Dear Experimenter,
Here’s a fresh set of answers to frequently asked questions (FAQs) about design of experiments (DOE); plus timely alerts for events, publications and software updates. Check it out!
- Mark
Blogs
*(My wry look at all things statistical and/or scientific with an engineering perspective) Also, see the Stat-Ease blog for tips on making DOE easy. For example, a recent posting provides “Four Questions that Define Which DOE is Right for You”. Check it out!
Alerts
Topics in the body text of this DOE FAQ Alert are headlined below (the “Expert” ones, if any, delve into statistical details):
  1. Software Alert: Version 11.1.2 of Design-Expert® software released with enhanced capability to model historical data!
  2. Software Alert #2: Beta Test for Version 12 of Design-Expert softwareyou are invited to join.
  3. FAQ: What is the best way to assess the goodness of fit for a response surface method (RSM) model?
  4. Info alert: Design-Expert's definitive screening designs (DSD) featured in Chemical Engineering.
  5. Book giveaway: Winners of autographed copies of the DOE/RSM/Formulation Simplified trilogy and more!
  6. Events alert: A great line-up of talks and appearances in the first half of 2019—culminating in the (Premier!) Analytics Solutions Conference co-sponsored by Stat-Ease and Camo
  7. Workshop alert: See when and where to learn about DOE—Sign up now before classes fill.
P.S. - Quote for the month: Lab automation taking the tedium out of running experiments so researchers can focus on the design and analysis.
(Page down to the end of this e-zine to enjoy the actual quote.)
Software Alert: Version 11.1.2 of Design-Expert software released with enhanced capability to model historical data
All you current Design-Expert v11 (DX11) users will do well by updating to 11.1.2 via "Help" from within your installed program. View the Read Me file for details on the changes—mainly maintenance items with one notable exception: expanded capability to model historical data.* If you want to receive notice when an update becomes available, go to Edit on the main menu of your program, select Preferences and, within the default General tab, turn on (if not already on by default) the “Check for updates on program start” option. (This alert feature is not available in network versions.)

If you remain mired in an obsolete version or not yet a user of Design-Expert, click here and scroll down to the “Features” tab for what puts the heaven in DX11. Then, assuming you like what you see, scroll back up to the link for the free 30-day trial or buy it directly, taking advantage of upgrade pricing if eligible. You will do well by gaining the leading edge on DOE capability!

*While teaching our Modern DOE workshop on-site at a mining-industry client in Jamaica last fall, I achieved a big success by modeling historical data from their refining operations and uncovering a key factor for reducing losses of valuable ore. This is a powerful feature provided by the latest version of DX11. Check it out with your own data! - Mark
Software Alert #2: Beta Test for Version 12 of Design-Expert—you are invited to join
We are getting Version 12 of Design-Expert ready. How would you like to join the Beta Test? It's free. You get a sneak peek at the new version. There's only one catch: you need to share your thoughts on the new version with us. That's it. If you are interested, you can sign up at https://beta.statease.com.
FAQ: What is the best way to assess the goodness of fit for a response surface method (RSM) model?
Original question from an Aerospace Engineer:
“What is the best way to check the “firmness”/”goodness” of a RSM? Obviously, I could just try a bunch of points in the design space and see how good the RSM predictions are, but is there a method to this madness? I’m sure there’s a handy answer for this question, but I couldn’t readily find it in the DX11 tutorials, and I am only just starting to read the other books I got from you guys with more detailed treatments of RSMs.”

Answer from Stat-Ease Consultant Martin Bezener:
“As long as you have some replicates in your design and more than enough unique points to fit the model,* there will be a "lack of fit" test on the ANOVA. This test essentially does what you propose - it looks at the predictions at all points in the design space and formally tests if the deviations are too large to have occurred by chance. If this test is significant, then there's probably some signal in the data that you're not picking up.

In your case, it turns out than even though the statistical fit (R-squared, etc), is very good, doing a “sweep” across the ranges of one of the variables while holding the other two constant shows that the RSM loses a lot of accuracy at non-DOE points away from the center (i.e. towards the edges design space)—something to keep in mind when deploying your model.”

*Provided with default choices in Design-Expert for standard RSMs (central composite and Box-Behnken) and optimal custom designs.


(Learn more about RSM by attending the three-day computer-intensive workshop on Modern DOE for Process Optimization. Click the title for a description of this class and registration details.)
Info alert: Design-Expert's definitive screening design (DSD) featured in Chemical Engineering
Those of you who work in the process industries can learn about Accelerating Six Sigma Research with the Definitive Screening Design (DSD) Technique in this Chemical Engineering online article. The case studies were set up and analyzed using Design-Expert software tools for DSDs.
Book giveaway: Winners of autographed copies of the DOE/RSM/Formulation Simplified trilogy and more!
(Sorry, due to the high cost of shipping, this offer applied only to residents of the United States and Canada.)

Several dozen readers took me up on my offer to simply reply the November-December DOE FAQ Alert email for a chance at winning a free autographed copy of DOE Simplified, 3rd Edition, RSM Simplified, 2nd Edition or Formulation Simplified, 1st Edition; plus three slightly used (display copies), but still very valuable textbooks: Myers, et al, Response Surface Methodology, 3rd Edition, Cornell’s Experiments with Mixtures, 3rd Edition; and Smith’s Experimental Design for Formulation.

