SPC (Statistical Process Control) Overview

Jim Wynne

Leader
Admin
Steve Prevette said:
But first - Do not calculate new average and control limits unless the existing one has been proven guilty by the data! Make sure your baseline has not become some sort of moving average - it needs to stay fixed in time, even if that was three years ago.

Con - People can get over anxious to rebaseline. And, in the timeframe just following a trend, you don't have much data to make a new baseline from. This may also be a production line where you want to take corrective action to get the data back to the old baseline. Dr. Wheeler tends to default to don't change the baseline unless you have both a data shift, and you know why the data shift occurred.

Pro - "The job of management is prediction" (Deming). The baseline average and control limits provide prediction. If the current baseline has been proven no good, it is no longer a good prediction. We want to detect when the data stabilize out again, and set up new predictions.

In either case, SPC is a remarkable self-healing, self-correcting process. If you shift the baseline too early, on a false alarm, the new baseline will be proven guilty and you will be back on the old baseline.
I don't think we disagree on this. And permit me to reemphasize the thing about prediction. If you're doing inferential statistics and you can't make good predictions, something is wrong.
 
T

tristan

i think there is an interpretation cut short here in terms of using control charts. re-calculating baselines must also be linked to final product performace, reliability, and customer specifications. any process change leading to baseline change must be checked against these. by doing these, the changing baselines and control limits do tell us 2 things: feedback to current process & confidence in how the process will perform in the future.
 

Jim Wynne

Leader
Admin
good info.
want more information regarding SPC.

Welcome to the Cove, Nehal :D

This place is loaded with information on SPC. Do a search of the Cove and you'll find lots of stuff, and if you have a specific question, someone here will be glad to help.
 
A

artichoke

Steve,

Wonderful to see that you are one of the very few people who seem to be aware of the 99.7% control chart myth. This myth appears to have been propagated by popular authors such as Montgomery and has led to much misunderstanding.
 
Last edited by a moderator:

Steve Prevette

Deming Disciple
Leader
Super Moderator
NEW!

Based upon a request originally by Bechtel, but also supported by my employer Fluor, and other Department of Energy contractors here at Hanford, I've accumulated a lot of my papers into a two-day statistical training session. This covers SPC, and also choosing indictors, Dr. Deming's Red Bead Experiment and System of Profound Knowledge, some hands-on exercises, and some computer exercises. Fluor has given me permission to make the materials available on the internet, and the URL is

https://www.hanford.gov/rl/?page=1156&parent=1144

Since the materials were generated under US Government contract, there is not a copyright protection. You are welcome to make use of the materials if you find them useful. I would, of course, appreciate a mention as the source of the materials if you use them.

{Semi-Advert} Of course, if you'd like me to come make the training for your group, that could be arranged.
 
S

shank

I am getting confused between stability and capabitlity. Can someone list out the attributes to capability and stability separately?
 

Jim Wynne

Leader
Admin
I am getting confused between stability and capabitlity. Can someone list out the attributes to capability and stability separately?

Welcome to the Cove, shank :D

It's pretty simple. "Stability" refers to a state of statistical control, meaning that the process is subject only to inherent random variation. "Capability," on the other hand, has to do with how likely the process is to produce output meeting the specification limits. A process may be stable without being capable.
 
Top Bottom