Abstract: In order to effectively control the accident risk in hazardous chemicals enterprises and prevent hazardous chemicals fires and explosions involving hazardous chemicals, a Bayesian and CA coupling model was built based on the data on the "8·12" special fire and explosion accident in Tianjin Port. The accident scenarios were simulated and reconstructed. All aspects of the evolution of the accident were analyzed. The law of accident evolution was explored. Hidden dangers of potential accidents in the storage industry of hazardous chemicals were identified. Based on the ACO, an emergency evacuation model was built to simulate the evacuation of residents in the worst affected communities on the Matlab platform. The best evacuation route was identified. The response time point was calculated for evacuating the surrounding residents and related personnel to the safe area. The results show that according to the results of computer simulation, emergency evacuation routes can be planned, which can improve the efficiency of emergency evacuation could be improved and the accidents impact could be reduced by planning the emergency evacuation routes according to the results of the computer simulation.

Key words: hazardous chemicals, cellular automaton(CA), emergency evacuation, ant colony optimization(ACO)algorithm, Tianjin port