Subway station facility layout optimization for efficient evacuation: a hybrid method of adaptable queueing network and heuristic retrieval

Safe and efficient evacuation is a key issue for underground transportation systems, especially subway stations with overcrowded passengers and narrow spaces. This work aims to provide a subway station facility layout design adjustment method for evacuation performance optimization. An adaptable queueing network is constructed to simulate the evacuation process. A heuristic search workflow based on the genetic annealing algorithm is designed for the optimal evacuation facility layout search. A verification framework for facility layout adjustment is constructed from both daily operation and emergency evacuation situations. A real subway station in Guangzhou is adopted as a case study to testify the applicability of the proposed method. The results suggest that (1) The evacuation facility layout could be effectively optimized with the proposed method. The optimal facility layout presents a 6.85% evacuation performance improvement compared with the current design in the peak hour scenario. (2) There is a broad space for facility layout optimization. About 90% of the randomly generated facility layouts have a better evacuation performance compared with the current design. (3) The robustness of the facility layout optimization method is validated in evacuation scenarios with a stable optimal facility location and daily operation scenarios with unvaried operation capacity. This hybrid method could become a quantitative evaluation tool for decision-makers and designers to identify the optimal facility layout and promote the efficiency of evacuation in buildings with complex structure.