Due to the fact that it was difficult to obtain the relevant data of high-speed railway delay, the research on the factors of high-speed railway delay mostly adopted the qualitative analysis method, and the factors considered were limited, so it is less effective in improving the punctuality rate of high-speed railway. In this study, we paid much attention to explore the contributing factors on the delay of high-speed railway and quantify the effects of these factors based on Chinese high-speed railway late actual data. Firstly, the Bayesian network structure model of the delay of high-speed railway was established by means of expert experience and Dempster-Shafer evidence theory. Secondly, the established Bayesian network structure model was modified by test for conditional independence. Finally, the posterior probability of each contributing factors in the Bayesian network was calculated based on 390 Chinese high-speed railway delay data. On the basis of the above analysis, the relationship between different factors can be objectively judged, and the key contributing factors can be identified by quantitatively measuring the influential degree of each factor. The results showed that the device failure was the most significant factor leading to the delay of the train. Among the device, Automatic Train Protection, platform door, and catenary were the three factors that have the greatest impact on train delay. These findings can help managers improve and optimize the work priorities and processes of high-speed railway operation, so as to reduce the delay rate of high-speed railway and improve the service level. Keywords: High-speed railway, Delay, Bayesian network, Dempster-Shafer evidence theory

CICTP