ABSTRACT In recent years, the one‐way car sharing system has gained popularity as a common method of automobile sharing in metropolitan traffic systems. However, issues such as uncertain user requests and imbalanced vehicle distribution pose significant operational difficulties. This study proposes a rolling horizon‐based systematic integrated scheduling planning (RSISP) framework, which integrates vehicle relocation and staff movement to enhance operational efficiency. The scheduling problem is formulated as a spatiotemporal network flow model to maximize operator profit, incorporating real‐time traffic conditions and fluctuating user requests. To handle the computational complexity of the model, we design a decomposed heuristic optimization algorithm that decomposes the scheduling problem into two subproblems: vehicle relocation and staff movement. Through iterative optimization, this approach achieves near‐optimal solutions within a feasible timeframe. Finally, extensive numerical experiments based on actual traffic data from Xi'an validate the effectiveness of the proposed model and solution method. The results demonstrate that the RSISP framework effectively determines the optimal fleet size and staff allocation, ensuring user demand fulfilment. Furthermore, the study highlights the importance of selecting an appropriate rolling horizon cycle length for balancing operational costs and performance.
Chen et al. (Thu,) studied this question.