One-way car-sharing services have drawn much attention in recent years in transportation research and practice. The flexible operating scheme stimulates patronage but causes unbalanced distributions of the shared cars and consequently low utilization of the services. In light of the empirical evidence that users are willing to cooperate with the operators for relocations, this paper formulates the supply–demand dynamics of shared cars under four user-based relocation strategies, including a naive one without incentive and three alternatives with incentives. We formulate the triggering conditions and travel disutilities responding to the incentive-based strategies based on the concept of load-space–time paths. To systematically assess the effects of the strategies, we propose an incentive-based boundedly rational dynamic user equilibrium (IBR-DUE) model in a multimodal transportation system. A cluster-based column generation algorithm is developed to solve the IBR-DUE model without path enumeration in a space–time supernetwork. Numerical examples demonstrate that the alternative drop-off incentive outperforms other strategies and can improve the utilization and service rates about 2 times on average.

近年来，单程汽车共享服务在交通研究和实践中备受关注。灵活的运营方案刺激了客流量，但导致共享汽车分布不平衡，从而导致服务利用率低下。根据用户愿意与运营商合作搬迁的经验证据，本文制定了四种基于用户的搬迁策略下的共享汽车供需动态，包括一种没有激励的朴素策略和三种有激励的替代策略。我们根据负载时空路径的概念制定了响应基于激励的策略的触发条件和旅行负效用。为了系统地评估策略的效果，我们在多式联运系统中提出了一种基于激励的有限理性动态用户均衡 (IBR-DUE) 模型。开发了一种基于簇的列生成算法来解决时空超网络中没有路径枚举的 IBR-DUE 模型。数值例子表明，替代的下车激励优于其他策略，平均可以提高约 2 倍的利用率和服务率。