公共自行车系统车辆库存管理研究

Managing Inventory in Public Bicycle Sharing System

考虑初始车辆对公共自行车系统运作成本的影响和站点未满足需求的溢出效应,针对站点租还车需求为随机变量且服从泊松分布,应用离散马尔科夫链理论提出基于车辆库存的站点服务损失成本的计算方法,根据溢出效应设计需求泊松参数调整方程组,构建最优车辆库存模型,并应用两个数值算例演示模型分析框架的应用,验证模型的有效性.两站点算例表明在一定条件下,未满足需求的站点间影响会极大降低损失成本,合理的站点间需求转移可作为运作优化的有效措施.华盛顿公共自行车系统多站点实例则进一步分析站点租还车需求泊松参数,成本系数,容量和初始车辆库存等因素对系统服务损失成本的影响机理.研究成果可为公共自行车车辆库存管理提供依据.

In this paper,the authors consider the impact of the initial vehicle inventory on the operation performance and the spillover effect of unsatisfied demand between stations in the public bicycle sharing system.the authors take into account the random rental and return demand following a Poisson distribution,and propose a calculation method for station service loss cost based on vehicle inventory using discrete Markov chain theory.The authors design demand Poisson parameter adjustment equations based on the spillover effect,and construct an optimal vehicle inventory model.Two numerical examples are provided to demonstrate the application of the model analysis framework and to verify the model’s effectiveness.The results of the two-station example indicate that under certain conditions,the impact between stations can significantly reduce the loss cost,and reasonable demand transfer between stations can serve as an effective measure for operational optimization.Furthermore,the authors conduct a multi-station example analysis of the Washington public bicycle sharing system to explore the impact mechanism of factors such as station rental and return demand Poisson parameters,cost coefficients,capacity,and initial vehicle inventory on the system’s service loss cost.This research provides a foundation for inventory management in public bicycle systems.