A Simulation-based Research on Emergency Pedestrian Evacuation for Urban Residential Districts

Evacuation may be carried out before, during or after natural disasters, terrorist attacks, etc. And it is a common strategy for handling such emergency situations. This paper presents a simulation-based research on emergency pedestrian evacuation for urban residential districts, which takes into account the behavioral input. This study assumes that residents in threatened living districts are evacuated to safe places on foot. A temporal distribution curve of resident's evacuation decision time is obtained based on studying the decision-making behavior of residents before evacuation. The degree of rationality and familiarity are introduced to predict spatial distribution of residents. Evacuation simulations are conducted with different values of the degree of rationality and familiarity. The network clearance time is obtained to measure the effect of people's different behavior on evacuation efficiency. The simulation incorporating behavior input is much more realistic than purely engineering-type models.