FW: Detect Vs. Predict

["Prevette, Steven S" <Steven_S_Prevette@rl.gov>]

The comment has been made about control charting, and detection and
predication. 

Yes, I would agree that both are important roles of the control chart.
I can take past data and detect if there are any trends, separating
common cause and special cause variation.  If the data are stable, I can
use the average and control limits as a prediction.  Even if it's not
stable, that provides some information for future prediction (even
confirming that I can't make a prediction).

Also, the control chart becomes a way to tell if you are having any
effect on the process.  If you are stable at an unacceptable level or
variation, you can use the control chart to determine if the change you
make has any effect, or if the process is continuing along unaffected.


Steve Prevette
Occupational Safety and Health
Fluor Hanford, A Fluor Government Group Project ASQ Certified Quality
Engineer steven_s_prevette@rl.gov
509-373-9371