Analyzing Complex Societal Problems





Dorien J. DeTombe

Cor van Dijkum


This book is the result of the First International Conference on Methods and Tools for Analyzing Societal Problems that took place in November 1994 at Delft University of Technology and Utrecht University in the Netherlands.




DeTombe, Dorien J. & C. van Dijkum

Analyzing Complex Societal Problems: a methodological approach / Dorien J. DeTombe & C. van Dijkum (Editors)

Munich; Mering; Hampp, 1996 With ref. - With authors description



ISBN 3-87988-189-8 NE: DeTombe, Dorien J.[Hrsg.]

Subject headings: methodology/ tools/ simulation/ complex societal problems

ã 1996 Dorien J. DeTombe



Dr. Dorien J. DeTombe

Chair International - , Euro - , West-Euro- & Dutch Operational Research Research Group Complex Societal Problems & Issues

P.O. Box. 3286, 1001 AB Amsterdam, The Netherlands, Europe

Tel: +31 20 6927526 E-Mail:

http: //



Publisher: Rainer Hampp Verlag Munchen und Mering Meringzeller Str. 16 D-86415 Mering Germany Fax + 49 8233 307 55



See for other parts of the book: 








Many societal problems are complex and interdisciplinary. They involve many parties and phenomena, and have a large impact on society. These problems are seldom well defined, change constantly and are very hard to handle. The data, knowledge and know-how needed for adequate handling of these problems are often insufficient, incomplete and/or in contradiction with each other.


Recently a method has been developed which supports the analysis of societal problems, namely Compram, which stands for Complex Problem Analyzing Method developed by DeTombe. The method is based on the idea that complex interdisciplinary societal problems should be handled cooperatively by a team. This team discusses the problem and gives suggestions for changes. The approach the methods dictates, starts to analyze the problem with a team of 'neutral' content experts in order to get a clear picture of the problem. The content experts are selected based on their knowledge of one of the domains, or of that of one of the parties which are involved. After the 'neutral' team has analyzed the problem, the problem will be discussed by the different parties. Then a representative selection of all the teams select the interventions and guide the implementations. In order to support the information exchange between the team members the problem will be expressed in different models using different languages. The problem handling process is supported by a facilitator, who uses special group facilitation tools to support the information exchange. The method can be applied for all kind of policy problems.

click here for reading the article





Cor van Dijkum


From the viewpoint of the methodology of science the question will be reflected in what way disciplines, such as psychology, sociology, economy and computer science can work together to handle complex social problems. Special attention will be given to the question which languages (or set of languages) are best suited for the communication and exchange of information between disciplines. The language of cybernetics (of the second order) is suggested as the most promising candidate. The practice of simulation and gaming is pictured as a sensible model of this language. Methodological questions and problems which could arise in this point of view are handled in the context of constructive realism. The idea of strangification, which has been developed in constructive realism, is used to clarify the process of dealing with concepts e.g. the transformation of concepts from one discipline to another discipline.





Frans A.J. Birrer

Research tools are instruments through which we look at the world. As such they may transform problems, and they may even shift attention away from a problem without it having been solved. Sometimes such a shift is desirable, but sometimes it is not : the original problem is repressed simply as a way of ostrich policy, or the researcher is paying only lip-service to a problem. Undesirable problem shifts are particularly persistent at interfaces where different cognitive spheres have to interact, e.g. when decision making by different societal groups is required, or when the problems are of a complex and interdisciplinary nature.

The phenomenon of problem shift is hardly exposed in the literature on problem solving, and the language in use tends to obscure it. In this article, we will take a few explorative steps towards an understanding of problem shift. Some types of problem shift will be discussed, and it will be examined what can be done about them.




Using system dynamic MODELING techniques for constructing scenarios of societal problems

Dorien J. DeTombe & Harm 't Hart

A system dynamic simulation model expressing the relations between the aspects of a societal problem can be used to make scenarios. Scenarios can be defined as explorations of future development (Jager, 1990). Scenarios are useful for making policy decisions only if the model and the data in the model represent reality. Given the uncertainties in a model of a societal problem we raise the question in how far a system dynamic simulation model of a societal problem can be used for future exploration and future prediction of the development of a problem. In reflecting on this question we discuss criticism of the use of system dynamic modeling for scenarios coming from the theory of complex societal problems (DeTombe, 1994), from chaos theory (Gleick, 1987), and from system theory (Flood & Jackson, 1992). We conclude that models which express societal problems contain much uncertainty. Scenarios based on these models will also contain a large amount of uncertainty. This makes it hard to make reliable predictions based on these scenarios for the future development of the problem.




