Issue: Volume 5, Number 5
Date: May 2005
From: Mark J. Anderson, Stat-Ease, Inc. (http://www.statease.com)

Dear Experimenter,

Here's another set of frequently asked questions (FAQs) about doing design of experiments (DOE), plus alerts to timely information and free software updates. If you missed previous DOE FAQ Alerts, please click on the links at the bottom of this page. If you have a question that needs answering, click the Search tab and enter the key words. This finds not only answers from previous Alerts, but also other documents posted to the Stat-Ease web site.

Feel free to forward this newsletter to your colleagues. They can subscribe by going to http://www.statease.com/doealertreg.html. If this newsletter prompts you to ask your own questions about DOE, please address them via mail to: [email protected].

Here's an appetizer to get this Alert off to a good start—a commentary about quantity versus quality from Josh Wolfe in his April 8th Nanotech Weekly Insider (http://www.forbeswolfe.com). "An art teacher split her class into two equal groups. Both would work on making ceramics. Group A would be graded only on how many pieces they produced: the quantity of their work. Group B would be graded only on how good their finished products were: the quality. The teacher would base her decision on Group A (the quantity people) by weighing their final output on a scale. 100 pounds of pottery would get an A, 90 pounds would get a B and so on. Group B only had to make one pot, but it had to be flawlessly perfect for an A...Surprisingly, the best quality pottery came from Group A, the group focused only on quantity. While they were all hammering away spitting out piece after piece—they were unconsciously learning from their mistakes and improving their quality. Meanwhile Group B, focused on quality, sat around discussing how best to perfect the design, and ultimately had nothing but a pile of clay to show." Put that in your pot and stew on it!

Here's what I cover in the body text of this DOE FAQ Alert (topics that delve into statistical detail are designated "Expert"):

1. FAQ: Why solutions vary from numerical optimization performed by Design-Expert® software
2. Info alert: Link to several new DOE articles and textbooks
3. Reader reply: A suggestion to uniformly disperse center points
4. Events alert: Come see us at the ASQ Quality & Improvement Conference in Seattle and a conference on coatings in New Orleans
5. Workshop alert: See when and where to learn about DOE

PS. Quote for the month: Keeping it simple and stupid

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

1. FAQ: Why solutions vary from numerical optimization performed by Design-Expert® software

-----Original Question-----
From: United Kingdom

"Why is it that I get a slightly different solution set each time I perform numeric optimization, even if I don't change any of my goals or criteria for desirability?"

Answer (from Stat-Ease Consultant Shari Kraber):
"Design-Expert starts its search for the most desirable solutions from a number of random points within the experimental region. Picture this: An airplane flies by night over mountainous territory where it drops 10 paratroopers. Unfortunately it is very windy, so the troopers are scattered randomly. Each has been given the same mission: Climb to the nearest high ground. At dawn, some troopers find themselves grouped on the same peak, but others stand alone on localized high points. On plateaus a number of the troopers are found very near each other, but not at the exact same locations. As you can imagine, each time you repeat this process, the results vary somewhat.

Design-Expert version 6 allows users to increase the number of random searches from the default of 10 to a maximum of 99. (Version 7 will increase the allowable searches to 999.) The program also provides a duplicate solution filter for which "Epsilon," the difference between two results considered to be the same, can be adjusted somewhat. As it's increased, the number of unique solutions decreases and vice-versa."

(Learn more about optimization by attending the three-day computer-intensive workshop "Response Surface Methods for Process Optimization." See http://www.statease.com/clas_rsm.html for a complete description. Link from this page to the course outline and schedule. Then, if you like, enroll online.)

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

2. Info alert: Link to several new DOE articles and textbooks

Read "Experimental Design Optimizes Assay Automation" posted on the Internet by "Inside Biomed" (a Quality Digest publication) at http://makeashorterlink.com/?F51552AFA to learn how an innovative blend of hardware, software and the right training in statistical know-how simplified research automation. The authors are Thomas Erbach and Lisa Fan from Beckman Coulter’s Biomedical Research Division, and Stat-Ease consultant Shari Kraber.

The Online Exclusive for the April issue edition of "Solutions!" magazine features an article by Richard Burnham on how Don Smith of Stora Enso brought in DOE training to banish their "Purple- Paper Blues." It is available for viewing at http://www.tappi.org/index.asp?pid=32071&ch=1.

On March 28 I received a press release announcing publication of "Experimental Design for Formulation" (ASA-SIAM Series on Statistics and Applied Probability 15) by Wendell F. Smith—a frequent contributor to Stat-Ease. Wendell's new book garnered this rave review from fellow author John A. Cornell, Professor Emeritus of Statistics, University of Florida (and reigning guru of mixture design): "...heartiest congratulations on a job well done. I am very impressed by the amount of discussion in the text devoted to the topics of model building and model evaluation and I hope the software companies that support the fitting of mixture models will incorporate the topics discussed in the text in their software." I can assure Professor Cornell that the upcoming new version 7 of Design-Expert software from Stat-Ease will indeed incorporate many new features instigated by Wendell Smith. His book can be ordered direct at http://ec-securehost.com/SIAM/SA15.html. It's intended for formulators in industry as well as senior undergraduate and beginning graduate students in statistics. However, as noted in the PR, be forewarned that before reading this book you'd best familiarize yourself with basic experimental designs and methods
for modeling and interpreting data. Wendell does not shy away from mathematics!

Also, I should mention that Doug Montgomery, Professor of Engineering at Arizona State University, has published his 6th Edition of "Design and Analysis of Experiments." He showed it to me in February at the American Society of Quality (ASQ) Six Sigma Conference. The most obvious change is the cover — it now fits into a new scheme by the publisher that ties together the whole series of statistical books authored in whole or part by Professor Montgomery (see http://makeashorterlink.com/?G25B53EFA). As a
companion to "Design and Analysis of Experiments," the publisher (Wiley, New York) sells an education version of Design-Expert at http://www.wiley.com/WileyCDA/WileyTitle/productCd-0471394114.html.

