The system dynamics group at the Rockefeller College of the University at Albany has been developing techniques to create system dynamic models with groups of managers during the last 25 years. Building upon their tradition in decision conferencing, the group has developed a particular style that involves a facilitation team in which people plays different roles. Throughout these years of experience, the group has also developed several scripts to elicit knowledge from experts based on small-groups research, and well-established practices in the development of system dynamics models. This paper constitutes a detailed documentation of a relatively small-scale modeling effort that took place in early 2001, offering a soup to nuts description of Group Model Building at Albany. The paper describes in detail 8 of the scripts that the group has developed, offering some reflections about their advantages and limitations.
System dynamics requires the intense use of qualitative data and human judgment in all stages of model development. Most approaches to the formal inclusion of qualitative data have been developed with the purposes of knowledge elicitation during the conceptualization or formulation stages of model development. Although the importance of using expert judgment to assess the validity of system dynamics models is well recognized, the development of approaches to use this kind of judgment is not well developed. In recent years, efforts to develop tools to assess the validity of system dynamics models by interviewing experts have been explored in some doctoral work. This paper reviews the basic concepts of model validation, and explores the use of interviews as a research and knowledge-acquisition technique. Finally, it documents and compares four applications of interviewing as a tool to assess system dynamics models, ending with recommendations for both the practitioner and researcher.
The author has created a system dynamics model to investigate how health care providers can and should respond to increases in patient demand for treatment above usual levels. This response by the health care system is called surge capacity and is an important issue in emergency and disaster planning and response. The model describes how hospital and home care treatment providers can alter their internal staffing and patient treatment policies as well as movements of staff and patients between each other. These providers can fail to respond adequately to surge events by exhausting their staff or by moving too much burden from the hospital sector to the home care sector.
Based on an in-depth field study in an electronics plant in Singapore, this paper examines the dynamic interaction between two of the key functions, Production (P) and Manufacturing Engineering (ME). P and ME are responsible for process execution and process development respectively, and for process smoothness jointly; their relationship is asymmetrical if judged from organizational and structural aspects. The paper reveals the causes and effects of three types of short-sighted functional behavior burden-shifting, resource-fighting, and corner-cutting. The resulting P-ME conflict due to short-sighted behaviors is analyzed in a qualitative system dynamics model. Although this research is based on a single firm, the findings have implications for many contemporary plants where the proliferation of new processes puts stress on the P-ME interface. Future researchers can use more samples to test and theorize the findings of this research.
Agent-based modelling seems to be an alternative way of modelling to System Dynamics. Criteria for discriminating the methodologies, and criteria for the choice of which one to use, still remain vague. This study compares both approaches on an empirical basis, utilizing an exploratory experiment aimed at investigating the respective comprehensibility of each methodology. The gained results, considering all the observations, show no significant differences between the two treatments. Nevertheless if the subjects are grouped into SD students and non-SD students, differences are observed. Interestingly it shows an advantage of the AB approach for the SD student group, whereas the non-SD students seem to have an advantage with the SD methodology.
We explore organizational forgetting, the notion that firms knowledge can be lost through human capital decay. An in-depth case study research, which is guided by the conceptualization of a system dynamics model, is conducted. The evidence appears to support the presence of forgetting. This gives rise to the possibility of productivity falling in spite of continued output accumulation, due to changes in the characteristics of the resource where experience resides. Most prior research on learning curves, however, assumes that productivity will always increase with cumulative firm output.
Education is considered one of the main drivers of welfare in society. However, countries in the world follow different paths when creating basic human capabilities, many of them not in the right direction. Linear extrapolation is still widely utilized to predict future behavior based on statistics like the Primary Completion Rate. This paper presents a dynamic model of primary education as a first step to understand the structure and behavior of educational systems and as an alternative way to extrapolate outcomes of this and other relevant key indicators, like the Gross Enrollment Rate. The model is calibrated for the case of Nicaragua.
It has long been thought that simulation could be used to design Command and Control (C2) systems, but simulations benefits have not matched their promise. Instead Enterprise Architecture Planning (EAP) tools have become ascendant in the design of C2 systems, though problems remain. EAP tools break down proposed systems into their low-level, constituent parts and place them into sophisticated relational databases. The resulting architectures however do not yield an intuitive sense of whether the proposed system actually solves the motivating problem. Consequently, fundamental conceptual issues continue to emerge deep into the design process. This study proposes using simulation early in the design process to envision the total system and avoid problems by generating requirements and metrics early in the design process. Issues regarding an Air Force Air Operations Center (AOC) are explored, most notably flow of control and the coordination of sensor, decision, and operator assets.
The objective of this study is to explore the factors that influence the quality of group learning and group effectiveness in organizations. Learning enables groups to acquire new skills, improve processes, find new ways of working, and enhance their decision-making process. However, group learning is affected by a set of structural, cognitive and interpersonal factors, which may foster or hinder the engagement of group members in learning-oriented activities. This study regards work groups as complex social systems and suggests that the explanation of the quality of learning and the effectiveness of a group lies in the interrelations of these factors. Existing research on group learning tends to follow an input-process-output approach; in contrast, this study offers a system dynamics model to explore the intricate relationships that arise from the factors such as group dynamics and leader behavior and that influence the outcomes of a work group. Although the model is highly aggregated, the simulation results can improve our understanding of the interrelations of key factors that influence group learning and effectiveness and farther the path for future research using system dynamics to study work groups as complex systems.