Study on Subway Station Evacuation Performance by the Improved Cellular Automata Evacuation Model

Abstract

By introducing toxicity coefficient to the static force field and pheromone model to the dynamic force field, an improved cellular automata evacuation model (ICAEM) is developed. Evacuation experiments in an 11 m × 8 m classroom are carried out, the time from their initial position to the single exit or double exits of the classroom and the stagnation time near the exits are recorded and compared with the simulation results to validate the ICAEM model, to check the influencing factors of the exit position KS, the repulsive coefficient KR, and the pheromone damping coefficient ρ in the model. The results show that when the pheromone damping coefficient ρ = 0.5, the self-lining phenomenon is significant. If KS = 1, KR = 0.12, the simulation result is very consistent with the actual evacuation time in the experimental tests. Finally, the ICAEM model is applied in a 118 m × 36 m twenty-first-century subway station platform where there are a total of 2500 persons from two trains waiting for evacuation. The simulation error is reasonable comparing with the calculation results by engineering design code of PRC and the Togawa’s formula. It is found that the original layout of EXIT 2 obstructed the evacuation flow of the station platform at a certain degree, resulting in bidirectional counter flow and stagnation near the exit. After improving the layout of the EXIT 2 on the platform based on the simulation results by ICAEM, the RSET of the whole platform is shortened and the stagnation phenomenon is attenuated. The ICAEM model is applicable and meaningful to the crowd evacuation and performance-based safety design in high densely populated public places.