PS. For a great discussion on Taguchi versus more modern approaches to DOE, see Montgomery's supplement to Chapter 12 of his book at http://makeashorterlink.com/?J62B42EFA.

(For a top-shelf selection of books on DOE, including the new one from Montgomery, see http://www.statease.com/prodbook.html.)

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

3. Reader reply: A Suggestion to uniformly disperse center points

From: P. B. Dhanish, Lecturer, Department of Mechanical Engineering National Institute of Technology Calicut, India
Re: FAQ #2 in DOE FAQ Alert, Volume 5, Number 4—April, 2005

"I would like to point out another potential benefit of having center points in your design. If you arrange the center points in the run order such that they occur in the start, end and balance them uniformly distributed in between, they will give a better check on the stability of the process than what you would
learn from a diagnostic plot of residual-versus-run-order alone. Especially in new processes, learning effects and unknown factors can cause the process to drift, and the center point drift is not aliased with any other factor you have taken into account.

For more discussion, please see "Adding centerpoints" in the Engineering Statistics Handbook at http://www.itl.nist.gov/div898/handbook/pri/section3/pri337.htm."

I appreciate the insights and link from Professor Dhanish.
—Mark

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

4. Events alert: Come see us at the ASQ Quality & Improvement Conference in
Seattle and a conference on coatings in New Orleans

Come see Stat-Ease in Booth #414 at the ASQ World Conference on Quality & Improvement in Seattle, Washington, on May 16-18. Attend talk T205 titled "PCR Process Optimized via Split-Plot Design" by Pat
Whitcomb on Tuesday, May 17th. See http://wcqi.asq.org/ for details on the conference.

On May 19th in New Orleans, I present "How to Design and Analyze Mixture Designs that Include Process Factors and/or Categorical Variables" at a conference on Research Methods in the 21st Century sponsored by the Federation of Societies for Coatings Technology. For more details, see http://makeashorterlink.com/?E34663FFA.

Click http://www.statease.com/events.html for the complete list of appearances by Stat-Ease professionals. We hope to see you sometime in the near future!

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

5. Workshop alert: See when and where to learn about DOE

See http://www.statease.com/clas_pub.html for schedule and site information on all Stat-Ease workshops open to the public. To enroll, click the "register online" link on our web site or call Stat-Ease at 1.612.378.9449. If spots remain available, bring along several colleagues and take advantage of quantity discounts in tuition, or consider bringing in an expert from Stat-Ease to teach a private class at your site. Call us to get a quote.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

I hope you learned something from this issue. Address your general questions and comments to me at: [email protected].

Sincerely,

Mark

Mark J. Anderson, PE, CQE
Principal, Stat-Ease, Inc. (http://www.statease.com)
2021 East Hennepin Avenue, Suite 480
Minneapolis, Minnesota 55413 USA

PS. Quote for the month—Keeping it simple and stupid:

"For every complex question there is a simple answer, and it is wrong."
—H. L. Mencken

Trademarks: Design-Ease, Design-Expert and Stat-Ease are registered trademarks of Stat-Ease, Inc.

Acknowledgements to contributors:
—Students of Stat-Ease training and users of Stat-Ease software
—Fellow Stat-Ease consultants Pat Whitcomb and Shari Kraber (see http://www.statease.com/consult.html for resumes)
—Statistical advisor to Stat-Ease: Dr. Gary Oehlert (http://www.statease.com/garyoehl.html)
—Stat-Ease programmers, especially Tryg Helseth (http://www.statease.com/pgmstaff.html)
—Heidi Hansel, Stat-Ease marketing director, and all the remaining staff

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Interested in previous FAQ DOE Alert e-mail newsletters?
To view a past issue, choose it below.

#1 Mar 01
, #2 Apr 01, #3 May 01, #4 Jun 01, #5 Jul 01 , #6 Aug 01, #7 Sep 01, #8 Oct 01, #9 Nov 01, #10 Dec 01, #2-1 Jan 02, #2-2 Feb 02, #2-3 Mar 02, #2-4 Apr 02, #2-5 May 02, #2-6 Jun 02, #2-7 Jul 02, #2-8 Aug 02, #2-9 Sep 02, #2-10 Oct 02, #2-11 Nov 02, #2-12 Dec 02, #3-1 Jan 03, #3-2 Feb 03, #3-3 Mar 03, #3-4 Apr 03, #3-5 May 03, #3-6 Jun 03, #3-7 Jul 03, #3-8 Aug 03, #3-9 Sep 03 #3-10 Oct 03, #3-11 Nov 03, #3-12 Dec 03, #4-1 Jan 04, #4-2 Feb 04, #4-3 Mar 04, #4-4 Apr 04, #4-5 May 04, #4-6 Jun 04, #4-7 Jul 04, #4-8 Aug 04, #4-9 Sep 04, #4-10 Oct 04, #4-11 Nov 04, #4-12 Dec 04, #5-1 Jan 05, #5-2 Feb 05, #5-3 Mar 05, #5-4 Apr 05, #5-5 May 05 (see above)

Click here to add your name to the FAQ DOE Alert newsletter list server.

Statistics Made Easy™

DOE FAQ Alert ©2005 Stat-Ease, Inc.
All rights reserved.

 


Software      Training      Consulting      Publications      Order Online      Contact Us       Search

Stat-Ease, Inc.
2021 E. Hennepin Avenue, Ste 480
Minneapolis, MN 55413-2726
e-mail: info@statease.com
p: 612.378.9449, f: 612.378.2152