An FYI thread:

--> From: Guy Rodgers - Subject: RE: EVOP -
--> Evolutionary Operation of Processes
--> Any experience/Mullins/Rodgers
-->
--> Mullins wrote; "Does anyone have any information on a course
--> called "Evolutionary Operation of Processes (EVOP)" which is
--> referenced in QS-9000, section II.2- Continuous Improvement?"
-->
--> The requirement for EVOP knowledge in Section 2.3 Techniques
--> for Continuous Improvement has been removed from the 3rd
--> edition Quality System Requirements of QS-9000. The section
--> has been moved to Element 4.2 Quality System under paragraph
--> 4.2.5.3.
-->
--> Other techniques eliminated from the requirement include;
--> Capability Indices (Cp,Cpk), Cost of Quality, and Problem
--> Solving. Finally, a note for interpretation was added to
--> clarify that the list only shows examples and other methods,
--> which meet the suppliers needs, may be used.
-->
--> To further explain, EVOP is not a course, but rather a
--> process or technique of systematic experimentation.
--> Evolutionary Operations (EVOP) is based on the understanding
--> that every production lot has the ability to contribute
--> valuable information on the effect of process variables on a
--> particular product characteristic or feature.
-->
--> Typical methods used involve structured designs of
--> experiments (DOE) which may result in interrupting production
--> flow to conduct the trials or experiments. EVOP, on the other
--> hand, is intended to introduce small changes in the process
--> variables during normal production flow. These changes are
--> not large enough to result in non-conforming product, but are
--> significant enough to determine the optimum process ranges.
-->
--> Best regards,
-->
--> Guy Rodgers, QA Administrator Paulo Products Company

Subject: RE: EVOP - Any experience/Mullins/Brauer

Sheila Mullins asks -- Does anyone have any information on a course called "Evolutionary Operation of Processes (EVOP)" which is referenced in QS-9000, section II.2- Continuous Improvement?

Reply -------------------------

While I cannot recommend a "course", you will find information on EVOP in statistics books on experimental designs. Look for one that has Response Surface Methodology in it. Here's a brief description of EVOP from one such book --

EVOP consists of systematically introducing small changes in the levels of operating variables, usually 2^k factorial designs, such that serious disturbances in yield, quality or quantity will not occur, yet large enough that potential improvements will eventually be discovered.

-- Design and Analysis of Experiments, 3rd Edition
-- by D.C. Montgomery, 1991, John Wiley & Sons, Inc.
-- ISBN 0-471-52000-4

Also, you should note that in QS 3rd Edition, Continuous Improvement is now 4.2.5 and EVOP was deleted from the examples list. However, I believe it would still be acceptable to use EVOP since it is a type of DOE (Design of Experiments) which remains on the examples list.

Hope this helps.

Les B.

>> "Greg House"
>> wrote in message
>> news:3bcc8c5c$0$227$cc9e4d1f...
>>
>> I have heard of most things in QA and CIP, but what is EVOP?
*****************

EVOP stands for "EVolutionary OPeration". Basically, it is a process (mainly manufacturing) improvement method developed by Box in de '50s. In this method, the process control variables (e.g. machine speed, temp, etc.) are continuously and systematically modified, the effect of these perturbations on the process output is monitored and the conditions when output is improved are noted. IMHO a combination of experimental design and SPC.

Jean-Philippe Hubin
*******************

"LamarJL" wrote in message news:JoJz7.12110$B82.3776493699...
> Howdy Greg,
>
> EVOP stands for EVolutionary OPerations. The concept was also commonly
> referred to as the "string of pearls" approach to finding the sweet spot.
> Using EVOP you make a change to your process, if the change was good then
> you keep going in that direction. If the change was bad you backed up and
> went the other way. Like the myth of biological evolution - it would take
> forever to find the best spot for each process parameter and you would
> completely miss any interactions between parameters. EVOP gave way to DOE
> and Taguchi style experimentation years ago.
>
> Regards,
>
> Jim Lamar, CQE, CQA, QS-LA, MQM, ABC, Do-Re-Mi, 1-2-3, yada, yada, yada,

EVOP is a special application of DOE so it can't be said to replace it. It is often a 2 level, 2 factor factorial.

