Modeling of Spatial Transfer of Affected Passengers to and from Stations Under Delayed Urban Rail Trains

The rapid development of urban rail transportation requires more and more emergency events such as train delays. This paper proposes a spatial transfer model for affected passengers entering and leaving stations under train delays in urban rail, aiming to provide a basis for emergency response under train delay type of emergencies. The study analyzes the influencing factors of passenger travel under train delay conditions, constructs physical network and spatio-temporal expansion network under train delay conditions, and establishes effective physical path set and spatio-temporal path set. Based on the spatio-temporal path set, the generalized cost of passenger travel under train delay is calculated, and the spatial transfer probability of affected passengers in and out of stations is obtained by Logit model, and the spatial transfer model of affected passengers in and out of stations is established by using particle swarm algorithm for parameter optimization. The prediction model was validated using the data of a train delay event on Beijing Metro Line 10.