As an important lifeline system in the city, urban rail transit system has a high degree of uncertainty in the process of earthquake damage and post-earthquake repair. In order to evaluate its seismic resilience more reasonably, considered a variety of uncertain factors and based on Monte Carlo simulation, a quantitative evaluation framework of seismic resilience of urban rail transit system was concluded. The 'ladder' recovery path was selected, and the network service efficiency index was used to represent the performance level of the system. The spatial difference and uncertainty of ground motion were considered based on the ground motion prediction equation. The Monte Carlo method was used to simulate the post-earthquake damage state and recovery process of the system, and the uncertainty of system damage and recovery was considered. The uncertainty of repair time was considered by the method based on 'distributed sampling'. Zhengzhou rail transit network was taken as a typical case, the seismic resilience under the 8-magnitude earthquake was evaluated. The results show that the seismic resilience index obeys the normal distribution with a mean of 0.34 and a standard deviation of 0.03. The total repair time after the earthquake obeys the normal distribution with a mean of 276.86 days and a standard deviation of 21.05 days. This study can provide a reference for the resilience assessment of urban rail transit system.