Evacuation Entropy Particle Swarms Optimization (E-PSO) Evacuation Model Considering Panic Factors

Abstract

As the existing crowd evacuation model seldom considers the influence of crowd confusion degree on the evacuation results. This paper proposes evacuation entropy particle swarms optimization (E-PSO) crowd evacuation model. This model considers the impact of obstacles and the surrounding people on evacuees' evacuation behavior in the evacuation process. And the concept of evacuation entropy is introduced to quantitatively describe the degree of chaos in crowd movement and is used to correct the individual speed direction in the model. The model further considers the influence of individual panic psychology and the presence of leaders in evacuation. The simulation results show that the panic of evacuees will aggravate the degree of chaos on the event, but the addition of leaders will reduce the degree of confusion and improve the evacuation efficiency. At the same time, compared with the traditional PSO model, the modified PSO model with evacuation entropy can reduce the degree of confusion on the evacuation process, effectively avoid the occurrence of crowded stampede accident and improve the evacuation efficiency.