Agent-Based Model for Pedestrians’ Evacuation after a Blast Integrated with a Human Behavior Model

Abstract

Agent-based modeling (ABM) is a powerful tool for model complex and heterogeneous systems as pedestrian evacuation after extreme events. Several studies on Emergency Management exist in literature, but the majority of them do not consider how human behavior impacts the evacuation plan. Indeed, these can change drastically a critical scenario because are driven by emotions, knowledge and perception which are unpredictable. The aim of this study is to develop an ABM of an evacuation scenario due to a blast in a public area integrated with a mathematical dynamic human behavior model. Netlogo, which is a multi-agent programmable environment, was used to build the ABM. ABMs simulate two phases: the normal phase and the agent’s evacuation process after the blast. The evacuation time is the main parameter of response and it is used to evaluate the efficiency and safety of an infrastructure. Each agent has its own behavior according to a layered framework mathematical model. The first layer, thanks to an empirical statistical distribution, simulates the “agent’s state” function of the role performed by the agent and his age. Then the” individual module” describes the emotional aspects using the Decision Field Theory. This ABM becomes an important tool that enables designers or policy makers to test, estimating and improving the response of infrastructures in hazardous scenario.