Window Change Data

[Steven_S_Prevette@RL.gov]

>From Robert A. Ferrell's post with window change data:

How many windows did each inspector inspect during this time period that
had no flaws?

Did any one window have more than one flaw in any of the data?  If so,
can we identify those cases and back out the data to how many flawed
windows there were?

Are the 15 failures in the date sequence that they occurred?  Do you
have the dates?

Proposed methodology to answer your question:

1.  A p-chart by inspector for proportion of windows found with flaws,
to check differences between inspectors.

2.  A chi-square control chart to check the distribution of the flaws
versus time.

3.  A chart to check the time interval between failures (see
http://www.hanford.gov/safety/vpp/lowrate.htm)

4.  A Pareto chart for the different flaw types.

Steve Prevette
FDH ESH&Q Planning and Evaluation
ASQ Certified Quality Engineer
========================================================================