Quality eLine Newsletter
April 2005
Vol. 7, No. 04

Quality Quiz

Click here for Quality Quiz Video Explanation


As NCAA competition heated up last month, our Marty Graw moved closer to his own goal, scoring one for capability analysis--or at least, learning how to calculate Cp:


The chart reflects the relevant data:


Marty believed he'd made a slam dunk with this calculation, but since he made a strategic error in believing that a Cp greater than 1 assured good parts, he found himself fouled at the goal. In fact, the process is producing 2.28% out-of-spec parts--too high to predict quality output. Furthermore, the process itself was not centered on the spec limits.

Fouling out of that game, Marty is back in uniform this week, armed with his SPC reference guide, Practical Tools for Continuous Improvement (Graham and Cleary). He has discovered yet another capability index that takes into account the center of the data relative to the specifications. This is not Cp, but Cpk.

where is the smallest of:

Marty pursues this calculation for the problem he is working on:

The lower of the Z values is at 2, so the Cpk for this problem is:

Armed with this impressive calculation, he makes a full court press to his boss, and dramatically writes "Cpk = .67" on the whiteboard in his office.

"This proves that 67 percent of our products are within spec limits," he announces triumphantly.

Was this a free throw gone wrong, or has Marty Graw scored one for his team?


a) A Cpk of .67 does not mean that 67% of parts within spec. Marty is wrong.

b) Marty has scored a three-pointer for his team, with his impressive calculation of Cpk.


Copyright 2005 PQ Systems.
Please direct questions or problems regarding this web site to the Webmaster.