基于订单数据分析的共享单车重置调度优化研究

无桩共享单车打破了传统公共自行车基于固定桩点提供骑行服务的限制，但如何重置调度随机分布于整个区域面上的共享单车则成为决策者面临的一大难题，而数据驱动的优化方法则为解决这一问题提供了可能。本文首先通过对居民使用共享单车的订单数据进行深入分析，挖掘用户出行规律，继而界定出区域内共享单车的收车点与放车点；进一步地，借鉴报童模型思想构建放车点单车投放的收益函数，在此基础上建立带取送货的共享单车重置路径规划模型并设计算法进行求解。通过算例测试，本文从重置车辆的最大载重量、单位距离成本、重置车辆里程限制、收车点个数以及收车点所拥有的共享单车数等多角度，为共享单车的重置调度提供有效的决策建议。

Modeling and Optimizing Method for Rebalancing the Dock-less Bicycles based on Order Data Analysis

The dockless sharing bicycle system breaks the limitation of traditional public sharing bicycle system because the latter relies on the fixed stations to provide travel service. Therefore, it greatly improves the user's riding experience. However, how to rebalance the dockless bicycles, which are randomly distributed on the whole area, has become a big problem faced by decision makers. The data-driven operation research method provides the possibility to solve the above problem. In this study, the large-scale order dataset of the dockless sharing bicycles is analyzed to explore the travel rules of the users. Then the collection points and the release points are defined. A newsboy model is referenced to construct the innovative revenue function when releasing dockless bicycles at the release points. After that, an optimization model for rebalancing the dockless sharing bicycles by considering both pick-up and delivery aspects is proposed. The test results demonstrate that the proposed method can provide effective decision-making suggestions for rebalancing the dockless sharing bicycles from the perspectives of the maximum loading capacity, the unit distance cost, the driving distance limitation, the number of collection points and the number of bicycles in the corresponding collection points.