Protecting Rail Transit Systems from Flood Risks: A Two-Stage Stochastic Programming Approach

Climate change leads to more frequent and intense climatic extremes in recent years, such as rainstorm and sea level rise. The increased flood risk will do damage to rail transit systems, threatening system reliability and safety. In this paper, we propose a new method to develop flood prevention measures for rail transit systems using stochastic programming approach. A two-stage stochastic programming model considering uncertainties in inundation depth is proposed. The first stage identifies which station exit to protect and the corresponding level of protection. Given the first-stage prevention measures, the second stage optimizes closures of stations and model commuter flows after the realization of uncertain variables. Particularly, we model the statuses of exits, platforms and trains under extreme flooding events, to ensure the safety of commuters. The proposed model is demonstrated on the core part of Hangzhou Metro network.