As heavy rainfall increasingly disrupts services of urban rail transit systems (URTSs), enhancing their resilience to flood risks is crucial to sustaining reliable public transport, particularly amid growing climate challenges. To investigate the effectiveness of potential interventions for mitigating post-flood impacts on URTS operations, this research introduces a novel application of genetic algorithms to optimise recovery resource scheduling for URTSs following large-scale flood-induced disruptions. The objective is to reduce economic impacts related to revenue loss and operational impacts concerning disruptions to passenger travel. By systematically integrating network topology, operational performance, flood disruption scenarios, and recovery profiles, the methodology is demonstrated through the London URTS under 30-year, 100-year, and 1,000-year flood risk scenarios. Compared to a topological attribute-determined benchmark, the optimised resource scheduling solutions have a tangible effect in reducing post-flood impacts. In the London case study, revenue loss can be reduced by 10.9%, 10.7%, and 6.7% across the respective flood scenarios, corresponding to savings of approximately £337K, £708K, and £760K, along with decreased unmet travel demand of 197K, 404K and 470K. These results demonstrate the significance of strategic resource scheduling in ensuring effective recovery from large-scale flood disruptions, offering valuable insights for disaster risk management, especially for extreme weather scenarios.