Time Domain Optimize in an Urban Rail Transit Line Based on Passenger Flow Spatial and Temporal Distribution

Abstract

Focusing on the time domain optimization problem of an urban rail transit line, this paper constructs a passenger travel network with OD passenger flow data as input, by using a multi-path search algorithm based on dynamic cost to deduce the passenger space-time path. The passenger travel path is restored and the spatial and temporal distribution of passenger flow is calculated. Based on this, considering the influence of passenger flow spatial and temporal distribution on the time domain division, the orderly clustering method is used to optimize the time domain. Factoring in the influence of line capacity constraint, train running sequentially on time domain division and bidirectional time domain, a time domain optimization framework for an urban rail line is proposed in this study to integrate the time domain optimization results and improve the adaptability of optimization method. A practical line is taken as an example to verify the effectiveness of the proposed framework. Compared with the traditional time domain division method, the time domain division result accuracy is significantly improved and lays a foundation for the formulation of train service scheme which accurately matches transport capacity to demand.