Informal Approaches to Developing Simulation Models

Abstract

This chapter describes an approach commonly taken by most people in the social sciences when developing simulation models instead of following a formal approach of specification, design and implementation. What often seems to happen in practice is that modellers start off in a phase of exploratory modelling, where they don’t have a precise conception of the model they want but a series of ideas and/or evidence they want to capture. They then may develop the model in different directions, backtracking and changing their ideas as they go. This phase continues until they think they may have a model or results that are worth telling others about. This then is (or at least should be) followed by a consolidation phase where the model is more rigorously tested and checked so that reliable and clear results can be reported. In a sense what happens in this later phase is that the model is made so that it is as if a more formal and planned approach had been taken.

There is a danger of this approach: that the modeller will be tempted by apparently significant results to rush to publication before sufficient consolidation has occurred. There may be times when the exploratory phase may result in useful and influential personal knowledge, but such knowledge is not reliable enough to be up to the more exacting standards expected of publicly presented results. Thus, it is only in combination with a careful consolidation of models that this informal approach to building simulations should be undertaken.

Further Reading

Outside the social sciences, simulation has been an established methodology for decades. Thus, there is a host of literature about model building in general. The biggest simulation conference, the annual Winter Simulation Conference, always includes introductory tutorials, some of which may be of interest to social scientists. Good examples are Law (2008) and Shannon (1998).

For a comprehensive review of the currently existing general agent-based simulation toolkits, see Nikolai and Madey (2009); other reviews focus on a smaller selection of toolkits (e.g. Railsback et al. 2006; Tobias and Hofmann 2004; Gilbert and Bankes 2002).

The chapters in this volume on checking your simulation model (Chap. 7, Galán et al. 2017), documenting your model (Chap. 15, Grimm et al. 2017) and model validation (Chap. 9, David et al. 2017) should be of particular interest for anyone intending to follow the exploration and consolidation approach to model development. However, if you would rather attempt a more formal approach to building an agent-based simulation model, Chap. 6 (Siebers and Klügl 2017) discusses one such approach in detail. You could also consult textbooks on methodologies for the design of multi-agent systems, such as Luck et al. (2004) and Bergenti et al. (2004) or Henderson-Sellers and Giorgini (2005). After all, any agent-based simulation model can be seen as a special version of a multi-agent system.