考虑驾驶行为的租赁预订控制与激励策略研究

通过构建随机动态规划模型分析了考虑两类顾客驾驶行为可转变的汽车租赁预订容量控制问题,系统的考察了当企业运用智能设备对顾客行为进行监测,并事后对顾客实施价格补贴策略时,对汽车租赁的预订限分配以及顾客行为转变的动态影响过程.由于动态规划模型维度较高,提出单日决策收益与多日决策收益(周期性决策)两种近似算法进行求解,并通过数值模拟验证了两种算法的有效性.研究给出了提前期随机和租期不确定前提下,预订限的基本分配原则,得到以下结论:1)多日决策收益更逼近最大期望总收益;2)当顾客行为不变时,租赁企业在实施补贴策略时,期望总收益会随着补贴的增加而单调不增;3)当补贴策略促使顾客选择良好行为的可能性增加时,增加补贴反而会增加企业的期望总收益.研究结果将为汽车租赁企业的预订决策提供支持.

Capacity booking control and incentive strategy in the rental system based on driving behavior

Considering the transfer of the customers' behavior, this paper has builded a stochastic dynamic programming model to study dynamic booking capacity control in the car rental system. When the customers' driving behavior can be monitored by intelligent equipment, we have researched how the price subsidy policy affected the capacity booking control process and customers' behavior. Because of the higher dimension of the model, we have proposed two dynamic programming decomposition approaches. One decomposition has approached dynamic programming by daily (ADPD), the other has approached dynamic programming by periodicity (ADPP). Finally, numerical simulations have verified the effectiveness of the proposed algorithms. The finding gives the principal of the capacity booking control, and the conclusions are as follows:1) The expectation profit of ADPP gets closer to maximum expectation profit; 2) When the customers' behavior has not been changed, the expected total profit will not increase with the increase of subsidy; 3) When the subsidy strategy increases the possibility of choosing good behavior, the increase of subsidy will increase the expected total profit of the enterprise. The results will provide a support for capacity booking control in the car rental system.