Incorporating Human and Social Behavior in Computational Egress Analysis

Recent studies of past disastrous accidents have shown that human behavior plays a crucial role affecting evacuation time and patterns. For example, in the 9/11 World Trade Center studies, people demonstrated helping behaviors, even towards strangers, while evacuating [1]. Decision to evacuate depends on many human factors, such as presence of authorities and individual’s experience and familiarity with the environment. Such factors have significant influences, for example, resulting in delayed reaction to initial evacuation. Crowd density and group relationships can greatly affect evacuation patterns [2,3,4,5,6]. There is a clear need to better comprehend the different human and social behaviors for emergency evacuation analysis, simulation, design and planning. Based on analyses of recent fire and emergency events, social scientists have developed theories that attempt to capture human behavior based on individual characteristics, group relationships and crowd behaviors. Our research proposes to incorporate some of these theories in a computational framework that will potentially lead to more accurate virtual simulation of emergency evacuation and to better support “whatif” design scenarios for emergency planning and management.