The whole idea is that it is done with a running process so you have to keep the changes to the independent variables small so the the output variable stays in spec.

You can't optimize a full-scale production process except by running it, unless in the unlikely case that you know in advance that a small scale or pilot run exactly represents full scale production.

It can certainly take a while to run enough replications to extract the significance of the variables and their optima, but what's the alternative? You can certainly find the interactions. I know there are at least 3 approaches to analyzing the data. One method uses ranking but I have never used it. Maybe that doesn't show the interactions.

Regards,

John Duffus
###########
Some old EVOP messages are Here (http://Elsmar.com/pdf_files/evop/) in case anyone is interested.

Anyone here doing any EVOP experiments?

From: Suzanne Fiorino
Subject: RE: EVOP - Any experience/Mullins/Bigelow/Fiorino

One can think of EVOP as 'baby steps" experimentation. Typically a small full factorial design is used to set up the EVOP test and the intent is to get meaningful information while still producing saleable product. QualPro in Knoxville, TN, has an excellent experimental design course, very practical, and a section of it covers EVOP. Also George Box, who is at the University of Wisconsin, may have a course on EVOP.

From: David Thibault
Subject: Re: EVOP - Any experience/Mullins/Thibault

One point not mentioned so far: Since the changes for EVOP are small relative to those for DOE, EVOP must rely on many replications to separate the signal from the noise.

Date: Sun, 23 May 1999 12:00:56 -0400
Subject: Electronic reprint of textbook about quality improvement using simplex EVOP

The textbook "Sequential Simplex Optimization: A technique for Improving Quality and Productivity in Research, Development, and Manufacturing" by F. H. Walters et al has been out of print for some time. CRC Press LLC has now licensed us t he right to make electronic reprints of this excellent text. We wish to continu e the work to popularize the simplex EVOP methodology, and make it an everyday tool for many more professionals and researchers. We have therefore decided to publish and distribute the electronic reprint through our web site. You may fre ely browse the whole text as HTML or download it as an Adobe Acrobat PDF-file from Grabitech.com

Best regards,
Tomas Oberg
MultiSimplex AB

And you can always check out the always faithful "Juran Handbook".

You know - I never looked in Juran's tome before.... Now I see it is discussed in some length.

Picking it up (literally) is an experience you will probably never forget!!:biglaugh: Darned thing is big and heavy!!

Definition of tome
*
Part of Speech * noun

Definition 1. a large thick book, often one of a multivolume scholarly work.

I'll admit I don't see a weight specification, and I admit it's not technically multi-volume (note the qualifying word often in the definition which preceeds "...one of a multivolume..."), but I sorta was alluding to its heft - so to speak - calling it a tome. (Al: Actually I was just showing off my great vocabulary having been raised on Scrabble...)

I can't state it's bias, linearity, calibration to a reference standard or anything else fancy, but my bathroom scale indicates about 6 to 6.5 pounds (maybe one of our English compatriots here will translate that into stones for us).

One of you folks have a calibrated scale close by - How much DOES the book weigh? Do let us know which revision level it is... And the uncertainty of the scale... Etc...

:topic: A Scrabble Master - I am in awe!!!

On my SAT, I ranked in the 37 percentile in vocabulary. I was much better at Trivial Pursuit!!:biglaugh:

Kev

:topic: My mother and little sister are the Scrabble Masters... They still play. It's their common ground, if you will.

Is anyone currently into EVOP (Evolutionary Operation of Processes)?

I see a lot of various discussions here - but I don't see much discussion of methods or tools to look at feedback from proceses in what I consider process 'monitoring' and evolution - Fine tuning and predictive methods.

