SPC (Statistical Process Control) Overview

Steve Prevette

Deming Disciple
Leader
Super Moderator
This presentation gives a little more detail on the nuts and bolts of SPC. This was made for my monthly performance indicators session here at Hanford. I alternate months with a classroom topic one month, and a hands-on computer topic the next.

This presentation is provided for peer review and comment.
 

Attachments

  • SPC_June2005.ppt
    317.5 KB · Views: 6,057

Wes Bucey

Prophet of Profit
Steve Prevette said:
This presentation gives a little more detail on the nuts and bolts of SPC. This was made for my monthly performance indicators session here at Hanford. I alternate months with a classroom topic one month, and a hands-on computer topic the next.

This presentation is provided for peer review and comment.
Excellent non-technical presentation. Even the statistically-challenged can see the process is not "smoke and mirrors" - that it is a simple and straightforward process that returns more than its cost as value to the organization.

Thanks for sharing.
 

Jen Kirley

Quality and Auditing Expert
Leader
Admin
This is an excellent overview of a tool that befuddles or intimidates many into careless neglect of the procedure. I especially enjoyed likening it to a smoke detector (feel safer?) and the visuals/examples were excellent. Nice work Steve, thank you!

Jennifer (non-statician) Kirley
 

Govind

Super Moderator
Leader
Super Moderator
Steve,
I find the presentation very useful. A good refresher as well.
I am very interested in the slides 19 to 21.
"The existing average and control limits should NOT be changed unless a significant trend is detected."
I would appreciate if you can explain with an additional bullet point as to how you propose to detect the significant change in average?
Thanks,
Govind.
 

Jim Wynne

Leader
Admin
Govind said:
Steve,
I find the presentation very useful. A good refresher as well.
I am very interested in the slides 19 to 21.
"The existing average and control limits should NOT be changed unless a significant trend is detected."
I would appreciate if you can explain with an additional bullet point as to how you propose to detect the significant change in average?
Thanks,
Govind.
Not answering for Steve, of course, but changes in the process mean are always the result of a significant change in the process; this would show up in a histogram as a bimodal distribution. The problem is that if you wait for this kind of thing to show up on a chart, it's too late. The type of change I'm talking about is often deliberate--a tooling change is a good example. In such cases, it's advisable to allow the process to stabilize and then recalculate control limits. Steve's point about not changing the average is perhaps misworded; the average changes when it changes.
 

Steve Prevette

Deming Disciple
Leader
Super Moderator
On the issue of changing the baseline average and control limits - Davis Ballestracci gave me a great quote once - the baseline is innocent until proven guilty. You do not adjust the baseline until there is a significant change or trend in the data. And the signal for a significant change or trend is tripping one of the SPC trend rules, such as 7 points above average. Now, sometimes the data return to the previous baseline after corrective actions are taken, and in that case the baseline would remain the same. But if the data steady out at a new baseline average (or standard deviation), then the baseline would be changed.

Have I posted "The Life Cycle of a Trend" here? I wrote that up with an illustrative example a few years back.
 
Q

qualityboi

I liked the presentation, it gives a good overview of SPC without going into the how the UCL and LCL are derived which is a sleeper for non quality folks. In your discussion you may want to mention the whole idea of decision making is making those decisions by fact, not by feel. Good job!
 
Q

qualityboi

I understand that the center line average is recalculated from the process data, however, how would you know if your process was out of control if you keep on readjusting the center line? We had an SPC specialist that had set up a control chart for our out of box quality defects. The center line and control limits kept changing with the data, the chart was useless to me. Maybe I missed something? Is it here that was posted that the center line is innocent until proven guilty?
 

Steve Prevette

Deming Disciple
Leader
Super Moderator
I am going to attach here "The Life Cycle of a Trend". That will help explain what is happening.

The idea is the EITHER a circle, or a shift in the baseline is indication of a shift in the process.

If the center line is proven guilty, your prediction of the future based upon that center line and control limits is no longer useful. You must revise the center line and control limits so that your prediction of the future is more accurate.

The Life Cycle of a Trend
 
Last edited by a moderator:
Top Bottom