Mechanism and strategies of post-earthquake evacuation based on cellular automata model

Abstract

The evacuation procedure during an earthquake can likely lead to a stampede among a crowd of people involved in it and consequently casualties. The understanding and simulation of this procedure are essential to study how unexpected disasters affect casualties, behaviors and safety levels in an earthquake. This paper proposes an improved cellular automata model considering people's subjective perceptions (e.g. familiarity, behavioral preference) and implementation of safety protocol (e.g. protocol robustness) during an earthquake. This study simulates an evacuation procedure done in a room and a whole building; the effects of the model parameters, evacuation priority, and collapsed staircases are studied. The minimum evacuation time is associated with higher level of evacuees’ familiarity with the surroundings where the earthquake occurs, undifferentiated habitual preference, and a suitable range of protocol robustness. A multi-floor case study based on the proposed model shows that the total evacuation time in the building after an earthquake occurs can be significantly reduced with more emergency staircases. Adjusting the evacuation order of pedestrians on different floors can increase the efficiency of evacuation. The effect of collapsed staircases during or after earthquake occurrence is simulated to examine the pedestrians in a collision. Finally, strategies for improving the efficiency of evacuation during earthquakes are proposed.