Static vs. Dynamic Modeling in Policy Analysis:

The Social Amplification of Risk

Giampiero E.G. Beroggi & Markus Aebi

Many different modeling approaches in policy analysis have emerged over the last few decades. Examples are statistical data analysis, covariance structures, influence diagrams, mathematical programming, simulation modeling, and expert systems. Most of these modeling approaches have been implemented into user-friendly software systems which promoted the widespread use among policy analysts. The purpose of this paper is to compare two frequently used modeling paradigms in policy analysis: static and dynamic modeling. The static modeling approach is discussed with the concept of probabilistic influence diagrams using the software DAVID, while the dynamic modeling approach is analyzed with the simulation modeling approach using the software iThink. After introducing and comparing the two modeling paradigms, the findings are applied to policy analysis in risk management. The social amplification of risk, as introduced by Kasperson et al. [1988] and operationalized for influence diagrams by Burns [1991], is used as a case study. The paper closes with a research agenda.



Some aspects of management of organizations in turbulent and complex environments.

Tineke P. Bahlmann

Organizations, whether for profit or not-for-profit, are more and more faced with rapid changing environments. The environment of organizations consists of clients, suppliers, competitors, bankers, shareholders, employees, society, politics etc.: the stakeholders. In this article will be described how organizations in rapid changing environments can survive and how these surviving organizations are managed. That does not mean that the organization only needs a couple of methods and tools to cope with the rapid changes, but it also means total different ways of organizing. The characteristics organizations need in rapid changing environments will be described, and the tools and methods that the organization can use in order to survive. The information in this article is based on research published in a book called 'Denken & Doen'(Bahlmann en Meesters, 1988), and action research in a lot of organizations in the profit and non-profit sector: banks, universities, highschools, production companies, commercial companies.

First 'rapid changing' environment will be explored. Then the old model of organizations will be described and why organizations according to this model are not able to survive in changing environments. The third part will elaborate on stable and dynamic qualities. The fourth part is about the qualities of organizations in rapid changing environments based on the above-mentioned research. In the fifth part some tools and methods will be described for organizations in fast changing environments to increase their ability to survive. In the last part some differences between profit and non-profit organizations will be discussed.



Simulation as an approach to support institutional design

Ruurd R. Meinsma

Decision makers within organizations are looking for means to deal with markets and with other organizations in order to realize organizational goals. One way to handle problem situations in which more organizations are involved is by institutional design. From a design perspective a decision maker needs insight into interactions of the organization with its markets, formal and informal relationships with other organizations, and its relevant organizational processes.

Simulation can be used to analyze problems of organizations in their changing environments and to evaluate designed solutions. A case study has been carried out in order to acquire guidelines for the development of appropriate support tools. In this case study, the approach consisted of three tasks: problem identification, model specification, and looking for solutions. A simulation model has been developed for analyzing the current situation and examining responses to several scenarios. Because of the fact that decision makers need to deal with multi-actor situations, multiple perspectives, and dynamic environments simulation appears to be promising as a support to the process of institutional design.




Danko Roozemond

As managing traffic becomes increasingly complex and interdisciplinary, more information is needed by more different people at the same time. Therefore a need arises for a new, more sophisticated, information system accessible to all. This research deals with the preliminary exploration for a dynamic traffic management system that can handle some of the social problems caused by traffic. Chaos theory and self organization form the basis for our approach to deal with the behavior and interactions of the different individuals and in(ter)dependent traffic participants. Furthermore the traffic management part is addressed; the part that can deal with the not always precise information gathered in the system. Finally the complex process of validating such expert-based simulation system is addressed.



Design of a Dialectic Policy Analysis Method for the Management of Production and Consumption Chains.

Remke M. Bras-Klapwijk

Human interpretations and judgments are extremely valuable in both everyday life and in politics. However, positivistic thinking values only objective knowledge in which human aspects are absent. A growing number of scientists now realize that we need new notions of knowledge that do value human judgments. This paper elaborates on these new notions and on their implications for the theory and the methodologies of policy analysis. The focus is on the implication for methodologies of chain management of an appreciation of human interpretations. Policy analysis methods -such as a life cycle method- do not take subjective perspectives explicitly into account, nor do they support the exchange of perspectives. For this reason, stakeholders are not enabled to learn about the goals, values and interests of other stakeholders or about political feasible measures. This is especially important in situations with different stakeholders. Therefore a dialectic, participative method for chain management has been developed in which different perspectives are illuminated, exchanged, confronted, and integrated. The new method employs participative, conceptual modeling techniques.




Joseph Wolfe

While social critics deplore the excesses and evil deeds perpetrated by business organizations in the name of profit, a more balanced view can often lead to a better understanding of the issues involved, the compromises which must be made, and methods by which practical solutions can be found and implemented. The use of simulation techniques may lead to this balanced view if the simulations used embrace and model the underlying issues facing what Parsons and Katz and Kahn have termed adaptive or production organizations. These organizations are uniquely qualified to solve certain societal problems although in reality they also create other problems due to the pursuit of their own needs for survival.

