移动互联背景下共享单车投放点动态布局与调度问题研究

As a novel mobile-internet travel mode, bicycle-sharing make up for the shortcomings of “the first kilometer” and “the last kilometer” service provision for the public transport in cities. Meanwhile, new research topics related to the bicycle-sharing systerm have also been brought in the field of the public transport development and urban management. Due to lack of experience in the aspects of bicycles deployment, operation and maintenance, and dispatching management, enterprises have caused several operational problems such as unreasonable layout of bicycle stations, the excessive number of bicycles placed in the stations, and the unsatisfactory scheduling plan for bicycles which cannot meet riding requirements in some districts of the city. This project intends to conduct the research effort from the following aspects: By obtaining the massive data related to the shared bicycles, we estimate the shared bicycles demand in different regions and time windows through the mobile-internet big data analysis technology; According to the space-time characteristics of the shared bicycles demand, we develop the layout model of the shared bicycles stations within the region to fully meet the needs of bicycles in this region; To meet the needs of shared bicycles in different time periods, we develop a dynamic scheduling model of shared bicycles among different bicycles stations. Moreover, based on the characteristics of this model, we design an efficient heuristic optimization algorithm to solve realistic problems in large-scale. Our objective is to reduce the operating costs and improve the management level of shared bicycle systems, while fully guaranteeing the balanced demand of the bicycles station network.

共享单车作为新兴的移动互联出行方式，弥补了城市公共交通出行“第一公里”和“最后一公里”服务供给短板同时，也给公共交通发展和城市管理带来了新的研究课题。由于企业在车辆投放、运营维护、调度管理等方面缺乏经验，造成了部分地区不同程度出现了投放点布局不合理、投放数量过多、车辆调度不及时无法满足骑行需求等运营问题。本项目拟借助海量共享单车骑行数据，通过移动互联大数据分析技术进行不同区域和时间窗的共享单车需求预测；根据共享单车需求的时空特征，构建区域内共享单车投放点布局模型，最大限度满足区域内骑行需求；为满足不同时间段共享单车的需求，构建投放点间的共享单车动态调度模型，并根据模型特征设计高效的启发式优化算法以解决大规模问题，在最大限度保障投放点网络需求平衡的同时，减低运作成本，提高共享单车系统管理水平。

本项目研究移动互联背景下共享单车投放点动态布局与调度问题研究。共享单车作为新兴的移动互联出行方式，弥补了城市公共交通出行“第一公里”和“最后一公里”服务供给短板，起到承载居民短距离出行和通勤换乘接驳的服务主体作用，有效提升了公交地铁换乘接驳的效率，在城市公共交通中发挥的作用日趋明显。在竞争的初期阶段由于企业在车辆投放、运营维护、调度管理等方面缺乏经验，造成了部分地区不同程度出现了投放点布局不合理、投放数量过多、车辆调度不及时无法满足骑行需求等运营问题，增加了共享单车系统的运营成本，影响了城市公共交通管理秩序。研究共享单车投放点布局、投放数量以及车辆调度优化等运营问题是必要且迫切的。本项目具体研究内容包括：一是基于共享单车骑行产生的移动互联大数据，预测不同区域和时间窗的共享单车需求，构建需求模型；二是在需求预测基础上研究区域内共享单车投放点布局优化问题；三是基于不同时间段的骑行需求研究共享单车动态调度优化问题。本项目研究旨在最大限度保障投放点网络需求平衡的同时，减低运作成本，提高共享单车系统管理水平，具有重要的理论价值和实际应用意义。本项目已发表3篇SCI/SSCI期刊论文，出版1部学术专著，培养硕士研究生5名。