System Dynamics as a methodology has traditionally been concerned with the study of processes that can be described by continuous variables. Discrete or integer events, such as the number of sales made in a day or the number of factory closings in a year have either been approximated as continuous variables or else not dealt with. This paper examines another way of dealing with discrete events through the realization that any discrete event has a certain probability of occurance. These probabilities are continuous and conserved quantities and can be modeled as system dynamic levels. Treating probabilities as levels in dynamic simulations is a standard technique in stochastic modeling, markov models being one example. System dynamics' advantage over these other methods is that it can represent the impact of the results of the probabilistic study of the social feedback systems. This paper focuses on examples demonstrating the use of system dynamics to model uncertain events. These examples deal with the simple case of a Poisson process with a time varying event arrival rate. Extensions incorporating conditional and independent probabilities are also considered.
The premise of this paper is that System Dynamics has, in the past, been primarily perceived, both by external observers and by most of its own practitioners, as a technique of computer simulation. Although this situation is changing, there is still little wide scale recognition of its true generality and relevance as a complete subject of systemic enquiry. The purpose of this paper is to explore the merits of system Dynamics as a total systems methodology. Specifically the presentation will undertake to review the need for and the requirements demanded of such problem solving methodologies, to briefly explore the dilemma resulting from historic attempts to create them and to present changes to the existing Systems Dynamics method which might improve its conformity and acceptability as such a methodology. These include the formal definition of Qualitative System Dynamics and the presentation of a set of rigorous rules to provide much needed guidance in its application; firstly, Stepwise Influence Diagramming, aimed at enhancing problem exploration and model development and secondly, Qualitative analysis, aimed at identifying critical system components and exploring the effects of change.
In the health sciences, concepts are shifting toward system models which recognize multiple factors interacting to determine health phenomena. The hybrid biomedical disease model has proven insufficient for the analysis of modern health problems. A population perspective and an expansion in the influence of the behavioral and social sciences have required conceptual models with greater breadth, and facility in relations between models. Morbidity is portrayed here as two domains of phenomena, the disease process and the illness state, each seen as part of a socio-ecological dynamic. Applied to major disease problems, the utility of these propositions can be examined. In the McMaster M.D. program, this set of models has been translated into a curricular structure which has the individual in all her/his healthy or morbid aspects as the interface between biological and social systems. Perplexing dilemmas in health care thus become not only understandable but predictable. Adopting this approach creates a new generation of problems. Just as our students have become familiar with the critical appraisal of evidence, the testing of conceptual models becomes a necessary skill. The background of this analysis is the socio-ecological niche of concepts. A model of models is proposed in which concepts interact with problem environments and modern medicine emerges as a case study for socio-ecological epistemology.
Much of the literature on model evaluation focuses on what amount to absolute measures, that are independent of the context in which a particular model is used. This paper argues in favor of situation-dependent measures. Whether or not a model is “good enough” depends on the job it is being asked to do and the mid set of the people who must use the results. The relationships between model adequacy and successful implementation of model-based recommendations are discussed. While rejecting the classical paradigm, the author emphasizes model realism and historical accuracy as important determinants of implementation. The life of the model involves many evaluations of whether it is “worth the costs”, “believable”, “useful”, and “right”. Issues surrounding these judgements are explored. How differences in circumstances can lead to different, but in each case quite adequate, models is illustrated by contrasting two models developed five years apart for the same organization. The paper concludes that successful models are persuasive, not simply to modeling technicians but to high-level decision makers.
A disaggregate population model of China is presented. The age structure is represented by one-year cohorts. Urban and rural populations are distinguished. Birth and death rates, family size, life expectancy, and other demographic variables are determined endogenously. The model can be used to analyze population problems and to project population size, the age structure, the adult labor force, the elderly population, and so on. The model can be used in two modes. It can be used to project the consequences of various exogenous fertility levels. Alternatively, birth rates and fertility can be determined endogenously by economic inputs such as food supply, GDP, and services. The model incorporates socioeconomic factors important in the demographic transition, such as the effect of perceived life expectancy on fertility, the effects of traditional values, and the ability of government to influence family fertility choices. The model can be used to evaluate policies and programs designed to control population growth, such as delayed marriage age, improved contraception, and restrictions on family size. The model requires industrial, service, and food output per worker as inputs, and also the level of pollution. The model should be thought of as a component of a comprehensive planning model which generates these inputs endogenously. Based on the system dynamics approach to modeling complex systems, the model is implemented in the DYNAMO simulation language.
