【作者】 李雪；

【导师】 高阳；

【作者基本信息】 中南大学 ， 管理科学与工程， 2013， 硕士

【摘要】 科学技术的进步,环保意识的增强,促使越来越多的国家、企业将目光转向废旧品的循环再利用。再制造作为废旧品循环再利用的高级处理方式,广泛的应用于现代制造业中。再制造的成功实施,离不开物流设施的支撑,合理的设施分布,不仅能够帮助企业节约成本,还能方便消费者,提高顾客满意度。目前,再制造物流网络主要有三种设计模式,即扩建设计、独立设计和集成设计。当前的研究主要集中于独立再制造物流网络的构建,且仅从单周期、单目标、静态层面进行探讨,忽略了集成物流网络的构建以及合理的物流运作计划是基于多个周期、涉及多方利益主体、存在动态变化等问题。针对以上问题,论文以导师承担的国家自然科学基金项目为选题背景,拟对现实生活中应用广泛的两种网络—独立网络和集成网络的设施选址问题进行研究。在考虑网络中废旧品回收量和再制造品需求量不确定的基础上,研究了基于多个周期、多方主体的独立和集成网络设计问题。主要工作及成果如下：(1)针对废旧品回收量的不确定性,构造等维无偏灰色马尔科夫预测模型,用以预测废旧品的回收量。论文分析了灰色GM(1,1)模型和马尔科夫模型的优缺点,柔和两种模型,给出等维无偏灰色马尔科夫模型的原理及计算步骤,并通过实际算例,验证模型的有效性和精确度,结果表明：等维无偏灰色马尔科夫预测模型能够有效克服灰色GM(1,1)和马尔科夫模型的缺陷,其精确度明显优于其他两种模型。(2)给出了独立再制造逆向物流网络的物理结构图,在此基础上,建立了同时考虑制造商成本最小和再制造过程对居民负效应最小的多周期设施动态选址模型。模型中,废旧品回收量及再制造品需求量的不确定性通过第三章的预测模型和不确定规划理论加以处理。通过算例求解,验证了模型的有效性,并对比分析了动态选址和静态选址的异同以及库存对选址决策的影响。(3)对独立逆向物流网络进行了适当扩展,给出了集成物流网络的物理结构图。在此基础上,建立了集成网络的多周期选址模型。模型同时考虑构建成本最小和居民负效用最小。通过算例求解,验证了模型的有效性,并分析了不确定因素对选址策略的影响,对比分析独立网络和集成网络的异同。更多还原

【Abstract】 The progress of science and the enhancement of environment protection consciousness prompt a growing number of countries and companies turning to the recycling of waste products. Remanufacturing is widely used in modern manufacturing as an advanced recycling approach. The successful implementation of remanufacturing is inseparable from the support of logistics facilities. Reasonable facilities distribution not only can help companies save costs, but also convenient for consumers, improve customer satisfaction.At present, there are three main kinds of remanufacturing logistics network design model which are expansion design model, independent design model and integration design model. The current research mainly focuses on the construction of independent remanufacturing logistics network with single cycle and single objective, and models are static. All those ignore the construction of integrated network and the fact that reasonable logistics operation plan is based on the multiple times, involving different interest subjects and can dynamic change.To solve these problems, the independent and integrated networks which are widely applied in real life are studied. Considering the uncertainty of the number of waste products and the demand of remanufactured goods, this paper gives independent and integrated network facilities location models which based on multiple cycles and various participants. Main work and results are as follows:(1)To solve the uncertainty of waste products, Unbiased Grey Markov forecasting model was proposed. His paper analyzed the advantages and disadvantages of GM (1,1) model and Markov model. And then, it gave the principle and calculation steps of Unbiased Grey Markov model which was the combination of Grey model and Markov model. An actual example was given to verify the validity and accuracy of these models. The results show that Unbiased Grey Markov forecasting model can overcome the defects of Grey and Markov models, and its accuracy was better than other two models. (2)Gave the physical structure of independent network and established an dynamic location model which targets are to minimize the remanufacturing costs and the side effects on residents. In this model, the number of waste products and the need of remanufactured goods are uncertainty which can be solved by the forecasting model in chapter three and the uncertain programming presented in chapter two. To verify the validity of the model, a numerical example was given. What’s more, through studying the example, we could realize the similarities and differences between static model and dynamic model as well as the inventory’s impact on the location decisions.(3)An integrated network structure was given which was the extension of independent network. And on this basis, a multi-time and multi-objective facilities site model was built with the same targets. A numerical example was given to high light the differences between independent network and integrated network and the impact of uncertainty factors on site strategies.更多还原