ABSTRACT Energy‐efficient operation is a fundamental objective for intelligent urban rail transit. However, conventional train scheduling approaches typically assume a constant train mass, overlooking the influence of passenger flow on traction energy consumption. This paper presents a comprehensive scheduling framework that explicitly models the interaction among passenger flow, train mass, and energy consumption. The model focuses on the effect of reducing traction energy consumption rather than the overall energy consumption and dynamically updates train mass according to time‐varying passenger demand and determines the optimal distribution of section times to minimise general energy consumption while satisfying operational and service constraints. A dynamic programming (DP) algorithm is developed to obtain the optimal schedule efficiently without encountering dimensionality issues, enabling near‐real‐time applicability in the train scheduling approach. The proposed approach is validated through case studies on an actual subway line. Results show that the proposed schedule achieves significant energy savings by accounting for passenger‐flow‐induced mass variation, while the multi‐train schedule further adapts to time‐varying demand. Compared with the original and passenger‐flow‐independent optimal schedules, the proposed method reduces energy consumption by 10.83% and 1.56%, respectively, while maintaining passenger transport capacity and ensuring constraints. Analysis results based on power flow also show that this method can reduce system‐side energy, which demonstrates that integrating passenger flow into train scheduling enables more accurate modelling and yields substantial energy savings for urban rail systems.
Geng et al. (Thu,) studied this question.