Analysis of risk factors affecting delay of high-speed railway in China based on Bayesian network modeling

Abstract

This study aims to investigate the effects of risk factors affecting high-speed railway delay using Bayesian networks. At first, we determined risk factors based on real time data about high-speed railway delay in China. Then, the Bayesian network structure model of the delay of high-speed railway was established according to expert experience and Dempster-Shafer evidence theory. The established Bayesian network structure model was modified by test for conditional independence in the next step. Finally, the posterior probability of each contributing factors in the Bayesian network was calculated based on the real time data. Compared to other statistical analysis methods, Bayesian network took interaction of risk factors into consideration by analyzing the correlation and influence degree of influencing factors under incomplete information. It has been found that the device failure is the most significant factor leading to the delay of the train, among which, failure of automatic train protection, platform door and catenary are the three specific factors that affect the delay most. The findings in this study provide useful and valuable information for high-speed railway operation managers to take effective countermeasures to reduce the delay rate of high-speed railway.