基于质保数据的备件库存控制策略研究

发布时间：2018-02-14 02:37

本文关键词： 质保数据支持向量机备件分类备件预测库存控制策略　出处：《南京理工大学》2017年硕士论文　论文类型：学位论文

【摘要】：备件资源配置是提高售后服务绩效十分有效的工具。通过分析企业售后服务系统中的质保数据可以较为准确地获取备件可靠性信息,如平均故障间隔时间(MTBF)。本文以备件响应性和经济性为目标,综合考虑零备件使用中的MTBF等,构建基于质保数据的备件分类、预测以及库存控制模型对备件库存控制策略进行优化。引入支持向量机(SVM),创新备件管理模式。利用S公司的质保数据进行算例研究,以达到一定现实约束前提条件下提高售后服务的响应性及企业经济效益的现代管理要求。本文研究内容如下:首先,介绍了本文的研究背景和意义。总结了备件分类、备件预测和备件库存管理的研究现状,并对相关理论与方法进行了阐述;其次,建立了基于质保数据的备件分类模型:将由质保数据提取的可靠性信息MTBF应用到备件分类中,构建两阶段分类模型,引入SVM对其智能分类,为后续备件需求预测和分类库存控制策略奠定基础;再次,对于间断需求的备件,建立了基于质保数据的备件需求预测模型:以因素分析法为基础,将MTBF作为影响因素输入LS-SVM进行预测;对于平稳需求的备件,提出应用基于时间序列的预测方法;最后,建立基于备件分类的库存控制策略以及基于需求分布的库存控制模型:利用需求预测获取控制策略中的需求参数,使用不同的库存控制策略进行分类管理,详细介绍了泊松分布及正态分布下的库存控制模型的计算。从而得到在一定服务水平下对应的订货量、订货点等参数,兼顾了响应性和经济性。
[Abstract]:Spare parts resource allocation is a very effective tool to improve the after-sales service performance. Through analyzing the quality assurance data in the enterprise after-sales service system, the reliability information of spare parts can be obtained more accurately. For example, the average fault interval time (MTBF). This paper aims at the response and economy of spare parts, synthetically considers the MTBF in the use of spare parts, and constructs the classification of spare parts based on quality assurance data. Prediction and inventory control model are used to optimize spare parts inventory control strategy. Support vector machine (SVM) is introduced to innovate spare parts management model. In order to improve the responsiveness of after-sales service and the modern management requirements of enterprise economic benefits under the condition of certain realistic constraints, the research contents of this paper are as follows: firstly, the research background and significance of this paper are introduced, and the classification of spare parts is summarized. The research status of spare parts prediction and spare parts inventory management, and the related theories and methods are expounded. Secondly, the classification model of spare parts based on quality assurance data is established. The reliability information extracted from quality assurance data, MTBF, is applied to spare parts classification. Build a two-stage classification model, introduce SVM to its intelligent classification, for the following spare parts demand prediction and classified inventory control strategy to lay the foundation. Thirdly, for the spare parts with intermittent demand, The prediction model of spare parts demand based on quality assurance data is established: based on factor analysis method, MTBF is input into LS-SVM as an influential factor to forecast; for spare parts with stable demand, a prediction method based on time series is proposed. The inventory control strategy based on the classification of spare parts and the inventory control model based on demand distribution are established. The calculation of inventory control model under Poisson distribution and normal distribution is introduced in detail.
【学位授予单位】：南京理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：F274

【相似文献】