Enhancing Rail Transit Safety Under Extreme Rainfall Risks Based on Stochastic Programming Model

ABSTRACT The increasing frequency and severity of extreme rainfall events have significantly heightened waterlogging risks for rail transit systems. In this paper, we investigate the optimization of the emergency response plan for rail transit networks under extreme rainfall risks to enhance the railway system's safety and resilience. A two‐stage stochastic programming model considering uncertainties in inundation depths of railway stations is proposed, in which the weighted sum of the total expected unsatisfied demands and passenger flow time is minimized. The protection level and the closure time of railway stations are determined in the first stage, while the passenger's flows are modeled in the second stage under the given first‐stage plan. To make the two‐stage model solvable, we derive a computationally tractable mixed‐integer linear programming (MILP) formulation. Since the proposed mathematical programming model and solution method are based on perfect prior information, to further evaluate the practical performance of solutions derived from the MILP model, we develop a time‐driven simulation model. The proposed approach is validated using a real‐world case study based on the Hangzhou Metro system, demonstrating its effectiveness in optimizing emergency plans and maintaining service continuity.