Spatiotemporal path inference model for urban rail transit passengers based on travel time data

Although the 'one ticket transfer' mode brings convenience to passengers, it also poses challenges to the passenger flow assignment. The current widely used multi-path probability assignment model based on traffic cost realizes the passenger flow distribution from the macro-perspective but lacks the consideration of passenger attributes and the time-varying characteristics of passenger travel time. Here, a novel spatiotemporal path inference model is proposed for Urban Rail Transit (URT) passengers based on the travel time data and train operation information. In this study, the impact of passengers detained at the platform on passenger travel itinerary is considered by characterizing the passenger detention rate. The proposed method realizes the reverse inference of passenger path from the micro-perspective, and can accurately describe the specific travel process of each passenger. Moreover, the real-world data of Shenzhen Metro in China is taken to verify the rationality of the proposed model. The results show that the model is in good agreement with the existing clearance model and can accurately infer the passenger travel itinerary from the micro-perspective. The proposed method provides a more refined solution for the spatial-temporal assignment of URT passenger flow.