RE: SPC Theory Question

[Steven_S_Prevette@rl.gov]

Question posted:  isn't it better to have many managers start using the
X-mov R
chart to have them apply variation concepts, even in situations where it is
not the right chart, instead scaring them away by trying to be statistically
rigorous?

Another way to put it would be: in practice, how statistically rigorous do
we have to be?
------

I would completely support anyone using X-mov R on all data.  Should be good
enough in 99% of the cases, especially in manufacturing.  I tend to run into
a lot of event counting in the Environment Safety and Health world, so in my
case the majority of my charts are c or u charts.

"Rigor" is an interesting word.  I would say that yes, we need to be
statistically rigorous.  The X-mov R is plenty rigorous.  Unfortunately, I
have seen the word "rigor" used as an accusation - that this has been made
too hard.  I prefer seeing the word rigor in a positive light.

On the discussion of sigma vs mov R - at least in my experience the two
calculations have given the same results, as would be expected.  I have
found the sigma calculation "good enough" and considerably eases
calculations on Excel spreadsheet where the sigma formula is built in, but
the mov R is not.


Steve Prevette
QA Engineer, ESH Radiological Compliance
Fluor Hanford, A Fluor Global Services Company
ASQ Certified Quality Engineer
steven_s_prevette@rl.gov
509-373-9371