機械不稼働原因の検討におけるクラスター分析の利用について

Abstract

It is imperative for a manufacturing business to maintain constant and high level machine utilization in order to perform the Just-In-Time manufaturing system effectively and to achieve higher productivity. Besides functional machine breakdowns, there frequently happenes machine stoppages due to minor mulfunctions of the equipments as well as by comitted / omitted faulty operations by operators that can be restored to the normal condition in a short time. In conventional approach for detecting causes, "Cause analysis table for defect machine utlization ratio" as well as Pareto-ABC-Analysis are adopted. This paper presents a result of the application of Cluster Analysis Technique for the given subjects which demonstrates the following advantages. (1) Critical cause factors are effectively isolated and the total picture is illustrated with comprehensive visuality. (2) It detects causes of similar categories rather easily and quickly thereby multiplier effects are expected when a single action is taken for eliminating one selected cause.