基于需求特性的备件分类模型研究

Classification Models of Spare Parts Based on Demand Characteristics

备件分类是备件需求预测研究的基础。只有在正确的备件分类的基础上才能更好地对备件需求预测进行研究。主要从不常用备件的特点入手,从需求数据中的平均需求量、需求间隔时间以及需求量的变化程度对备件进行了粗分类。在粗分类的基础上,提取平均需求间隔和非零需求值的变异系数两个参数对分类模型进行量化分析,上述需求属性不仅仅要考虑需求的间断性和块状性,还要考虑突发性需求。将突发性需求考虑在内的拓展的突发性需求属性空间将大大提高需求分类的准确性。该分类方法的应用是通过将分类属性变量映射到最佳预测方法,通过映射得到给定需求序列的最佳预测方法。

Spare parts classification is the basis for spare parts demand forecasting research.Only on the basis of correct spare parts classification,the research on spare parts demand forecasting can be carried out better.Mainly starting from not commonly used spare parts features the needs of the average demand data,time and extent of changes in demand interval demand for spare parts are rough classified.On the basis of rough classification,average demand interval and coefficient of variation of the two parameters of non-zero values are extracted for classification model needs quantitative analysis,these needs not only to consider the needs of property intermittent and massive resistance,but also consider the unexpected requirements.The sudden demand for property space to expand into account sudden demand will greatly increase demand classification accuracy.The classification method is applied by the classification attribute mapping variables to predict the best method,the best prediction method for a given sequence is obtained by mapping needs.