RE: REQUEST: Histograms and process capability

[Steven_S_Prevette@rl.gov]

   I am asking, why one would not simply use histograms instead of Cp and
Cpk indices to measure process capability?

I assume you are proposing to collapse the process data into one
distribution histogram and calculated Cp and Cpk from it.  Mathematically it
can be done, but it can prove to be meaningless.  One of Dr. Deming's major
thoughts about statistics is the importance of maintaining the time series
of the data, and the determination of whether or not the process is stable.
The histogram will not tell you if the process is stable or not.  Cp and Cpk
taken from a process that is not stable are meaningless numbers.   

This is a similar situation as Dr. Deming's discussions of the failures of
"hypothesis testing" and tests for level of significance.  Cp and Cpk are
basically forms of hypothesis testing.

Steve Prevette
Site Technical Authority for Statistical Trending
Environment, Safety and Health
Fluor Hanford, A Fluor Global Services Company
ASQ Certified Quality Engineer
steven_s_prevette@rl.gov
509-373-9371