According to a survey by Faria North America's business students are exposed to at least one computer-based simulation during their undergraduate or graduate-level business educations. These simulations are almost always found in either an introductory marketing course or an end-program course in strategic management. This paper reviewed the set of commercial simulations available to those teaching these courses to determine the degree those simulations exposed players, either explicitly or implicitly, to the problems solved and caused by business organizations.

It was generally found that business's positive societal contributions were never explicitly highlighted although business's good deeds where implicitly featured through the introduction of new products, greater market penetration, lower product prices, and corporate dividends. Regarding societal problems created by business organizations, most exposure was again implicit through such decision results as fired workers, returns of defective products, and wasted funds caused by the oversaturation of markets and extravagant advertising campaigns. A few simulations, probably due to an emphasis on business ethics and corporate citizenship by North America's major business school accrediting body, feature explicit demonstrations of these topic areas although the direct effect of these demonstrations on the firm's operating results is problematical. Given this lack of explicit emphasis on the benefits and liabilities of business organizations as social entities, and a questioning of whether an implicit emphasis is ever felt by participants in a simulation given their attention to directly observable phenomena, the paper concludes with a set of recommendations as to how a typical game's deficiencies can be corrected.



Computer Support for ORGANIZATIONAL Diagnosis; the Design of a Diagnosis Instrument for the Analysis of Teacher Absenteeism ’

Eugene.G.M. Westerhof, J.C. Frowein, Theo.J.B.M Postma.

Setting out policy to prevent or reduce employee absenteeism is considered to be an important aspect of managerial activity, due to the substantial costs associated with absence behavior on an individual, organizational, and (not in the least) societal level. In the process of this policy formulation, knowledge of absence-inducing factors is needed. Our current research is aimed at supporting this knowledge enhancement process, which we call organizational diagnosis.

Support for organizational diagnosis can be provided by means of a (problem) specific diagnosis tool or instrument. In our research, we concentrate on the possibilities information technology offers to design and implement such a diagnosis instrument for a specific type of problem, teacher absenteeism.

This paper discusses the concept of organizational diagnosis, the content and design of diagnosis instruments, and the particular design-approach, based on knowledge acquisition methodology, used in our research.




E A. Stuhler - M. Vezjak

Building on previous research the article refers to common difficulties encountered by people handling complex problems and then deals with various complementary concepts usable in order to analyze the need for managers, administrators and politicians to regulate their actions. The findings could contribute to the available basis for a theory on complex problem solving.




Wim L. M. van Luijn

Consensus Engineering (CE)® is a method of structuring and facilitating collective decision-making in organizations. The CE method uses the knowledge and skills of those involved rather than that of external experts. It incorporates a method of ranking objects. This is done by comparison of object pairs based on agreed criteria. This is followed by the presentation and statistical analysis of the data. Together with a structured discussion of the results (Knowledge Engineering). The discussion aims at achieving further consensus where applicable.

The CE method may be applied to complex inter-disciplinary social problems that involve (a degree of) collective decision-making (for a definition of 'complex inter-disciplinary social problems', see De Tombe 1994). The method was originally developed for job evaluation by grouping functions (positions) in organizations. It has been applied, for example, to the ranking of budget items in a municipal council, the selection of a software package in industry, and the ranking of project priorities in a government ministry.

The Consensus Engineering method, a tool for structuring and facilitating collective decision-making, is described. A case history is reported and results are presented. The value of the tool in the area of Group Decision Support Systems is discussed.



Transfer, Simulation and the Limits to Education

Jan Gerrit Schuurman

In the faculty of Instructional Science and Technology (IST) of the University of Twente and in the Center for Innovation and Cooperative Technology (CICT) of the University of Amsterdam methods for control are studied that use as their medium for delivery the process of interaction itself (and take positive transfer and increase of quality as control-invariants). A computer simulation environment (called TILTPLAY) is built. Human participants (players) steer individual agents visible on their screens. Each player has its own screen and computer in a network. Players are instructed to avoid dropping off a simulated platform while having to move simultaneously. In addition they are encouraged to collect packages visible on the platform, again doing so without loosing balance. The main dependent variable is the form and amount of information that players can provide to each other, and are willing to use to control their interaction. The main independent variables are ‘the number of interacting players’, ‘urgency to act’ and mode of interaction. In the future positive transfer for coordinated and cooperative behavior should become a prime dependent variable.




