Enhancing Efficiency and Fairness: Rolling Stock Scheduling with Virtual Coupling and Short-Turning Strategies

In this paper, focusing on an urban rail transit line, we propose a novel operational mode that combines the short-turning strategy and flexible train composition with the technology of virtual coupling, in order to effectively match transportation capacity with passenger demand in both time and space dimensions, while ensuring fairness in service provision. Using a spatio-temporal representation, we formulate a Mixed Integer Linear Programming model that generates the rolling stock schedule with virtual coupling and short-turning strategies, with the goal of optimizing the operational cost, passenger service quality, and fairness. Considering the complexity of the model in addressing large-scale problems, we develop three types of exact algorithms, that is, Branch-and-Benders-cut (BBC), Section-based decomposition (SBD), and Branch-and-Section-Add (BSA), on the basis of the mathematical properties of the proposed model. To validate the proposed approaches, a series of experiments are conducted by using the real operation data from Xi’an metro Line 3. Experimental results demonstrate that our designed algorithms are able to solve the large-scale instance effectively, outperforming the commercial solver. Moreover, our presented strategy also exhibits superior performance in comparison with the currently used operation scheme on this experimental metro line. Funding: This research was supported by the National Natural Science Foundation of China [Grant 72288101] and the Fundamental Research Funds for the Central Universities [Grant 2024YJS103]. Supplemental Material: The online appendix is available at https://doi.org/10.1287/trsc.2024.0687 .