Urban rail transit operation risks intelligent identification algorithm based on dispatching fault log data mining

In this work, we proposed a new systematic metro operation risk identification method (MORIM) and risk grade classification method (RGLM) based on the daily dispatching fault log. We collected and analysed the operation risks during Metro operation, and the database SQL was designed for calculating the probability of risks. Then, we converted the equipment malfunction, train delay, large passenger flow etc. to time delay, so as to realize the quantitative calculation of the risks. We clustered risk sources by data mining, from which, we can get the risk cluster centre, and the emergency response scheme can be accurately made to match the clustered risks. Finally, the systematic method was validated by a case study. It was found that the method was accurate and the conclusion was reliable. This paper can provide theory and decision support for Metro operation safety management and it has good practical significance for larger cities to dispose the conditions of large passenger flow.