基于双向协同算法的公交时刻表及排班优化  

Bus timetable and scheduling optimization based on bidirectional collaborative algorithm

作　　者：李光春 聂磊[1] LI Guangchun;NIE Lei(School of Traffic and Transportation,Beijing Jiaotong University,Beijing 100044,China)

机构地区：[1]北京交通大学交通运输学院,北京100044

出　　处：《北京交通大学学报》2023年第6期57-64,共8页JOURNAL OF BEIJING JIAOTONG UNIVERSITY

基　　金：国家自然科学基金(U1934216);国铁集团重大课题(K2022X029)。

摘　　要：针对公交时刻表生成及排班过程中,以往研究大多建立优化模型,并采用遗传算法等启发式算法求解,存在公交时刻表与排班模型的求解效率低、求解结果不确定性,不便于公交企业的实际运营组织等问题,提出了双向协同算法的时刻表及排班优化方法.首先,建立公交单向时刻表模型、双向协同时刻表及排班优化模型;然后,融合高峰小时识别、发车频率累计曲线法、最大断面平移算法等方法,提出公交时刻表及排班双向协同优化算法(Bidirectional Coordination Optimization Algorithm,BCOA),通过双向协同算法将排班计划与时刻表进行联动,在排班的过程中同时对时刻表进行调整,对公交车次链进行协调优化,高效求解双向协同公交时刻表及排班优化模型;最后,以四川省宜宾市公交线路10路数据为例进行分析.研究结果表明:与遗传算法比较,基于双向协同算法的时刻表算法结果确定、求解效率高,求解时间减少5倍;与实际运营时刻表比较,双向协同公交时刻表在使用车辆数相同的情况下,运营车次减少了9%,下行线路运营时长增加了5%,解决了实际使用时刻表下行方向结束运营时间较早的问题,且能保证司机休息时长.In the context of generating bus timetables and scheduling,previous research primarily relied on establishing optimization models and employing heuristic algorithms like genetic algorithms.How-ever,these approaches have exhibited limitations concerning the efficiency of solving bus timetables and scheduling models.A timetable and scheduling optimization method on bidirectional collaborative algo-rithm is proposed.Furthermore,they often resulted in uncertain outcomes,rendering them less practical for operational deployment within bus companies.To address these issues,this study first establishes a unidirectional bus timetable model and extends it into a bidirectional coordinated model for bus time-tables and crew scheduling optimization.Subsequently,a novel Bidirectional Coordination Optimization Algorithm(BCOA)is introduced.This algorithm incorporates methods such as peak-hour identification,cumulative curve analysis of bus departure frequencies,and the maximum section translation algorithm.BCOA facilitates an interactive adjustment of scheduling plans and timetables,enabling the coordinated optimization of bus service chains.Finally,data from bus line No.10 in Yibin City,Sichuan Province is analyzed.The results indicate that the bidirectional cooperative algorithms outperform genetic algorithms by providing more reliable results and a fivefold reduction in solving time.Moreover,when compared to the actual operational timetable,the bidirectional cooperative bus timetable,using the same number of vehicles,reduces the number of operations by 9%and extendss the operating duration of the down-stream route by 5%.This effectively addresses issues like early termination of operations in the down-stream direction in the actual timetable and ensures adequate rest time for drivers.

关 键 词：智能公交 公交时刻表 排班优化 发车曲线平移算法 双向协调算法 

分 类 号：U492.4[交通运输工程—交通运输规划与管理]