Agent-based simulation of fire emergency evacuation with fire and human interaction model

Abstract

Fire emergency evacuation in an underground mass transit station is a great concern especially in developing countries. The interaction between fire and human is very important in the analysis of fire emergency evacuation. In this study, an agent-based fire and human interaction model is used. The grid resolution is analyzed to determine an appropriate grid size that will optimize the solution accuracy and time. Different fire locations, heat release rates, occupant loadings, ventilation conditions and material properties are considered under fire condition in an underground subway station. It shows that the heat release rate has a weak influence on fire emergency evacuation, but the fire location, occupant loading, ventilation condition and material property have a great influence on fire emergency evacuation. Furthermore, the five parameters have a coupled function on fire emergency evacuation.