The cellular automata evacuation model based on Er/M/1 distribution

Abstract

In the evacuation process of indoor space, it is often easy to form congestion and queuing phenomena at the exit due to the limitation of space capacity. For this reason, in this paper, the queuing condition of pedestrians at the exit is described based on Er/M/1 distribution and a crowd evacuation model is established using cellular automata. The model firstly gives a calculation method for movement benefits of pedestrians at the next moment according to field domain and queuing time, and deeply deduces the queuing time using the queuing phase. In addition, a simulation platform is established based on the evacuation model to deeply analyze the relationships among the parameters such as queuing time, actual phase, average queuing length and average velocity. The results show that when the exit flow rate is about to reach a saturation condition, the average queuing length presents a trend of steady increase followed by sharp increase after further increase in the number of pedestrians; and meanwhile, the queuing time increases rapidly when the aggregation degree near the exit increases to 0.4–0.6.