【作者】 毕娅；

【导师】 李文锋；

【作者基本信息】 武汉理工大学 ， 机械制造及其自动化， 2012， 博士

【摘要】 全球经济使企业间的竞争与合作成为常态,企业希望能充分利用手中的资源降低制造成本,提高客户满意度。在此内因的驱动下,随着高性能计算、物联网等新兴IT技术的发展,形成了一种新的制造模式-----云制造。云制造模式将各类制造资源和制造能力互联起来,封装成标准化的制造服务,用户可以按需随时获取,同时这种制造服务安全可靠、质优价廉、绿色低碳。由于物流与制造活动存在天然的相关性,物流模式需要与制造模式一致,因此云制造模式的产生和发展带来了物流模式的变革。2010年,马云入股民营物流企业“星辰急便”便昭示了物流己经正式进入“云”时代。选址作为企业管理和决策的重要部分是企业战略成功的保证。选址决策的好坏不仅会影响设施的建设成本,而且会影响企业未来的运营成本和竞争力大小。因此,如何在云物流模式下进行科学的选址和对资源实行优化配置具有非常重要的理论意义和实践价值,对竞争环境下企业的生存和发展有重要的作用,本文希望在该领域做些有益的研究和探索。论文首先介绍了研究的背景和意义,描述了选址相关问题的经典模型以及常用的启发式算法的原理和应用技巧,在对国内外文献分析的基础上总结了以往研究中的不足。然后,研究了云物流模式的概念、系统框架以及关键技术,提出了云物流模式下的协同库存机制,指出云物流是将各种物流资源和能力虚拟化、服务化并进行集中的、智能化管理与经营,服务于多客户,实现高效协调与多方共赢的一种新的物流模式,协同库存机制是云物流模式下的库存控制方法,它强调物流资源在逻辑上的虚拟化集成和对资源的统一优化调度,目的是提高物流资源的利用率,降低企业物流成本。云物流概念和协同库存机制是后述选址-配送模型的理论基础。接着,在云物流模式和协同库存机制下分别构建了基于集合和最大覆盖的选址-分配模型。模型解决了配送中心的选址问题和需求量的协同分配问题,并在时间和容量约束、单一商品下需求点的优先级以及不确定需求问题上对模型进行了扩展。设计了基于GA-PSO的混合式启发算法。其中改进的遗传算法解决了离散空间的选址问题；改进的粒子群算法解决了连续空间的需求量协同分配问题。通过Benchmark实验验证了算法的可行性和有效性；通过真实算例、三种不同问题规模的随机算例以及与LINGO的对比实验验证了云物流模式下选址-分配方案的先进性,通过参数敏感性分析提出了云物流模式下选址-分配方案的策略。最后,对全文内容进行了总结,指明了未来有待进一步研究的方向。更多还原

【Abstract】 Under the global economy, competition and cooperation among the enterprises become a normal state, but enterprises hope to make full use of resources to reduce the production cost and make customers satisfy. For this purpose, along with the new IT technology development, such as high performance computing and Internet of things etc., a new manufacturing mode was formed, namely cloud manufacturing. Cloud manufacturing mode connect all sorts of manufacturing resources and capability together and form standard manufacturing services which customers can obtain according to their demand at any time. Meanwhile, this kind of manufacturing service is safe and reliable; good quality with low price; green and low carbon. Logistics mode should coincide with manufacturing mode for the certain correlation between logistics and manufacturing process, so the production and development of cloud manufacturing mode had the logistics mode reformed. In2010, Alibaba invested "Star Express" who is a private logistics company indicated that logistics has formally entered into a new era. The location, as an important part for enterprise’s management and decision-making, ensures the success of company’s strategies. Therefore right decision of location will not only influence the construction cost of facilities, but also affect enterprise’s future operation cost and competition. Thus how to decide the right location and optimize the allocation of resources in cloud logistics mode has very important theoretical significance and practical value, and is also important for company’s survival and development in competition. This paper hopes to do some useful research and exploration in this field.Firstly, the paper introduced the study backgrounds and significance; described classical models of something about location-allocation and fundamental principles and application skills of heuristics algorithms that are frequently used; summarized some shortcomings and insufficiency in the previous research, which was based on analysis of domestic and foreign references.Secondly, we studied the concept, system framework and key technology of cloud logistics mode, put forward collaborative inventory system under cloud logistics mode. Then we pointed out that cloud logistics is a new logistics mode which can make a variety of logistics resources and capabilities virtualization, servicization and integrated and intelligent management and operation, and service for different customers in order to achieve efficient coordination and win-win. Collaborative inventory system is an inventory management method under cloud logistics mode, which emphasized the virtual integration of resources in logic and unified optimization. The purpose is to improve resource utilization rate and reduce enterprise logistics cost. The concept of cloud logistics and collaborative inventory system is a theoretical foundation for the following location-allocation model.Thirdly, in the cloud logistics mode and collaborative inventory system, location-allocation model was respectively constructed on basis of set covering location and maximal coverage location. The models solved the problem about location of distribution center and collaborative allocation in demand. Then, we expanded the model in the following three aspects:the time and capacity constraints, the needs priority of single products; the uncertain demand. We also designed the hybrid heuristic algorithm based on GA-PSO, in which the improved GA solved the location problem of discrete space and the improved PSO solved the collaborative allocation of continuous space in demand. Through Benchmark test, we certified the feasibility and effectiveness of algorithm; Through real data test, three groups simulation data tests of different problems and comparison test with LINGO, we certified the advancement of location-allocation scheme in the cloud logistics mode; Through data sensitivity analysis, we put forward the strategy of location-allocation scheme in cloud logistics mode.Finally, we concluded the whole content of this paper and presented direction for further studies.更多还原