Quality eLine Newsletter
May 2004
Vol. 6, No. 5

Quality Quiz
from Professor Cleary

Congratulations! 
You're right!

The best way to explain Cpk in a way that makes it seem intuitive is by example. If, for example, you have a process that is in control (and data follows a normal distribution), and have garnered the following information:


thus USL = 29 LSL = 14

Calculation for Cpk:

A way to understand Cpk is to note that a Cpk of 1 represents an assurance that the process is producing at least 99.73 percent good parts, since if spec limits were exactly plus 3 and minus 3 sigma (), one would be capturing 99.73 percent of the normal curve. In this case, the lower spec limit is exactly 3 sigma from the mean, and the upper spec is 4.5 sigma from the mean. Weasel words such as “at least” indicate the possibility that even more than 99.73 percent of parts are good—as is the case in this example.


Copyright 2004 PQ Systems.

Please direct questions or problems regarding this web site to the Webmaster.