【作者】 夏新海；

【导师】 张培林；

【作者基本信息】 武汉理工大学 ， 交通运输规划与管理， 2004， 硕士

【摘要】 物流配送是物流活动中直接与消费者相连的环节。在物流的各项成本中，配送成本占了相当高的比例。配送车辆调度的合理与否对配送速度、成本、效益影响很大，特别是多用户配送车辆调度的确定更为复杂。采用科学、合理的方法来进行配送车辆调度，是物流配送中非常重要的一项活动。因此，车辆调度问题(Vehicle Scheduling Problem，简记VSP)成为众多学者竟相研究的热门话题。在高度发展的商业社会中，特别是随着Internet的普及和电子商务的发展，消费者对时间的要求越来越严格，以往的到货“日”已转换成到货“时”。 VSP是一个典型的NP-难题，高效的精确算法存在的可能性不大，启发式算法虽能快速求解大型问题，但对解的质量没有保证。近些年来，人们在用遗传算法解决现实中的各种组合优化问题上进行了探索，如在生产调度问题中的应用，但在车辆调度问题中的应用才刚刚开始。有专家断言遗传算法是用来解决NP完全问题和NP难题的趋势。 本论文主要对有时间窗的非满载VSP和供应商管理库存(Vendor Managed Inventory，简记VMI)管理思想下的VSP进行了研究。 对于有时间窗的非满载VSP问题，将货运量约束和时间窗约束转化为目标约束，建立了VSP模型，使用最大保留交叉、交叉率和变异率的自适应调整等技术，设计了给予自然数编码的可同时处理软、硬时间窗约束的遗传算法，实验分析取得了较好的结果。本论文丰富了遗传算法在组合优化中的应用，为继续深入研究VSP、JOB—SHOP和物流配送车辆调度优化的计算机实现等打下基础。 对于VMI下的VSP问题，可以看作上述VSP的问题的延伸。本文分析了VMI对于供应链物流配送系统优化的作用。在VMI管理方式下，存在库存和配送运输可以集成起来进一步优化配送系统成本这一实际情况。接着对此问题建立了数学模型和迭代优化算法，实例证明该模型和算法能够起到较好的效果。 武汉理工大学硕士学位论文 物流配送车辆调度优化，是物流配送优化中关键的一环，也是电子商务活动不可缺少的内容。对货运车辆进行调度优化，可以提高物流经济效益、实现物流科学化。对货运车辆调度优化理论与方法进行系统研究是物流集约化发展、建立现代调度指挥系统、发展智能交通运输系统和开展电子商务的基础。目前，问题的形式已有很大发展，该问题以不仅仅局限于汽车运输领域，在水运、航空、通讯、电力、工业管理、计算机应用等领域也有一定的应用，其算法已用于航空乘务员轮班安排、轮船公司运送货物经过港口与货物安排的优化设计、交通车线路安排、生产系统中的计划与控制等多种组合优化问题。更多还原

【Abstract】 Logistic distribution is an operation linking with consumer directly ,and takes account for considerable proportion in variable costs in logistics. The planning of vehicle scheduling in distribution will be take great effect on the efficiency ,cost ,and benefit, especially in distributing for multi consumers. A scientific and reasonable method to vehicle scheduling is an important operation in logistic distribution. So,vehicle scheduling problem had become focus of many scholars to study. In the developed commercial society . With popularization of internet and development of electronic commerce, requirement of comsumer for delivery time is higher and higher so that delivery day formerly had rum to delivery hour now.VSP is a typical strong NP- hard problem,high effective exact algorithm is immposible to it. Heuristic Algorithm can resolve large-scale problem,but cannot ensure the quality of the resolution. Recently, genetic algorithm has been tyied to resolve variable combinatorial optimization problems, such as job scheduling problem,but but has began just now in VSP. Some people assert that genetic algorithm has tended to NP-had problem.This paper attempts to take VSP with time window and VSP in VMI as the tow core problems for further research.On the VSP with time window,while the restraints of capacity and time windows are changed into object restraints,a mathematic model is established.We use technique such as maximum preserved crossover and selfadaptability change of probability of crossover and mutation,and design genetic algorithm on nature number,which can deal with soft and hard time windows.The excellent solutions are obtained in the application.VSP in VMI may be looked as the stretch of the VSP above. On the VSP in VMI,this paper analyses the contribution of the optimization of logistic distribution’s system.Being aimed at looking for the optimal distribution policy.this paper put delivery routing and inventory into consideration jointly,and brings forward an mathematic model and its algorithm.Consequently,results from the case proved the availability of the model and its algorithm.VSP is both a pivotal tache in logistic distribution optimization and indispensable in electronic commerce.It can increase logistic economic benefit and realize logistic rationalization.The systemic study on the theory and method of VSP is the base on the growth of logistic intensivism ,the establishment of modem chain of command,the development of ITS and EC.Now,the problem is not only applied to the field of auto transportation,but also to ship avigation communication electricity industry management computer application etc.The algorithm has been applied into many combinatorial optimization problems such as the trainman’s shift arrangement in avigation the optimization design of cargo arrangement in ship company . traffic routing arrangement and the plan and control in the production system.更多还原