Why model evacuee decision-making ?

The purpose of this article is to outline the theoretical and practical benefits of representing the evacuee decision-making process within an agent-based simulation tool. This rationale is important as the development of a comprehensive representation of evacuee decision-making will be expensive, requiring a great deal of time, expertise and effort. However, the theoretical and practical benefits are such that this effort is considered warranted. In previous work, Gwynne et al. demonstrated advances in the representation of evacuee performance and the potential for representing evacuee decision-making. Here, we show that realising this potential is critical to progressing future analysis and, in turn, the field. The paper concludes the following: Current understanding of evacuee performance suggests a decision-making process often in response to a complex, ambiguous and dynamic environment. This process connects the conditions experienced by an agent and the actions taken. Representing this is important if we are interested in what evacuees do and when they do it. Agent-based models have the potential to represent evacuees, their decision-making process, their sub-sequent actions and the resultant interactions between agents and entities in their environment. Generative Social Science (GSS) employs ABM (agent-based models) to produce subject matter insights using retrodiction and prediction. These approaches will help us enhance our subject matter understanding and the computational tools available to quantify evacuee performance, in turn aiding theoretical and practical efforts. The specific benefits of this approach to our understanding and quantification of evacuee performance are described in this paper, which include the expansion of the explanatory value of the tools available and the refinement of theoretical explanations of evacuee behaviour.