Pro's and con's of using precontrol charts

Home > Statistics Every Writer Should Know > The Stats Board > Discusssion

Pro's and con's of using precontrol charts
Message posted by Enrique (via 12.32.68.100) on July 3, 2001 at 9:41 AM (ET)

We are currently in the "learning" process of using process control charts. I was told by one of our corporate's experts that the precontrol charting method would be easier to use than conventional process control charting. I am not yet familiar with this methodology. What are the pro's and con's of using this precontrol method over SPC? Where can I find reliable literature on this?
Thanks

READERS RESPOND:
(In chronological order. Most recent at the bottom.)

Re: Pro's and con's of using precontrol charts
Message posted by JG (via 128.8.23.44) on July 3, 2001 at 5:02 PM (ET)

What is SPC ? A control chart has to be calibrated first. This may be information about the process that you already have or information collected in the preliminary phase of settin up a control chart. You can use exponential smoothing or cumulative averages to see if the process is stable enough for a control chart to work and to get the information required by a control chart.

Re: Pro's and con's of using precontrol charts
Message posted by Vern Myers (via 12.5.194.71) on July 5, 2001 at 6:10 PM (ET)

Don't waste your time with pre-control. Pre-control mandates adjusting your process based on deviation from specification limits, rather than statistical limits. In the SPC world this is called "tampering" with the process and is a SURE way of increasing overall variation. Go with the basic Shewhart-type control chart. Be wary of some of the books and advice out there on the subject of SPC-- there is a lot of misinformation out there. I recommend the book "Advanced Topics in Statistical Process Control", by Donald Wheeler (SPC Press, Knoxville, Tenn.) It is the best single SPC reference I've ever read, and I've read a lot of them. If you don't already have a fairly good understanding of control charts, also pick up "Understanding Statistical Process Control" by Wheeler and Chambers, same publisher.

Re: Pro's and con's of using precontrol charts
Message posted by Phil (via 216.175.112.5) on July 5, 2001 at 9:37 PM (ET)

I agree that you should stay away from pre-control. To even work as advertised, you have to have a very capable process already. IT IS NOT A SUBSTITUTE FOR SPC.

I agree that Wheeler's stuff is top notch and well written. I believe his stuff is available from SPC Press in Knoxville, TN.

SPC done right is great. Sadly, most of the time it is not used as originally intended.

Re: Pro's and con's of using precontrol charts
Message posted by Darius (via 200.23.217.10) on July 6, 2001 at 5:09 PM (ET)

PreControl is a methodology that has its uses.

One of the followers is Steiner: http://www.stats.uwaterloo.ca/~shsteine/papers/pre.pdf

I read the book "Advanced topics in Statistical Process Control" an I recommend it, but I am disagree about precontrol, I tink it usefull if you don愒 care about reducing variation, can愒 reduce it or the specification limits are too important to miss (control charts don愒 take specifications in account).

I am tink that every chart (Precontrol and SPC) has its uses, I agree the SPC uses are more sublime ( variation reduction), but you have to tink about what you are trying to do, don愒 use precontrol just because is simpler. And if you use SPC you have to look for variation reduction, because if you don愒, SPC will be just a waste of time.

Re: Pro's and con's of using precontrol charts
Message posted by Phil (via 216.175.112.5) on July 6, 2001 at 8:47 PM (ET)

The biggest proponents of Pre-Control Charts in recent history (Dorian Shainin, Kiki Bhote, et al) have promoted Pre-control charts as a substitute for SPC charts--not as a way to set up a machine as originally proposed.

While Shainin did do some great stuff in the area of mechanical problem solving, I think he was off-base in this particular area.

In the few conversations I have had with Pre-Control enthusiasts, the only common element was that they didn't know anything about statistical theory. I am not an elitist, but if you don't even know what a statistical hypothesis is or how to calculate process capability, how can you pass yourself off as an SPC consultant.

Other than for trying to establish a quick set-up for an already capable process, I see no value in Pre-control.

Re: Pro's and con's of using precontrol charts
Message posted by Phil (via 216.175.112.5) on July 6, 2001 at 9:03 PM (ET)

I skimmed the Steiner article. I failed to see the advantage of any of the three versions of Pre-control over SPC. Determining special vs. common cause variation is a critical matter.

I recommend Deming's "Out of the Crisis". I am re-reading it myself after seeing Dan Reid's article on the book in the latest Quality Progress.

Your $5 contribution helps cover part the $500 annual cost of keeping this site online.