Flood risk assessment and mitigation for metro stations: An evidential-reasoning-based optimality approach considering uncertainty of subjective parameters

This study presents an optimality approach based on evidential reasoning for flood risk assessment and mitigation of metro stations. To address the uncertainty of subjective parameters involved in risk assessment, we consider the subjective parameters to vary within certain intervals, and then employ evolutionary optimization to solve the maximum risk expectation, which is adopted as the index to measure the flood risk in the worst case under the uncertainty of subjective parameters. Furthermore, we propose a sensitivity indicator termed combined gradient to identify critical influential factors and correspondingly provide risk mitigation measures for the high-risk metro stations. A case study is conducted on Shanghai metro stations to verify the applicability and validity of the proposed approach. The results mainly indicate that: (1) The metro stations in eastern Shanghai have higher flood risks than those in central and western areas. (2) Improving the drainage capacity is of uppermost priority for flood risk mitigation. The novelty of the evidential-reasoning-based optimality approach lies in its capabilities of (1) considering the uncertainty of all kinds of subjective parameters in a unified framework; (2) considering both the short-term and long-term benefits of risk mitigation measures so that the most effective measures can be suggested.