Rules of incidental operation risk propagation in metro networks under fully automatic operations mode

Abstract

The frequent interruptions of network operation due to any incident suggest the necessity to study the rules of operational risk propagation in metro networks, especially under fully automatic operations mode. In this study, risk indicator computation models were developed by analyzing risk propagation processes within transfer stations and metro networks. Moreover, indicator variance rules for a transfer station and different structural networks were discussed and verified through simulation. After reviewing the simulation results, it was concluded that under the impacts of both sudden incident and peak passenger flow, the more the passengers coming from platform inlets, the longer the non-incidental line platform total train operation delay and the higher the crowding degree. However, train headway has little influence on non-incidental line platform risk development. With respect to incident risk propagation in a metro network, the propagation speed varies with network structure, wherein an annular-radial network is the fastest, a radial is moderately fast, and a grid-type network is the slowest. The conclusions are supposed to be supports for metro operation safety planning and network design.

Railway Safety Analysis: Trends and Challenges

“…(Ramirez, 2022) For this reason, artificial intelligence is a key tool for implementing safety automation instruments: autonomously operated trains, with the obstacle detection system, can increase the existing level of safety with the elimination of human errors and the recognition of obstacles in difficult conditions; and it can significantly increase transportation efficiency and decrease railway headway. (Chen, 2021). Another recurrent problem is the lack of communication between trains.…”