RE: Changing Windows

[Steven_S_Prevette@RL.gov]

I have put together 6 charts using excel spreadsheet.  It is saved in a
dual (excel 5 and excel 97) format.  I will forward the file to upon
request (I have already sent it to the originator of the data).

My conclusions from the data are:

The two repairs on 1/3/97 are significantly different than the others.
Many more defects, spread across all categories.  Since inspector A did
these two inspections, his/her results are also different than the
others.  

Inspector A does have more defects found than the others (using a
u-chart), but if you remove the two 1/3/97 inspections, his/her
remaining inspections are statistically in line with the other
inspectors (average number of defects per inspector).

The 12/30/98 to 3/18/98 inspections have a different distribution than
the 1/5/97 to 8/31/98 inspections.  Incorrect Installation, Loose or
Missing Hardware, and Safety appear to be much higher.  This was
determined using a chi-square control chart to find break points in the
distributions, then plotting Pareto charts for each time interval.

After the initial surge in 1/3/97, Documentation does not appear to
dominate as highly.  Incorrect Installation is actually now the leading
defect in recent inspections, with Documentation a close second.

After January 1997, the rate of window replacements have been vey
stable.  There is some indication of increase in defects per window for
Feb 97 to present, it is not quite significant on the c-chart, but does
show as a significant change on the chi-square chart starting in
December 1998.

The methods used are documented at
http://www.hanford.gov/safety/vpp/trend.htm

Steve Prevette
FDH ESH&Q Planning and Evaluation
ASQ Certified Quality Engineer
steven_s_prevette@rl.gov
==================================================================