Q-table, QA-table and QC-process chart etc. have been proposed as tools for quality goal seeking in the territory of quality control for mechanical industry. Though these tools have been used, the systematic prediction of quality is not necessarily sufficient. That's the reason why activities for quality goal seeking have a lot of trials and errors. This study proposes, in order to make activities for quality goal seeking easier and to fill up them, Relation Chart System (R. C. S.) which becomes the basis for the systematic prediction of quality. The basic structure of R. C. S. is outlined, its pictorial representation and mathematically strict definition of R. C. S. is presented. Based on its definition, some concepts of R. C. S., which can be regarded as significant and fundamental for quality goal seeking, is discussed.

View full abstract

In the practical situation, the coefficient of variation is used as a measure of stability of the product quality or the operation performance, and also it has many other applications. This paper presents approximate distribution of the coefficient of variation which can be calculated from random samples of size n having a normal universe. The approaches employed here are (1) normal approximatiom and (2) χ approximation, since the moment method can not be applied when n is small. The results are confirmed by means of simulation experiment. An application of this approximate distribution is exhibited in the process of analyzing the sample size determination to assure the desired level of precision. We analyze the distribution of square of the coefficient of variation by the same approaches. The necessary condition is also given that χ2 distribution is available for a practical and convenient approximation to the square of coefficient of variation.

View full abstract