Abstract: In this paper,the optimal fault distribution model and preventive maintenance plan of EMU bogie components were studied. Firstly, the fault data of a certain type of rolling stock were preprocessed to obtain the fault samples, the specific parameters of exponential, normal and Weibull distribution models were calculated according to the parameter estimation method, and the optimal fault distribution model of the component was determined by using the K-S test method. Secondly, taking the axle box device as an example, the reliability assessment of the optimal fault distribution model was conducted to obtain the best maintenance cycle for the secondary repair of this component, and compared it with the actual maintenance cycle on site, according to the research results and the existing maintenance plan of the EMU bogie, the adjustment and optimization suggestions for the maintenance cycle of the bogie components were put forward. Finally, taking the component reliability threshold as the decision variable and the minimum preventive maintenance cost as the optimization objective, the preventive maintenance plan of each component in a life cycle was obtained through an example.