A PEDESTRIAN EVACUATION MODEL BASED ON ANT COLONY ALGORITHM CONSIDERING DYNAMIC PANIC SPREAD

To explain the evacuation behavior characteristics of people in panic, an improved cellular automata pedestrian evacuation model is proposed by combining with the SIS (Susceptible-Infected-Susceptible) epidemic model and ant colony algorithm. First, to explain the spread of panic in dynamic pedestrian evacuation, an improved SIS infection model that considers individual movement is proposed. At the same time, a quantitative formula for panic mood is given. The infection critical value of the SIS model is analyzed by mean field theory. Secondly, the ant colony algorithm is introduced based on considering exit distance, obstacles, panic, and other factors. These factors are mapped with heuristic function and pheromone concentration in the ant colony algorithm. Finally, the evacuation model and parameters such as evacuation time, pedestrian density, and panic spread are analyzed and discussed through simulation experiments. The simulation experiment results show that the evacuation model proposed in this research can better reflect the actual situation. Simultaneously, the critical rate of effective propagation is closely related to the crowd density in the evacuation area, and individual movement has a significant impact on the panic transmission behavior.