【作者】 于楠；

【导师】 胡祥培；

【作者基本信息】 大连理工大学 ， 电子商务与物流管理， 2008， 硕士

【摘要】 物流配送过程中存在着大量不确定性,如交通事故、道路堵塞、车辆故障以及天气变化等诸多不确定因素,这些变化常常导致物流服务供应商最初制定的配送计划不能顺利实施,甚至变得不可行,给正常的物流配送服务造成时间延迟,使物流配送系统受到干扰事件的影响。如何在干扰事件发生与发展过程中实时有效的对干扰事件进行处理和应急决策,是目前干扰管理研究的难点问题之一。由于物流配送系统涉及了物流供应商、客户等多个主体,扰动发生后,如何权衡各方利益形成应对扰动的方案,如何快速有效地处理干扰事件并形成系统扰动最小的调整方案,是解决物流配送干扰事件应对方案决策问题的关键。另外,由于干扰事件的产生、发展、衰退、结束等具有一定的规律性,因此,如何在干扰事件演进中进行决策,是该领域值得研究的问题之一。本文的主要研究工作如下:(1)从物流配送系统可恢复性干扰事件的分析入手,建立了可恢复性干扰事件的延迟时间数学模型,以该模型为基础,并综合物流服务供应商和客户两个主体,提出干扰事件的时间延迟度量模型,为可恢复性时间延迟干扰事件应对方案决策问题提供定量化度量依据。(2)结合物流配送系统可恢复性时间延迟干扰事件的应对方案决策方法方面存在的困难,提出一种多阶段的决策方法——序贯决策方法,解决应对方案的分阶段实时决策问题,并给出了多阶段决策环境下应用时间延迟度量模型进行决策的具体步骤。(3)以序贯决策方法为基础,将成果应用于中石油大连配送分公司某片区成品油配送中的时间延迟干扰事件的应对方案决策过程中,说明了该方法的可行性。处理时间延迟干扰事件的序贯决策方法,根据干扰事件产生、发展、衰退、结束的一般规律,将延迟按时间段进行划分。在每一个阶段初,通过综合考虑物流配送公司及其客户两个主体,对所有可能的策略进行评价,形成相应的最优决策。该方法可以解决干扰事件发生与发展过程中进行分阶段实时决策的问题,为实际的物流配送干扰管理提供科学有效的工具。更多还原

【Abstract】 In the distribution system numerous uncertainties occur because of the disruption factors such as accident, traffic, vehicle breakdown, bad weather etc.. These troubles usually come beyond the original distribution plan and even make the routing infeasible, which cause the negative affections in distribution system. Researchers focus on the problem of how to effectively deal with the disruptions in real time and make decision in the beginning and development of the disruption events in the disruption management.The distribution system includes subjects like distribution service supplier and its customers. It is crucial to balance the benefits between both sides and to generate optimal adjustment scheme with the least deviation. Furthermore, for the reason that there exits regularity in the disruption process of causing, developing, declining, terminating, it is a valuable issue to provide a methodology of handling disruptions in the time with disruption evolution for decision makers.The main researches in this paper are as follows:(1) Begin with the analysis of recovery disruption in distribution system, a mathematics model of delay time is set up. Then a metric model for time delay is presented with the concerning of the distribution service supplier and its customers, which provides decision maker quantitative metrics basis for tackling scheme decision method of recovery time delay disruption in distribution system.(2) In the consideration of the problem of tackling scheme decision method for recovery time delay disruption in distribution system, a multi stage decision method called sequential decision method is presented to solve the real time decision making problem in multi stage. Decision making steps is detailed in multi stage environment with the application of time delay metrics model.(3) The time delay disruption of gas distribution system in the north area of Dalian Distribution Center of China Petroleum Corporation is illustrated on the basis of sequential decision method to explain its feasibility.The sequential decision method, which deals with the time delay disruption, divides the delayed time in intervals based on the disruption regularity of causing, developing, declining, terminating. At the beginning of each stage, optimal scheme is chosen after all possible tacking scheme evaluated considering both subjects of distribution service supplier and its customers. This method resolves the real time decision making problem in the process of disruption, and provides the decision maker effective tool for practical distribution disruption management.更多还原