I guess what I'm saying is: Your process(es) is(are) in control. Now what?

My vote was for the category the "most closely" matched my experience. The first part, "we", if meaning my present employer, is incorrect. I used it years ago when in the metal stamping business. Several applications were successful back then.

I voted that we embrace it and have reliable data...., but only because EVOP is pretty much our process for Design of Experiments (DOE). We, typically, don't do DOEs, per se. Our process engineers seem comfortable tweaking only a couple of parameters at a time, and feeling they have the issue or improvement taken care of :nope:

Bill

My first co-op job way back in the early 80’s was at a mine. I had an entry level job tending an on stream process analyzer. I had no idea what the thing did, so after a few weeks I followed the wires up to a control room.

The guys there said the whole mineral processing plant was run by computer. Teletypes and PDP-11s! My device sent information on the content of Lead and Zinc, and then the computer did EVOP to improve the yield.
They gave me the best description ever of how EVOP works.


If you were blindfolded and placed on the side of a hill, how could you most quickly get to the top? What you would do is plant you left foot (the origin) and step with right foot four times at 90 degrees. One direction would be up (increased yield). Now move your left foot to the new highest point and test with your right foot again. You only need to do this 3 times since you came from the low point. Continue on until you get to the top of the hill.

You are mapping out the hill topology or response surface.

The fun becomes when you apply it to more than one variable. It still works, but can no longer be easily described in words.

They had data that showed that starting Friday night and ending Monday morning the EVOP computer system continually improved things. When the process engineers got to work Monday morning “improving things” yields got worse.

Being a dumb student at the time I only now realize what an incredible thing they had set up.

I saw EVOP again 15 years later at a die cast shop. They ran 0% scrap, had tapped out SPC, (Cpks > 20 on key process parameters) and now had the operators running EVOP to decrease cycle times.

A very cool tool, one I hope to use some day.

So here it is 7 years later and my first exposure to this poll and I STILL have "never heard" of EVOP, despite the fact I once held QS 9000 status for my organization. The point being there are some terms and phrases, which once heard, keep passing through until they come out the other ear and then are never heard of again, mercifully and rightfully forgotten, regardless whether folks performed some process similar to the one described.

It must have been the fact that the phrase was so "contrived" that it just didn't stick. Perhaps if they had used the term "continual improvement" - it might have had better "stickiness."

:topic: The allusion to "string of pearls" brought a smirk - my memory having to do with one of the ways a string of pearls was purported to be used in bawdy houses in my youth.

My comment would be the same as post #10. I haven't used it since, but I always look for opportunities to use the right tool (in my present business it just isn't that useful).

But I did like the technique: EVOP, or Response Surface Methodology (RSM) does have its place, and would probably be used more if it was taught more.

If you were blindfolded and placed on the side of a hill, how could you most quickly get to the top? What you would do is plant you left foot (the origin) and step with right foot four times at 90 degrees. One direction would be up (increased yield). Now move your left foot to the new highest point and test with your right foot again. You only need to do this 3 times since you came from the low point. Continue on until you get to the top of the hill.


There are several other variations on this theme that can sometimes be more efficient.

1) Instead of heading in the direction of the single largest increase, head somewhere between the two best steps. For example, if a step to the north took you 4" up and a step east took you 1" up (and S & W took you downhill), then head NNE rather than due N. That is closer to the true uphill direction than either N or E.

2) Instead of stopping after every step and rechecking which way is up, just keep walking forward. After the first step, you may not be heading straight uphill, but you will still be close. Once you reach a point where you start down, then take the four steps NSE & W to determine the uphill direction and head off once again in the new uphill direction. You take more steps actually travelling toward the top, but you take fewer test steps along the way.

One challenge with any such method is if there is a "valley" between you and the top of the highest hill. You will reach the top of the closest hill, but you will get stuck there and never find the highest peak.

Tim F