RE: Periodic Control Charts

[Steven_S_Prevette@rl.gov]

I sent Jan some examples that I had run across in analyzing energy usage at
Hanford.

You can do a reasonable seasonal or cyclic analysis by accumulating all of
the Mondays together, determining that average and standard deviation for
Mondays, and subtracting the average and dividing by the standard deviation
for Mondays.  Do the same for each day of the week.  Then plot the resulting
"normalized" data in date sequence.  It should have an average of zero and a
standard deviation of 1.  If not, then there are other special causes to go
looking for.

I also did a similar analysis years ago looking at promotion rates in the
military, with comparisons by race and gender.  Each promotion each year had
a different average rate, but I needed to compare the genders and races
against that rate, and then compare across years.  Doing the same
"normalization" helped to see what was happening.  

This method is supported in Acheson Duncan's Quality Control and Industrial
Statistics, but I loaned out my book so I can't quote the page.

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