不确定环境下城市公共自行车动态网络设计优化问题研究

Public bike sharing systems have been developing rapidly in many Chinese cities to encourage people cycling and using this low-carbon transport mode to make their trips. It is therefore important to strengthen the public bike network planning and management in China in order to increase the attractiveness of cycling in Chinese Cities. In view of the existence of uncertainties and dynamics in demand and supply and the presence of inaccurate data in every bike sharing system, this project will investigate the dynamic public bike network design problems with uncertainties to improve the efficiency and the reliability of public bike sharing systems. This project will develop optimization models that consider uncertainties and dynamics for bike network design problems, bike demand management problems, and bike repositioning problems, and then extend the above models to bi-level models based on the analysis and the modelling of the cyclists' travel choice behavior. Based upon metaheuristics such as artificial bee colony algorithm, genetic algorithm, chemical reaction optimization algorithm, and tabu search, a hybrid metaheuristic will be proposed to solve the models and determine solutions efficiently and accurately. Finally, using the actual transport network data of big cities worldwide, case studies will be carried out to test the effectiveness of models and the hybrid metaheuristic. This project will help to develop the theory of public bike network design, and lay a good theoretical basis to boost the planning and management of green and low carbon transport systems in China.

公共自行车是我国城市正大力发展的低碳交通出行模式，加强公共自行车网络的规划和管理是提升自行车出行吸引力的关键所在。鉴于公共自行车网络供需的不确定性和动态性以及数据的不准确性，本项目拟针对公共自行车网络设计、公共自行车需求管理和公共自行车调配优化等典型问题，分别建立不确定环境下公共自行车动态网络设计优化模型；基于自行车用户出行选择行为的分析和建模，提出不确定环境下公共自行车动态网络设计优化问题的双层规划模型；以人工蜂群算法、遗传算法、化学反应算法、禁忌搜索等为基础，设计多元混合启发式算法实现公共自行车网络设计优化模型的快速、精确求解；选取国内外大城市中典型的公共自行车系统进行实例研究，检验提出的模型和算法的有效性。本项目的研究有助于城市公共自行车网络设计优化理论的发展，为提高我国城市绿色低碳交通系统的规划和管理水平奠定良好的理论基础。

公共自行车是国内城市正大力发展的低碳交通出行模式，加强公共自行车网络的规划和系统的管理是提升自行车出行吸引力的关键所在。鉴于公共自行车网络和系统供需的不确定性和动态性以及数据的不准确性，本项目考虑自行车网络基础设施合理布局问题、公共自行车需求管理问题和公共自行车调配优化问题的不确定性或动态性，建立了一系列公共自行车网络设计优化模型，必要时还会将自行车用户的出行选择行为纳入这些模型；以基于代理模型的启发式算法、人工蜂群算法、化学反应优化算法和遗传算法等为基础，设计混合算法实现公共自行车网络设计优化模型的快速、精确求解；选取国内外城市中典型的公共自行车网络和系统进行实例研究，检验提出的模型和算法的有效性。项目执行过程中，在“Transportation Research Part B”、“Transportation Research Part C”、“Transportation Research Part D”、“Transportation Research Part E”、“Networks and Spatial Economics”、“International Journal of Sustainable Transportation”等SCIE或SSCI源期刊上发表标注论文11篇，另外两篇论文已被接受发表在“Transportation Research Part A”和“Transportation Research Part E”上。本项目的研究有助于城市公共自行车网络设计优化理论的发展，为提高我国城市绿色低碳交通系统的规划和管理水平奠定良好的理论基础。