In station-based one-way carsharing system, the asymmetric demand-supply issue represents a toughing challenge. This problem affects the carsharing system’s level of service as well as the financial viability, and requires the engagement of a large amount of resources in redistributing the sharing cars to meet travelers’ need. Firstly, this paper proposes an approach involves the day-to-day dynamics of traveler’s ’learning behavior’ together with an adaptive incentivization scheme of the carsharing operator. Secondly, on each day, carsharing travelers make the route choice decisions according to their perceived travel costs, which can be affected by the past experience and the incentivization scheme of the carsharing operator. More specifically, the adaptive scheme does not require specific information about travelers’ behavior traits, is adopted by the operator so as to motivate travelers to rent their car from an over supplied station and/or return it to an under supplied station, thereby reducing the expected cost of relocating the cars using dedicated staff. What is more, travelers tend to discount the value of the incentive, making it less effective in relocations. Then, the equilibrium state and stability of the evolution model is examined. Finally, numerical experiments are conducted to illustrate the application of the approach.

在基于车站的单向汽车共享系统中，需求不对称问题代表了严峻的挑战。该问题影响到汽车共享系统的服务水平以及财务可行性，并且需要投入大量资源来重新分配共享汽车以满足旅行者的需求。首先，本文提出了一种方法，该方法涉及旅行者“学习行为”的日常动态以及驾车者的自适应激励机制。其次，拼车旅行者每天都根据其感知的旅行成本做出路线选择决定，这可能会受到拼车操作者的过去经验和激励计划的影响。更具体地说，自适应方案不需要有关旅行者行为特征的特定信息，由运营商采用，以激励旅行者从供过于求的站点租用汽车和/或将其返还至供不应求的站点，从而降低了使用专门人员重新安置汽车的预期成本。而且，旅行者倾向于打折奖励的价值，从而使其在搬迁中的效力降低。然后，研究了演化模型的平衡状态和稳定性。最后，进行了数值实验以说明该方法的应用。研究了演化模型的平衡状态和稳定性。最后，进行了数值实验以说明该方法的应用。研究了演化模型的平衡状态和稳定性。最后，进行了数值实验以说明该方法的应用。