Abstract: Identifying high-risk pipelines and reducing their failure probability is an important means to alleviate urban waterlogging.Therefore, this paper proposed a high-risk pipeline identification method based on the correlation between drainage capacity and failure probability.Firstly, the critical criterion for pipeline failure was that the fullness was equal to 1,and the failure probability of pipeline under the design rainfall scenario was calculated based on Monte Carlo sampling.Secondly, SWMM was used to evaluate the drainage capacity of the pipeline.Then, the K-means clustering algorithm and the piecewise regression analysis method were used to construct the correlation between the drainage capacity and the failure probability, and the discriminant threshold of the high-risk pipeline was deduced by the correlation.Taking a drainage area in Fengtai District of Beijing City as an example, the failure probability of 3.64% was deduced as the identification threshold of high-risk pipelines under the scenario of three-year rainfall.Combined with the longitudinal profile of high-risk pipelines, the causes of waterlogging were analyzed, and the rehabilitation plan was formulated.The simulation results showed that the proportion of high-risk pipelines in the total number of rainwater pipelines was reduced from 91.39% to 2.32% after rehabilitation.The proportion of the total length of the high-risk pipeline to the total length of the rainwater pipe network decreased from 88.74% to 1.11%.