Responding of metro stations to upcoming floods: To close or to protect?

Floods have become a pressing challenge for urban metro systems. Facing an upcoming flood, closing all high-risk metro stations is a straightforward solution, but it can negatively affect people's travel. Conversely, protecting all high-risk stations to ensure their operation for smooth public transportation comes at a significant cost of flood control resources. Hence, given limited resources, it is necessary to devise an optimal closure-protection scheme to reduce the impacts caused by station closure while ensuring safety. For this issue, current practice mainly relies on the subjective experience of metro managers, lacking scientific and reasonable decision-making. To address this problem, this study proposes a novel optimization framework, implementing the entire process from metro network modeling, high-risk station identification to network performance evaluation and closure-protection optimization. A case study is conducted on the metro network in Shanghai, China, to verify the effectiveness of the proposed framework. The results demonstrate that compared with the best baseline strategy, the optimization framework can improve the metro network performance by 3.6 percent points with 100 units of resources under minor floods, and 6.2 percent points with 300 units of resources under moderate floods. Under major floods, however, closing all high-risk stations is more cost-effective.