Can I Use Cpk to Calculate the Sigma Level Here?
- November 18, 2017 at 7:39 am #55872
I have a process where the USL is 100 units with no LSL. We collected one sample per day, 12 days per month. We repeated the process over many months. All the values were between zero and few units. I calculated the Cpk using the (USL – mean)/3sigma. The Cpk in this case was very large because the mean was much less than the USL (Cpk = 30). Is it logical to use this Cpk value to calculate the Sigma level, if so (3×30 = 90????). If not what is the alternative way to calculate the sigma level?
Thank you in advanceNovember 20, 2017 at 11:00 am #201988
It means you don’t have to focus on improving this process for quality control issues. However, are there cost benefits to raising the mean much closer to the USL? Less processing, cheaper processing, fewer ….
Just 2 cents.November 27, 2017 at 9:17 am #202013
In this process, we are looking at purified water microbial counts. The maximum allowed level is 100 counts. The lowest the count, the better the water quality. Therefore, we do not want to increase the number of the microbial counts to have near the USL. My question was, under such condition, is the calculated Sigma level realistic?
RegardsNovember 29, 2017 at 5:25 pm #202021
When you have a single-sided specification it makes little sense to calculate variation from the mean. Search on this site for “cpk with one sided tolerance” or look it up in a good statistics reference book.
You must be logged in to reply to this topic.