Enhancing resource sharing in urban rail transit: a rolling stock sharing strategy for multi-line timetable optimization in cross-line operations

Cross-line operation allows trains to travel between intersecting metro lines, which can provide direct travel for some transfer passengers, thereby alleviating transfer demands at transfer stations. Cross-line operation in urban rail transit allows trains to travel between intersecting lines, reducing transfer demands by enabling direct trips. However, existing studies focus on single-line optimization, lacking strategies for coordinated resource allocation across lines. This paper proposes a rolling stock sharing strategy to enable the sharing of train resources among different lines in cross-line operations. Taking the complex travel processes of both direct and transfer passengers into account, a mixed-integer nonlinear programming model (MINLP) is formulated to optimize train timetables and train resource allocation in cross-line operations, so as to minimize passengers’ waiting time and operating costs. A hybrid algorithm that combines a genetic algorithm, an adaptive large neighborhood search algorithm, and a train operation conflict elimination strategy is designed for the proposed model to find high-quality solutions. Finally, using Line 8 and the Changping Line of the Beijing Metro as a case study, results analysis and five sets of numerical experiments are conducted to prove the effectiveness of the proposed method. The experimental results demonstrate that the method can enhance train resource sharing among depots, improve transport efficiency, and reduce operating costs.