RE: PLEASE HELP

[Steven_S_Prevette@rl.gov]

I will make an initial try at
these definitions.  On those items asterisked (*), I have a fairly detailed
explanation on my web site, the Hanford Trending Primer at
http://www.hanford.gov/safety/vpp/trend.htm  I will attempt to define these
terms in the context of Dr. Deming's Out of the Crisis and The New
Economics.

* run-chart - A chart which plots the data in a time sequence.  Each point
is plotted as a raw value, there are no moving averages, cumulative year to
date, cumulative averaging, et al.

* control-chart - A chart which adds an average (center) line, an Upper
Control Limit and a Lower Control Limit to the run chart.
http://www.hanford.gov/safety/vpp/spc.htm provides a good introduction, and
documents some bases for calculation of control limits.

*c-chart - A control chart which plots "counts" data.  Usually this is
number of defects or events.  The poisson distribution is used to calculate
the control charts.  See http://www.hanford.gov/safety/vpp/cchart.htm for an
example.

*p-chart - A control chart which plots "proportion" or "percentage" data.
The binomial distribution is used to calculate the control charts.  See
http://www.hanford.gov/safety/vpp/pchart.htm for an example.

s-chart - A control chart which plots standard deviation data.  This is used
in conjunction with the x-chart to see if the variability of the data is
changing.  However, I have not found it very practical to explain and have
personally been able to do without.

*u-chart - A control chart which plots "counts" data per an area of
opportunity.  A classic example is injuries per 200,000 hours worked.  The
poisson distribution is used to calculate the control charts.  See
http://www.hanford.gov/safety/vpp/uchart.htm for an example.

*x-chart - A generic control chart, where the standard deviation of the data
(calculated either by moving range or the statistical standard deviation -
the choice of which has already been the subject of lengthy discussions on
the DEN) is used as a basis for the control limits.  See
http://www.hanford.gov/safety/vpp/xchart.htm for an example.

Cp and Cpk are "process capability" calculations.  They compare the spread
of the upper and lower control limits to the spread of the specifications
for the item being measured.  Since Dr. Deming was not in favor of numerical
targets, nor of hypothesis testing (which this is closely related to), I
will not go into detail.

In - Individuals?

Ind X/Mr, Xmr, X-movR - Individual X and Moving Range.  A pair of control
charts.  The Ind X chart is based upon the run chart and is the same as the
x-chart above, but is strictly limited to using the range to calculate the
control limits.  Rbar/d2 is one of the formulae used to determine control
limits.  The Moving Range chart plots the absolute value of the change from
point to point.  Dr. Wheeler's book "Understanding Variation" is a terrific
introduction to this series of charts.

Xbar - X bar is either the average of the data, or is a chart.  In the chart
context, instead of plotting individual X's, subgroups are plotted (usually
of size 4 or 5).  The subgroup average is plotted, and the range (maximum of
the points minus the minimum of the points) is plotted.  This is the
original concept Dr. Shewhart proposed in Economic Control of Quality of
Manufactured Product.  

Rbar/d2 - see explanation for IndX/Mr

R/xbar - R is range, and xbar is the average, but I don't know of any use in
doing this division.

Sampling Plan-1: n=13; c=0 - Probably refers to a sampling plan where 13
items are chosen at random, and the lot is rejected at the first detected
failure of the 13.  The acceptable number of failures is zero.  Dr. Deming
was generally suspicious of random sampling, see Out of the Crisis.

S/Xmr - S usually refers to standard deviation, and Xmr is the moving range.
These are two alternative methods for determining control limits.

I hope this is of assistance.  To fully answer your questions would require
writing a whole 'nother stats book.  I would recommend starting with Dr.
Wheeler's works.  Personally, I rely upon Acheson Duncan's "Quality Control
and Industrial Statistics", but this is probably not a good beginning book
to start with.  There are also plenty of qualified people out there that
offer training in the subject.

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