Revise Control Limits present Sigma level
May 26, 2004 at 3:43 pm #26124
Hi,We have a process where we need to calibrate a group of people to record customer call information based on set guidelines/parameters. As there are certain judgemental pieces of information, we have variation among different persons. Initially, we started taking the variation limits as +/-10.00%. Now, after we have some months of data, we checked the Cpk which was >1. Hence the limits were revised to +/-3 sigma (around +/-7%) with Cpk still >1. Can someone help me in clearing some doubts? 1) Can we find out the current sigma level? Is it 3 sigma? 2)How and when shall we revise the limits again?0June 3, 2004 at 3:07 pm #63477
Bob JParticipant@Bob-J Include @Bob-J in your post and this person will
be notified via email.
I’ll see if I can help…
If I understand correctly, you have currently established +/- 3 standard deviations around the mean as your guideline. Since your Cpk is based on the number of standard deviations between your mean and your nearest guideline, I am not surprised that it is approximately 1. This is essentially the same technique used in control charting to determine whether a process is consistent and repeatable….
Usually, the guidelines are established by the overall objective for the project or the needs of the customer. As such, guidelines are usually established as a hard target and not as a number of standard deviations from the mean. In this situation, the only time you change the standard is when the customer expectations change or a new improvement project has been established with a new goal.
Hopefully this helps with your second question…
As for the first question, the answer is yes…. Assuming a stable process (which it looks like you have), the sigma level will be 3 times the Cpk value.
The forum ‘Software/IT’ is closed to new topics and replies.