This paper uses a system dynamics simulation methodology to assess the potential effects of new accounting policies being considered by rule formulating bodies. The key objective of this paper is to demonstrate that current ex-ante intuitive assessment of the effect of proposed accounting rules is inadequate due to the counterintuitive nature of economic consequences in a complex social system. For this purpose a very simplified model of he US economy is developed and its parameters varied to reflect potential accounting policy changes. The effects of these policy changes are shown to be counterintuitive in nature, requiring consideration of second and third harmonics of the feedback loops for adequate ex-ante impact assessment. This paper is divided into six parts: the first part describes alternate approaches to economic consequence assessment and the advantages and disadvantages of utilizing the system dynamics methodology; the second part describes the skeleton of the system dynamics model; the third part examines the measurement problems of rates, levels, and delays as well as reviews the details of their computation for model formulation; the fourth part discusses the problems and results of model validation efforts; the fifth part describes some of the results obtained from application of the model, and their meaning in comparison with traditional methodologies for accounting impact analysis; the sixth part concludes by suggesting the next step for macro-accounting modeling: evaluating the potential and shortcomings of this methodology.
:One of the traditional obstacles to effective utilization of simulation models has been the great deal of time spent learning languages in which models are written and keeping track of the specific variable names and equations within models. To remove the excessive psychological burden from busy executives and to refocus attention towards the actual behavior being replicated, Inter/Consult has been researching development of highly supportive user interfaces to models. These interfaces prompt users by stating the nature of the model’s assumptions then asking what changes they would like to make. Through this on-line question-and-answer dialog users can build and compare scenarios without prior knowledge of computer languages and mathematical formulas or specific model components. Our paper presents reactions to the interface by members of the graphic arts industry who have used it. We discuss further improvements which are being made to the interface to make our models more accessible to non-expert users. Finally we explain why we feel that tightly-focused, easy-to-use, dynamic simulation models are of invaluable benefit to any industry such as graphic arts where craft-oriented skills are being replaced by rapidly evolving new technologies.
In this paper some of the ideas of Ortega y Gasset about the dynamics of history have been gathered and organized according to the system dynamics diagrams. A cyclic process, characteristic of every normal course of history, is described as well as the dynamics hypothesis responsible for it. Human life, as far as it affects history, is shown as being composed of five age groups each of them covering fifteen years of life. Two of these groups, two generations acting simultaneously in the field of history, are presented as taking the main responsibility for the dynamics of history.
Many firms use financial ratio analysis to monitor their control over the operating cycle and to serve as the basis for policy formation. Ratios are based on data produced through the accounting information system which is analyzed according to intuitively plausible concepts in order to make normative judgement about the financial health of the firm. A model is constructed to simulate the operating cycle of a business which generates financial ratios in a manner analogous to the accounting system. It is shown that noise and seasonality produce distortions in the ratio measures are spread throughout the system in a dynamic and complex fashion. Further experiments reveal that plausible control policies based upon financial ratios may make performance worse rather than better. System Dynamics appears to be a useful approach both to redesigning financial ratio measures and testing policies which could enhance out ability to manage such systems.
Sensitivity testing, according to the glossary of terms in a Congressional manual on simulation modeling, is defined as the “running of a simulation model by successively changing the states of the system…and comparing the model outputs to determine the effects of these changes” (Congress 1975, p. 129). Sensitivity testing is generally viewed as an important part of the modeling process because it helps researchers narrow down those areas where more data gathering would be useful. In our introductory remarks, we argue that detailed sensitivity testing is particularly important in system dynamics modeling efforts, and we list several obstacles that make detailed sensitivity testing difficult. We introduce a set of testing procedures developed at the Los Alamos National Laboratory and verified by the Control Data Corporation that can help system dynamicists perform detailed sensitivity testing on a routine basis. In the body of the paper, we present an illustrative application of the testing procedures, and we list six specific uses of the procedures. We describe the availability of the testing package, and we conclude with a set of practical guidelines for investigators wishing to make use of this unique set of procedures.