China Safety Science Journal ›› 2019, Vol. 29 ›› Issue (3): 180-186.doi: 10.16265/j.cnki.issn1003-3033.2019.03.030

• Public Safety • Previous Articles     Next Articles

Emergency evacuation path optimization model under multi-export conditions

ZHENG Xiazhong1,2, CAI Lili1, ZHANG Ming1, JIN Lianghai1,2, CHEN Yan1   

  1. 1 College of Hydraulic & Environmental Engineering, Three Gorges University, Yichang Hubei 443002, China;
    2 Center of Standardization Evaluation for Production Safety, Three Gorges University, Yichang Hubei 443002, China
  • Received:2018-11-10 Revised:2019-01-15 Published:2020-11-26

Abstract: To reduce the probability of congestion, considering the characteristics of evacuation route selection under multi-export conditions, an emergency evacuation route optimization model is established. Firstly, based on the evacuation channel arrangement orientation, the evacuation network is established, and the feasible evacuation path set is determined. Considering the dynamic relationship between crowd flow and queuing time, the population lag time is calculated, the path equivalent length is characterized, and the iterative feasible path set is represented. Then aiming at the shortest total evacuation time, the Dijkstra algorithm objective function is modified, the time search optimal path is globally, and the evacuation scheme is allocated. Finally, taking the multi-egress floor case as an example, the total time of the shortest and optimal evacuation path evacuation is compared to verify the validity of the model. The results show that the total time of queuing and the total evacuation time of the optimized evacuation scheme are smaller than the traditional evacuation model. The utilization of evacuation exit is higher, the distribution of people flow is more even, and the probability of overlapping of evacuation paths is smaller, which effectively improves the evacuation efficiency.

Key words: emergency evacuation, path optimization, dynamic network flow, point queuing, multiple exits

CLC Number: