RE: SPC Theory question

[Steven_S_Prevette@rl.gov]

Personally, I ignore the np chart.  The np chart gives the same statistical
result as a p chart, just the vertical axis is skewed to a different scale.

So, leaving the p, c, and u charts as the "special" candidates, we have

The p chart is used for percentages (or proportions).  Things equivalent to
flipping a coin.  Heads or Tails, Go or No go.

The c chart is used for counting.  Counting poisson events per a fixed time
interval, fixed area of opportunity each time interval.  Note the poisson
and binomial are related, as I can count the number of "heads" received per
so many times the coin is flipped, assuming the coin is flipped a (fixed)
large number of times, and the chance of heads is low on each flip.

The u chart is used for counting items per an area of opportunity.  Counting
poisson events per 200,000 hours worked for example.

See also http://www.hanford.gov/safety/vpp/spc.htm, first page.

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