RE: Shewhart's Maximum Control

["DeRoeck, Rich" <RDeRoeck@alphaind.com>]

I have also used XmR charts to measure manufacturing cycle times
(WIP/Thruput). Even though a histogram of the data are positively skewed,
the chart is able to detect special cause variation and this information has
been used to make improvements. I prefer to use the medium moving range
statistic for calculating the process limits because it is not influenced by
extreme values. Don Wheeler has just released a new book titled: "NORMALITY
AND THE PROCESS BEHAVIOR CHART" which may address some of the questions
raised (I haven't read it).

Rich DeRoeck
QA Eng-Alpha Industries 
==========================================================================