基于动态交通网络的城市物流配送路径优化研究

发布时间：2018-04-03 18:13

本文选题：城市物流　切入点：TDVRPTW　出处：《深圳大学》2017年硕士论文

【摘要】：车辆路径问题(VRP,Vehicle Routing Problem)是现代物流管理研究中一项十分重要的问题。随着社会和经济的快速发展,人们对货物运输和配送的需求不断增加,门到门物流服务的占比不断上升,配送压力大是负责末端物流配送的物流企业面临的突出问题,同时由于城市居民人数的不断增加,城市规模的不断扩张,交通拥堵已严重影响物流末端的配送效率,这是城市物流企业所面临的又一突出问题。如何在有限的时间里,有效的调度物流运输车辆,通过合理的安排配送车辆的配送路径和出行时间,在克服交通拥堵的情况下以最小的配送成本完成配送任务,实现物流配送系统高效低成本的运作,是当今物流业亟待解决的问题。因此,研究基于动态交通网络的车辆路径问题具有十分重要的现实意义。根据以上背景,本文基于动态交通网络对带时间窗限制的时变条件下城市配送路径优化问题(TDVRPTW,Time-dependent Vehicle Routing Problem With Time Windows)进行了相关研究。论文从城市物流配送及动态交通网络的相关理论出发,针对带时间窗约束的车辆路径规划问题(VRPTW,Vehicle Routing Problem With Time Windows),考虑了物流配送车辆的启用、配送的距离及配送的及时性对配送成本的影响,构建了带有惩罚函数的以最小配送成本为目标的VRPTW路径规划模型,然后考虑动态的交通环境,将VRPTW路径规划模型进行扩充,构建了时变环境下TDVRPTW路径规划模型。通过各种算法对比分析后,确定了以遗传算法作为本文中物流配送路径规划模型的求解算法,设计了适用于求解问题模型的染色体编码方式及具有遗传进化功能的选择、交叉、变异等遗传操作。基于上述模型和优化方法,利用国际公认的标准算例库和深圳市真实物流配送案例进行验证。实验结果表明:(1)针对非时变条件下的VRPTW模型,运用本文所设计的遗传算法对国际公认的标准算例库进行了求解验证,其求解结果已接近现有最优解,表明了该算法的有效性与可行性;(2)动态交通网络下,VRPTW模型及TDVRPTW模型运用本文算法求得的结果进行对比后,发现TDVRPTW模型结果中车辆行驶的总路程与总花费时间在一定程度上大于VRPTW模型结果中车辆行驶的总路程与总花费时间,证明了交通因素对物流配送产生了影响;(3)动态交通网络环境下的物流配送更加接近于现实的城市物流配送。
[Abstract]:Vehicle Routing problem is a very important problem in modern logistics management.With the rapid development of society and economy, people's demand for goods transportation and distribution is increasing, and the proportion of door-to-door logistics services is increasing. Distribution pressure is a prominent problem faced by logistics enterprises in charge of terminal logistics distribution.At the same time, due to the increasing number of urban residents and the continuous expansion of urban scale, traffic congestion has seriously affected the distribution efficiency of logistics terminal, which is another outstanding problem faced by urban logistics enterprises.In the limited time, how to effectively dispatch the logistics transportation vehicles, through the reasonable arrangement of the distribution route and travel time of the distribution vehicles, in order to overcome the traffic congestion with the minimum distribution cost to complete the distribution task.It is an urgent problem for logistics industry to realize the efficient and low cost operation of logistics distribution system.Therefore, it is very important to study the vehicle routing problem based on dynamic traffic network.Based on the above background, this paper studies the time-dependent Vehicle Routing Problem With Time Time problem of urban distribution routing under time-varying conditions with time window constraints based on dynamic traffic network.Based on the related theories of urban logistics distribution and dynamic transportation network, this paper discusses the vehicle routing problem with time window constraint (VRPTW) vehicle Routing Problem With Time Windows, and considers the use of logistics distribution vehicle.Based on the influence of distance and timeliness of distribution on the cost of distribution, this paper constructs a VRPTW path planning model with penalty function, which aims at the minimum distribution cost. Then considering the dynamic traffic environment, the VRPTW path planning model is extended.The TDVRPTW path planning model in time varying environment is constructed.Through the comparison and analysis of various algorithms, the genetic algorithm is chosen as the solution algorithm of the logistics distribution path planning model in this paper, and the chromosome coding method suitable for solving the problem model and the selection and crossover with the function of genetic evolution are designed.Genetic manipulation such as variation.Based on the above model and optimization method, we use the internationally recognized standard case base and Shenzhen real logistics distribution case to verify.The experimental results show that the VRPTW model under time-invariant condition is solved by using the genetic algorithm designed in this paper to validate the internationally accepted standard example library. The results are close to the existing optimal solutions.The validity and feasibility of the algorithm are demonstrated. (2) the model of VRPTW and the model of TDVRPTW under dynamic traffic network are compared with the results obtained by the algorithm in this paper.It is found that the total distance and total cost time of the vehicle in the result of TDVRPTW model is to some extent larger than that of the total distance and total time of the vehicle in the result of VRPTW model.It is proved that transportation factors have an impact on logistics distribution.) under the dynamic traffic network environment, the logistics distribution is closer to the reality of urban logistics distribution.
【学位授予单位】：深圳大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：F252.1;TP18

【相似文献】