使用聚类分析法截选产品结构的研究

The Study of Product Mix Selection based on Cluster Analysis

将基于工序的统计筛选法—聚类分析和Pareto图法相结合,在计算机辅助下,可加快计算过程,更有效地截取产品结构。在这一过程中,采用Pareto图法可以从产品的加工数量方面选取相对重要工序,采用聚类分析法则可从产品的工序相似方面选取重要工序。综合两种判别原则,以原待分析对象集合为产品集合,从而可以解得产品结构的选取这一制造加工中的复杂问题。在此特采用一实例详细讨论了使用Pareto图截选产品结构的局限性和使用该方法对车间布局设计的影响。

Proposed an enhanced method based on routing data-which integrates Pareto Analysis and computer-aided clustering technique in statistical analysis in order to facilitate the calculation procedures and obtain the product mix.During this process,the selection criteria of Pareto analysis and the clustering technique are comparative product quantity and similarities/dissimilarities between routings respectively.Then,by combining the two selection criteria,the practitioners could engineer a solution to a complicated problem–product mix selection based on the original product collection.A case study in this paper discusses the limitation of Pareto analysis for truncating a dataset and its impact on facility layout design.