Application of Naturalistic Decision Making to Emergency Evacuation Simulations

Abstract

Evacuation simulations are among essential analysis to assess the level of life safety in buildings. Currently, a number of evacuation modeling tools are available, each with unique characteristics. But, complexity of human behaviors made most of these tools ignore behavioral data in their models; and among those models which account for human behaviors, rule based modeling and rational decision making approaches are used for decision process. However, to achieve more realistic evacuation calculations, human-like decision making should be integrated to evacuation simulations. In this research we present a novel approach to build multi-agent evacuation simulations that exhibit more realistic simulation behaviors. We propose a model that integrates individual differences, such as personality, age, and gender, with naturalistic decision making, where human decision making in natural environments is investigated. The resulted decision process is applied to BDI agent architecture within a building evacuation system.KeywordsPersonality TraitEvacuation TimingPreference WeightingEmergency EvacuationNaturalistic DecisionThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.