Service Level Requirements for Real Life–Sized Bicycle Sharing Systems
提出一个两步法管理真实自行车共享系统的服务水平要求，先计算站点目标自行车数量以最大化出行满意度，再设计车辆路线调整自行车数量，并在九个城市的数据上验证了方法的有效性。
This paper presents a two-step approach to managing service level requirements (SLRs) in real-life bicycle sharing systems (BSSs). SLR is a general concept that broadly describes how to effectively manage a BSS to improve user satisfaction while minimizing system operation costs. The two steps consist of two proposed problems: First, the target-level problem computes target bicycle quantities for stations, maximizing trip satisfaction. Second, the bicycle rebalancing problem designs vehicle routes to adjust bicycle quantities. The SLR literature includes several variants of these problems, but to our knowledge, very few exact approaches such as the one we propose can successfully handle real-life BSS, which comprise thousands of stations, tens of thousands of bicycles, and nearly 100,000 daily trips. We gather data from real-life BSSs from Boston, Chicago, Madrid, Mexico City, Montreal, New York, San Francisco, Toronto, and Washington, DC. From a managerial perspective, our numerical results provide the decision makers of BSSs with several insights related to bicycle and station usage throughout the network of stations. Funding: This work was supported by the Canadian Natural Sciences and Engineering Research Council (NSERC) under [Grant 2021-04037].