When conducting long term process capability studies, and the data is from multiple machines and/or fixtures, should the process capability be looked at by combining all of the data or by looking at the data per machine, fixture, etc.?
You should never mix process streams unless first testing whether they are statistically different.
Study each stream separately first. Then you can perform a t-test or ANOVA depending on the number of streams to determine whether these streams are statistically different. If they are not, you could then combine them.
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to the use of cookies.