RE: Special Causes Indicators in Red Bead

[Steven_S_Prevette@rl.gov]

>From S Cheek DDS:  I agree that you the manager of the system determine
upper and lower limits 
(common and special causes) and make a judgment cause what to try to
improve. 
It is assumed to take action on special causes would be overreacting. I was 
looking for how one may break either down a small step in process 
improvement.     
--- end quote ---


Actually, it is the special causes you want to act upon.  If you get a
signal on a control chart, Dr. Shewhart's theory is that it is likely that
you will find something economically justifiable to work on.  Working on
special cause variation is relatively easy - there should be a readily
identifiable cause for the unusual result.

What is tougher to work on is common cause variation.  One should not go try
and "fix" individual results that exhibit common cause variation - exhibit
stability on a control chart.  To do so would be tampering.  What can be
done is to look at the long run performance of the system, and try to look
for common sources of defects, common issues across the board.  Pareto
charting can help considerably in this process.  Then what you find should
be applied to the entire system (not just certain individuals or certain
months).


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