公交接驳下共享单车投放点选址优化  

Location Optimization of Shared Bicycle Delivery Points under Bus Connection

作　　者：曹弋 张家龙 侯晓磊 CAO Yi;ZHANG Jialong;HOU Xiaolei(School of Transportation Engineering,Dalian Jiaotong University,Dalian 116028,China)

机构地区：[1]大连交通大学交通工程学院,辽宁大连116028

出　　处：《大连交通大学学报》2025年第1期15-23,共9页Journal of Dalian Jiaotong University

基　　金：辽宁省社会科学规划基金项目(L22BSH003)。

摘　　要：为了满足共享单车出行需求,改善与公交接驳的便利程度,研究了共享单车投放点的选址优化模型与算法。借鉴经典最短路径交通分配模型,在每次更新选址方案后,均重新分配共享单车流量。以投放点选址决策变量与单车配置数量为优化变量,以用户出行时间最少和企业运营成本最小为优化目标建立目标函数。基于公交接驳关系、投放点建设数量、各停放点容量和投放总量等多个因素构建约束条件,设计优化模型求解算法,开展案例研究。以厦门市共享单车出行订单数据为基础,采用自适应遗传算法(AGA)和枚举法优化求解,并分析求解精度与效率。结果表明,该模型与算法能够用较少的迭代次数求解得投放点最优选址方案的近似最优解。系统运营总成本的最小误差仅为4.4%,用户出行总时间误差为5%。就本案例而言,方案优化前、后运营成本降低12.9%,出行时间降低14.26%。此模型与算法能够处理共享单车停放点混乱与接驳公交的难题,减少出行者接驳成本,最大限度方便出行,增强接驳综合服务水准。In order to meet the demand for shared-bikes travel and improve the convenience of public transportation,an optimization model and algorithm for the location of shared bicycle placement points are studied.Drawing on the classic All-or-none traffic assignment model,the shared-bikes flow is redistributed after each update of the location plan.Using the decision variables for the location selection of shared-bikes placement points and the number of shared-bikes configurations as optimization variables,the optimization objective is to minimize user travel time and enterprise operating costs,while establishing an objective function.The constraints are established by comprehensively considering factors such as bus connection relationships,number of placement points,total placement amount,and capacity of each parking point.An optimization model is designed to solve the algorithms.Case studies are conducted.Based on the shared-bikes travel order data in Xiamen,an adaptive genetic algorithm(AGA)and enumeration method are used to optimize the solution and analyze its accuracy and efficiency.The results show that this model and algorithm can undergo fewer iterations to obtain the approximate optimal solution of the optimal location scheme for the placement point.The minimum error in the total system operation cost is only 4.4%,and the error in the total travel time of users is 5%.In this case,operating costs are reduced by 12.9%and travel times by 14.26%before and after the optimization.It is proved that the model and algorithm proposed in this article can solve the problems of chaotic shared-bikes parking points and difficulties in connecting public transportation,reduce the transportation cost for travelers,maximize convenience for travel and improve the comprehensive service level of transportation.

关 键 词：交通工程 共享单车 公交接驳 遗传算法 选址优化 

分 类 号：U491.225[交通运输工程—交通运输规划与管理]