优化公交车调度的多目标遗传算法模型  

Multi-objective genetic algorithm model for optimizing bus dispatch

作　　者：张思维 魏昕怡 邱桃荣[2] ZHANG Siwei;WEI Xinyi;QIU Taorong(Nanchang University Jiluan Academy,Nanchang 330031,China;Nanchang University School of Information Engineering,Nanchang 330031,China)

机构地区：[1]南昌大学际銮书院,江西南昌330031 [2]南昌大学信息工程学院,江西南昌330031

出　　处：《南昌大学学报（理科版）》2022年第1期60-65,共6页Journal of Nanchang University(Natural Science)

基　　金：国家自然科学基金资助项目(62066027)。

摘　　要：公交车是为市民出行提供服务的“准公共”产品,它反映了一个城市的管理水平。研究如何合理规划公交车的调度是一个至关重要的问题。针对由于公交客流量易受天气影响、不同时间段客流量及其变化程度相差较大等问题,本文提出了基于“平高峰”因素的多目标遗传算法对城市公交调度进行优化。模型由三个阶段构成,首先本文根据客流数据对公交的“平高峰”阶段进行定义,并验证其合理性;然后利用乘客在站等待时间与公交的各方面成本建立以乘客出行成本和公共交通的运行成本最小的公交调度优化模型。最后将“平高峰”融入至染色体基因序列中,利用多目标遗传算法不断迭代计算出最优公交调度方案。对比结果表明,基于“平高峰”优化后的模型较于传统算法的乘客出行成本降低了13.75%且公共交通的运行成本减少1.7%,均优于传统算法。证明该方法对优化公交调度的有效性。Bus is a"quasi-public"product that provides services for citizens to travel and reflects the management level of a city.Research on reasonably planning bus dispatch plays a critical role in the maintenance of bus normal operation.Aiming at the problems of public bus passenger flow being easily affected by the weather and the large difference in passenger flow and its degree of change in different time periods,a multi-objective genetic algorithm based on the"platykurtosis-peak"factor was proposed in this paper in order to optimize urban bus dispatching.The model is composed of three stages.First,platykurtosis and peak were defined in the article through passenger flow data,whose rationality has been verified.Secondly,the smallest passenger travel cost reflected by the waiting time of passengers at the station and the smallest public transportation operating cost represented by the cost of all aspects of the bus are used to establish the bus dispatching optimization model.Finally,the"platykurtosis-peak"is integrated into the chromosome gene sequence so that the optimal bus dispatch plan can be continuously iteratively worked out by the multi-objective genetic algorithm.The test results show that compared with the traditional algorithm,the optimized model based on"platykurtosis-peak"reduces the passenger travel cost by 13.75%and the operating cost of public transportation by 1.7%,which is better than the traditional algorithm as well as proving the effectiveness of it for optimizing bus dispatching.

关 键 词：公交调度 客流平高峰 多目标优化 遗传算法 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]