21st MIT-UAlbany System Dynamics Ph.D. Colloquium
Friday, October 29, 2010
University at Albany, State University of New York,
Milne 215

Organizers: Junesoo Lee (UAlbany) and David Keith (MIT)

Time Presentation Speaker
10:25 AM Introductory remarks
George Richardson, University at Albany
10:30 AM Potential of Group Model Building in Environmental Management Johannes Halbe, McGill University
11:15 AM Modeling Cases from Jared Diamond's "Collapse" Katrina Hull, University at Albany
12:15 PM Lunch .
12:45 PM Dealing with policy paradox by employing "Failure Management" & "Success Management" Junesoo Lee, University at Albany
1:30 PM A Systems Modeling Approach for Improving Oral Health Outcomes of Older Adults
Michael J. Widener, University at Buffalo
2:15 PM Completeness of Product Diffusion Models to Explain Dynamic Complexity
Stefan Grosser, University of St. Gallen (visiting fellow at MIT)
3:00 PM Closing Remarks George Richardson, University at Albany


Johannes Halbe, McGill University

Group model building for organizational learning has been applied frequently in business and the public sector over the last couple of decades. However, the application of group model building for the exploration of environmental problems is a more recent development that poses particular methodological challenges such as: the involvement of diverse stakeholders; the need to continue participatory processes in the long-term to achieve social learning; and the need to consider environmental processes in the modelling and management process. The Participatory Model Building (PMB) Framework was developed to overcome these barriers and to facilitate the application of systems thinking and system dynamics modelling in environmental management. The methodological framework consists of five stages: i) the structured analysis of a given problem (Stage One); ii) identification of stakeholder groups (Stage Two); iii) construction of Causal Loop Diagrams in individual interviews (Stage Three); iv) group model building via a system dynamics water balance model (Stage Four); v) and institutional analysis of the resource management regime (Stage Five). In particular, the third stage is considered as critical in the implementation of systems science in environmental management practice. A case study on water scarcity management in Cyprus demonstrates the relevance and effectiveness of Stages One to Three of this proposed framework. The results highlight the heterogeneity of perspectives, which in turn underline the need for participatory and interdisciplinary approaches in water management. Due to the active participation of the stakeholders and the diverse results of this study, future research in Cyprus will extend to Stages Four and Five. In addition a further case study will deal with flood and water quality management in Quebec, Canada.

Katrina Hull, University at Albany

Throughout history some societies, including the Maya, Anazi and Easter Island, have collapsed, while others facing similar challenges, such as New Guinea and Japan, have succeeded. The cases of the Maya and New Guinea were taken from Jared Diamond's study, "Collapse," of how these societies succeed or fail, to create a system dynamics model capable of producing both the collapse and success behavior. The endogenous pressures described by Diamond were used to develop the feedback story. The model consists of five sectors: food supply, natural resources, land management, population and the society's response.

Junesoo Lee, University at Albany

It is a common experience for anyone or any public policy to encounter two types of paradoxical situation. The first one is “negative policy paradox” in which a success can prevent us from improving mental model by blinding us through overconfidence. The second one is “positive policy paradox” through which we experience an unexpected benefit from a failure. One of such positive paradoxes – called “reforming” − was simulated in a model where income and health can be negatively correlated. And policy leverages which can be called “Failure Management (FM)” and “Success Management (SM)” were operated to deal with such policy paradoxes in the model. It is expected for this model to be developed further by involving other principles of failure management in the future.

Michael J. Widener, University at Buffalo (presenter) - Sara S. Metcalf, University at Buffalo - Mary Northridge, New York University and Columbia University - Bibhas Chakraborty, Columbia University - Ira Lamster, Columbia University - Stephen Marshall, Columbia University

As the population of older adults in the United States continues to grow with the aging of the Baby Boom generation, the advantages of developing coordinated and cost effective health policies is becoming increasingly salient. This interdisciplinary research focuses on improving the oral health of older adults as a means of enhancing their overall wellbeing and quality of life. Periodontal disease is a risk factor for other chronic illnesses, notably diabetes and cardiovascular disease. Despite this link, medical and dental disorders are rarely screened for and treated as related conditions. Additionally, access to dental care for older adults is oftentimes adversely affected by lack of insurance coverage and complex social, cognitive, and physical factors, which results in missed opportunities for intervention and prevention of more serious conditions.

In order to identify policies that improve oral health for older adults, a dynamic modeling approach that considers community and individual level factors is utilized. A causal map was developed in consultation with dental practitioners who lead the ElderSmile program at the Columbia University College of Dental Medicine to establish relationships among the social and environmental factors that influence access to and use of oral health screening and treatment services in northern Manhattan. The relationships between these factors are used to identify the critical feedback mechanisms to be implemented in a hybrid AnyLogic model employing agents, social networks, GIS, and stock-flow structures. Preliminary results from this model will be presented.

Stefan Grosser, University of St. Gallen (visiting fellow at MIT)

Most product diffusion models are simplistic in that they can generate s-shaped growth behaviour, but fall short to account for more dynamic behaviour patterns. A mathematical approach is used to measure the ability of an existing model to account for a wide variety of diffusion patterns and thereby reason for the completeness of the model. By this, the models can be compared to each other. This result seems to be a novel approach to evaluate product diffusion models. The presentation emphasizes, in addition, the distinction between “completeness of a model”, “dynamic complexity of reality”, and “the dynamic behaviour a model can generate”.

[ Back to the previous MIT-Albany colloquia ] [ Home ]

Last edited by NG 11/08/2010