Most industrial uses of process capability ratio focus on computing and interpreting the point estimate of the desired quantity. Furthermore, the process measures are estimated based on a single random sample of observations from the normally distributed process that is in statistical control. In this article, we not only make good use of a better estimator of σ for moderate large sample size, but also consider various sample size to construct the lower confidence bounds based on subsamples in some detail in connection with minimum values for the process capability index Cpm. Some examples are also given to illustrate the procedure for checking if the process meets the capability requirement based on the lower
confidence bounds.
Relation:
International Journal of Information and Management Sciences 19(4), pp.601-620