Abstract: To address the challenge of integrating regional network analysis with individual station hydraulic simulation in the inundation risk assessment of urban rail transit systems under extreme rainfall,this study aims to construct a comprehensive risk assessment framework.Based on the LEC (Likelihood,Exposure,Consequence) risk theory,a cross-scale assessment model is proposed,integrating complex network analysis,hydrological-hydraulic simulation,and expert evaluation methods.Taking the central urban area of Jinan as a case study,multi-source data—including 3D topographic data,rail transit network topology,BIM models,and expert surveys—are utilized to quantify the inundation likelihood,exposure,and consequences of stations.Stations classified as high risk within the assessment framework are selected for refined 3D hydraulic simulations.Under the specified rainfall scenario,without accounting for station-specific design flood protection levels or the effects of temporary emergency flood protection measures,stations including Shengchanlu,Bajianpu,and Jiluolu are classified as high risk based on passenger flows,network positions,and terrain conditions.For the typical station model under the specified inflow conditions,columns reduce the flow rate at the platform-tunnel interface by approximately 9%,while glass barriers without bottom gaps reduce flow toward the platform,indicating potential to delay floodwater spread.The “regional identification-individual simulation” assessment framework established in this study possesses strong operability and provides a scientific basis for flood prevention planning and emergency management in urban rail transit systems.