. On a second inspection, however, a gender effect influences the groups’ effectiveness significantly. Current psychological research supports this conclusion.

The ensuing discussion focused on the time available to the participants to solve the problem, one hour, which might be too short for the full effects of GMB to precipitate. Further research is required to determine why gender is relevant to the difference in effectiveness.

Group Model Building Effectiveness: A Qualitative Method to Assess Changes in Mental Models? by Brigit Fokkinga, Inge Bleijenbergh, and Jac Vennix
Brigit Fokkinga presented a paper on the changes in mental models occurring after a five week training course. The researchers used the Ford and Doyle definition of a mental model. They chose an approach to reconstruct individual mental models before and after the intervention. This way, they could make various comparisons between individual pre- and post-test models. With the help of written questionnaires, respondents were asked to answer the following four questions before and after the Group Model Building process: 1. Please write down the central problem at stake. 2. Please describe the causes of the problem. 3. Please describe the effects of the problem. 4. Please mention three options to solve the problem.

The written answers were analyzed and coded by two of the researchers and converted into cognitive maps. The comparison focused on the difference in the average numbers of variables and feedback loops. The researchers determined that the content of the variables changed for the GMB sessions participants, while they remained the same for the control group. They did not find a direct effect of the causal modeling on the amount of feedback loops respondents reported. Neither the group participating in the model building nor the control group reported more feedback loops after the causal diagram was produced and reported. The ensuing discussion brought forward the idea that the instrument used to elicit the mental models might have a significant effect on the results obtained. David Andersen, George Richardson, Colin Eden, and Fran Ackermann conduct research related to this idea (see respective publications).

Experimental Comparison of System Dynamics versus Traditional Facilitation, by Krystyna Stave and Marcia Turner
Marcia Turner addressed the question of whether system dynamics can improve group decision making. In an experimental setting with more than 200 participants, traditional facilitation methods and Group Model Building were compared. The researchers employed a pre-test and post-test control group design. On average, the groups using system dynamics as the facilitation method obtained better results in terms of identification of high quality leverage points (H1 supported). The researchers were also interested in how much the discussions focused on the subject during the decision making. The group facilitated with system dynamics had a higher degree of focus, which supports H2. The last hypothesis related to the degree of satisfaction of the participants. It turned out that the group using system dynamics had a lower average degree of satisfaction (H3 not supported). This might be because the time available for the group sessions was short, and, therefore, the simulations used for facilitation could not be explained sufficiently. The most interesting result may be that the groups using traditional facilitation methods, on average, demonstrated a lower performance and lower degree of focus, but were more satisfied and felt more confident than the group using system dynamics facilitation methods. The system dynamics groups performed, on average, better with a higher degree of focus, but were less satisfied and less confident in their decisions. This opens an interesting path for further research.

Stefan Groesser

Back to Contents in page 1

Page 1, 2, 3, 4, 5, 6, 7

How to Contact Us
System Dynamics Society

System Dynamics Society Logo
Milne 300 - Rockefeller College, University at Albany
Albany , New York 12222 USA
Phone: +1 518 442 3865                Fax: +1 518 442 3398
The System Dynamics Newsletter is published four times a year by the System Dynamics Society.