异质需求下响应型接驳公交车速引导与路径规划协同研究  

Collaborative Study of Speed Guidance and Path Planning for Responsive Feeder Transit with Heterogeneous Demands

作　　者：常玉林[1,2,4] 蔡宇航 孙超 王建 CHANG Yu-lin;CAI Yu-hang;SUN Chao;WANG Jian(School of Automotive and Traffic Engineering,Jiangsu University,Zhenjiang 212013,Jiangsu,China;Jiangsu Key Laboratory of ITS,Southeast University,Nanjing 211189,China;School of Transportation,Southeast University,Nanjing 211189,China;School of Automotive Engineering,Nantong Institute of Technology,Nantong 226002,Jiangsu,China)

机构地区：[1]江苏大学,汽车与交通工程学院,江苏镇江212013 [2]东南大学城市智能交通江苏省重点实验室,南京211189 [3]东南大学交通学院,南京211189 [4]南通理工学院,汽车工程学院,江苏南通226002

出　　处：《交通运输系统工程与信息》2023年第5期83-95,共13页Journal of Transportation Systems Engineering and Information Technology

基　　金：国家重点研发计划(2021YFB1600100);国家自然科学基金(71801115);教育部人文社会科学研究基金(22YJCZH153)。

摘　　要：为加强“轨道+公交”网络融合,进一步提升公交出行的乘客满意度和运营效率,本文设计了响应型接驳公交车速引导与路径规划协同优化模型。首先,基于乘客接驳出行需求的异质性请求信息,通过规范化处理统一纳入模型;以区间车速表征实际情况下的道路弹性行驶时间,通过车速引导策略协同规划公交运行线路,扩大模型优质解空间。其次,根据响应型接驳公交大规模和高动态的特性,采取滚动时域优化思想,通过周期驱动将服务时间划分时域,在每个时域开始时调用静态规划模型,并设计文化基因算法求解;通过事件驱动将动态请求分类,对即时请求采用最邻近算法插入已规划路径。最后,运用Sioux Falls网络算例对比验证模型和算法的有效性,并应用于以许昌东站为背景的实例进行分析。结果表明:所构建的协同优化模型能依据接驳出行需求异质性,利用车速引导策略优化响应型接驳公交路径规划;协同优化后,满载率提高10.8%,运营成本和人均时间成本分别降低12.85%、10.79%。To enhance the integration of"rail and bus"network and further improve passenger satisfaction and operational efficiency of bus services,this paper proposes a collaborative optimization model with responsive feeder bus speed guidance and path planning.First,the heterogeneous request information of passenger's transfer demand was incorporated into the model through a standardized processing.The interval speed was used to represent the road travel time in the actual situation,and the bus operation route was planned cooperatively through the speed guidance strategy to expand the high-quality solution space of the model.Then,according to the large-scale and high dynamic characteristics of the responsive feeder bus,the service time was divided into time domains by cycle driving.The static programming model is called at the beginning of each time domain,and the cultural gene algorithm is designed to solve the model.The dynamic requests are classified by event driven,and the immediate requests are inserted into the planned path using the nearest neighbor algorithm.At last,the effectiveness of the model and algorithm is verified using the Sioux Falls network example,which is applied to the case of Xuchang East Railway Station in China.The results show that the collaborative optimization model can optimize the responsive feeder bus route using the speed guidance strategy according to the heterogeneity of feeder travel demand.After the collaborative optimization,the load factor was increased by 10.8%,while operating costs and per capita time costs decreased by 12.85%and 10.79%,respectively.

关 键 词：城市交通 协同优化 文化基因算法 需求响应接驳公交 车速引导 异质性 

分 类 号：U121[交通运输工程]