Intelligent Train Scheduling Technology for Urban Rail Transit in China: A Multi-Aspect Spatiotemporal Supply-Demand Equilibrium Theory and its Application on Engineering Operations

Abstract

Train scheduling serves as a tangible reflection of urban rail transit capacity supply, shaped by the allocation of various resources (such as section passing capacity, station turnaround capacity, depot capacity, number of trains, and crew size) in response to passenger demand. The objective for operating enterprises is to efficiently allocate resources to precisely match passenger demand with capacity supply at the lowest possible cost. With urban rail transit expanding rapidly, passenger distribution exhibits notable spatiotemporal disparities. The conventional single-service route operation mode falls short of meeting the requirements to reduce both operational costs for enterprises and travel expenses for passengers. Therefore, there is a pressing need to explore resource allocation methods in multi-service route scenarios. To address this, this study establishes a multi-objective mixed-integer nonlinear programming (MINLP) model that comprehensively considers the interests of both passengers and enterprises. Furthermore, we propose the FSME-001 (Fast Solution Algorithm for Engineering) algorithm, which enables the simultaneous generation of multiple optimization results. Finally, the actual operation of the Changping Line in Beijing is analyzed in detail, and the train schedules for peak and off-peak hours are optimized, with the optimized train schedule being compared with the train schedule used in practice. Meanwhile, the proposed model and algorithm are applied to the operation lines in Chongqing and Nanning, respectively. Studies have shown that the theoretical framework proposed in this paper can be directly applied in practical engineering to allocate transportation resources effectively, simultaneously meeting passenger travel demands and reducing operational costs for enterprises. This framework provides a generalized theoretical basis for more specific and complex applications, such as overtaking operations and cross-line network operations.

Keywords: Urban rail transit, Train scheduling, Resource allocation, Passenger flow, Engineering application