M. Vezjak, E.A. Stuhler, A. Kositer

When the natural eco-system is in equilibrium there exist several kinds of species in more or less constant numbers. Natural resources are recycled and the energy use is rather low. With the arrival of tourists the natural eco-system becomes gradually overloaded. Therefore a transition takes place through which the eco-system becomes a "human eco-system". Numerous visitors caused a high level of pollution and energy consumption combined with a very low level of recycling. The following are the parameters for this simulated model: number of tourists, quality (state) of an environment, specific regional level of tourists, who are leaving, effect of advertising, specific level of environmental (over)load, specific environmental recovery rate and carrying capacity of the environment. The model is simulated with the language SIMPAS. The development of the described system is shown by trajectories in two or three dimensions. Since the United Nations Conference on the Environment and Development which was held in Rio de Janeiro, the leading concept for the environmental policy for the future has been sustainable development, which is closely linked to the concepts of carrying capacity and critical load concept. Complex problem solving in our case means the determination of the interlinking of the number of tourists and the state of an environment to the carrying capacity and the specific environmental recovery rate of one specific tourist area. For everyday practice in tourism it will be useful to find a more general method to get an assessment of the global carrying capacity of the environment in form of the environmental space. This value could be used for the tourism purposes on local, regional or country level. In the paper a concept of an overall (real-time) ecological monitoring system as a base for the development of quality (state) of an environment is suggested. Such a system is based on instantaneous measurements, processing, displaying and storing some important parameters of real state of an environment. The findings of the paper are clear: the new form of tourism should be sustainable (soft) tourism, which would based on new ecological ethics of tourists.



A model simulation to evaluate vaccination strategies against influenza

A. Reinders., J.C. Jager.

RIVM develops a simulation model for the evaluation of vaccination strategies against influenza. Influenza is a seasonal disease, causing an epidemic almost every year with varying effects upon the population. In the Netherlands each year an average of 1100 persons die of influenza. In the United States the total cost of influenza is estimated at 3 to 5 billion dollars each year, of which about 1 billion dollars are estimated to be direct medical costs of excess hospitalization. Influenza leads to excess mortality in heart disease, lung disease and some other diseases and is closely associated with age. Almost 95% of the persons who die of influenza or related disease are 60 or older. Influenza can be partly prevented by vaccination. People who are vaccinated have a lower probability of becoming infected. Because the efficacy of the influenza vaccine is not 100% and because immunization usually lasts for one or two years there is a debate whether influenza vaccination is effective or not.

In the Netherlands advice on influenza vaccination is given by the Dutch council of health in a yearly report to the government about the part of the population which should be vaccinated against influenza. Currently there are two points of view among experts: the majority point of view states that only risk groups (persons with heart disease, lung disease, etc.) should be vaccinated, while a significant minority wants everybody above 65 to be vaccinated. Their main objection against vaccinating risk groups is that it is often unknown whether someone belongs to a risk group or not.

To support decision-makers in choosing between these points of view, an interactive simulation model is being developed which simulates the effects of influenza and vaccination for the Dutch population. It is a so-called compartmental model in which several compartments are distinguished in the population (susceptible, diseased and immune) and the flow between these compartments is described by differential equations. The current version is able to compute the effects of vaccinating the different age classes. One of the preliminary simulation results shows that when 90% of the persons below 60 are vaccinated the mortality among people of 60 and older will be lower than when 90% of the persons above 60 are vaccinated.

In the near future excess mortality will be added to the model, so that risk groups can be vaccinated. The influenza model is one of the six models of the so-called integrated public health model, which is currently being developed at the RIVM. Excess mortality will be implemented by treating influenza as a risk factor for the relevant diseases (heart disease and diabetes). Another aspect to be added to the model will be health care. The model will then be able to evaluate the costs in public health and economy of both vaccination strategies.



Gaming/Simulation as a Tool for Analyzing Policy Problems:

The Diagnost Game and the Regional Network of Elderly Care

Cisca Joldersma ,Gerton Heyne & Jac L. Geurts

In interaction processes, interdependent stakeholders produce and reproduce policy problems. Gaming/simulation can stimulate actors to reflect in action on their problem-structuring activities. This chapter describes the design of the Diagnost game and reviews two runs of the game. It is concluded that a gaming/simulation can be used as a tool for analyzing policy problems, but, if stakeholders stick to their own action frames, the gaming/simulation should hold intervening elements.




A Master Equation Model to ANALYZE the Dynamics

at Austrian National Elections from 1970 to 1990

Kathrin Gruetzmann and Christoph Hofinger

The master equation formalism allows us to describe long term dynamics in societies by modeling transition probabilities. However, empirical studies are rare. This is a consequence of the difficulty in finding time series and transition matrices of sociological data that is empirically measured or estimated. With some effort we can obtain this data for the Austrian national elections from 1970 to 1990. In the proposed model we see electoral dynamics as a result of symmetrical distances between the political parties and asymmetrical attraction parameters. The model is able to explain a considerable share of the voting behavior of the given time period.




See for other parts of the book Analyzing Complex Societal Problems: 




See for more publications of Dorien J. DeTombe

See for lectures of Dorien J. DeTombe