Risk assessment of passenger flow dispersion in urban rail transit transfer stations

Urban rail transit has become an important part of the urban public transportation system. The transfer station serves as the intersection of urban rail transit lines to entire urban rail transit lines as an interconnected network. Once the passenger flow at the transfer station is not well dispersed, it will adversely affect the several lines and even the entire line network. Therefore, it is of great significance to scientifically evaluate the passenger flow dispersion risk of the transfer station. In this paper, a datadriven method is proposed to analyse the characteristics of passenger flow (e.g., the moving speed of different groups). A set of passenger flow diversion evaluation indicators for rail transit transfer stations is established, including the passenger flow density, emergency evacuation time, station passenger facility saturation, transfer time, and average transfer distance. And the software of AnyLogic and STEPS are applied to simulate the evacuation process. By comparing the observation data with the simulation results, the simulation model is verified to show that the simulation results have a good agreement with the actual situation. A case study with focusing on transfer Station X is adapted to illustrate the feasibility of the proposed approach.