Here are the lucky ones:
  • Bob B., Alpharetta, GA: DOE Simplified
  • Milda E., Hunt Valley, MD: DOE Simplified (bonus winner)
  • Ken H., Erie, PA: RSM Simplified
  • Peter H., Calgary, AB: Formulation Simplified
  • Robert R., Riverside, CA: Myers, et al RSM textbook
  • Chari H., Strongsville, OH: Cornell’s mixture design textbook
  • Maria L., Pasadena, TX: Smith’s mixture design textbook
Congratulations all!
Events alert: A great line-up of talks and appearances in the first half of 2019—culminating in the (Premier!) Analytics Solutions Conference co-sponsored by Stat-Ease and Camo
Stat-Ease Consultant Pat Whitcomb will present “Practical Aspects for Designing Statistically Optimal Experiments” and co-chair the session on Smart Data – Design of Experiments at the 2019 IFPAC (International Foundation Process Analytical Chemistry) conference in North Bethesda, MD (Washington, DC area) on March 3–6. Stat-Ease will be represented there by CAMO Analytics. Please stop by for a visit. For more details on IFPAC 2019, click here.

I will provide a briefing on the Strategy of Mixture Experimentation at the Southeastern Branch Meeting of the Entomological Society of America in Mobile, AL, at a lunchtime learning session on March 4th. I suppose we will be treated with pan fried crickets, bug salad and caramel mealworms for dessert. If this does not put you off entirely, check out these details on this summit meeting of SE USA entomologists.

Stat-Ease Consultant Martin Bezener is lined up for DATA (Defense & Aerospace Test & Analysis) Works 2019 in Springfield, VA, (Washington, DC area) on April 9–11, where he plans to speak on logistic regression of binary data. Details on DATAWorks can be found here.

(2nd Notice) Reserve June 18-20 to enjoy early summer in Minneapolis and discover how to transform your business with data-driven tools from R&D to production at the (Premier!) Analytics Solutions Conference. For more details including the call for speakers, registration and so forth, click here.

Click here for these and other upcoming appearances by Stat-Ease professionals in 2019.

P.S. Do you need a speaker on DOE for a 2019 learning session within your company or technical society at regional, national, or even international levels? If so, contact me. It may not cost you anything if Stat-Ease has a consultant close by, or if a web conference will be suitable. However, for presentations involving travel, we appreciate reimbursement for travel expenses. In any case, it never hurts to ask Stat-Ease for a speaker on this topic.
Workshop alert: See when and where to learn about DOE—Sign up now before classes fill.
You can do no better for quickly advancing your DOE skills than attending a Stat-Ease workshop. In these computer-intensive classes, our expert instructors provide you with a lively and extremely informative series of lectures interspersed by valuable hands-on exercises with one-on-one coaching. Enroll at least 6 weeks prior to the date so your place can be assured—plus get a 10% “early-bird” discount.

   Mixture and Combined Designs for Optimal Formulations (MIXC)
   Mar 19-21, Minneapolis, MN
   May 8-9, Cleveland, OH
   Experiment Design Made Easy (EDME)
   Apr 2-3, Minneapolis, MN
   May 29-30, Orlando, FL
   Modern DOE for Process Optimization (MDOE)
   Apr 2-4, Minneapolis, MN
   May 29-31, Orlando, FL
   Designed Experiments for Medical Devices (DEMD) <=New! 
   May 7-8, Minneapolis, MN

See this web page for complete schedule and site information on all Stat-Ease workshops open to the public. To enroll, scroll down to the workshop of your choice and click on it, or email our Lead Client Specialist Rachel Pollack at [email protected]. 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.*
*Once you achieve a critical mass of about 6 students, it becomes very economical to sponsor an on-site workshop, which is most convenient and effective for your staff. For a quote, e-mail [email protected].
I hope you learned something from this issue. Address your general questions and comments to me at: [email protected].

PLEASE DO NOT SEND ME REQUESTS TO SUBSCRIBE OR UNSUBSCRIBE—
FOLLOW THE INSTRUCTIONS AT THE END OF THIS MESSAGE.

Sincerely,
   Mark

Mark J. Anderson, PE, CQE
Principal, Stat-Ease, Inc.
1300 Godward St NE
Minneapolis, Minnesota 55413 USA
P.S. Quote for the month: Lab automation taking the tedium out of running experiments so researchers can focus on the design and analysis.

“Automation will free up scientists’ time, allowing them to spend much less time on grunt work in the lab and focus on experimental design and data interpretation, the crux of innovation.”
  - "How Automation Will Bring BioPharma Research Into the 21st Century", published Jan 2, 2019 by BioSpace.
Trademarks: Stat-Ease, Design-Expert and Statistics Made Easy are registered trademarks of Stat-Ease, Inc.

Acknowledgements to contributors:
  • Students of Stat-Ease training and users of Stat-Ease software
  • Stat-Ease consultants: Pat Whitcomb, Martin Bezener, and Shari Kraber
  • Stat-Ease programmers: Hank Anderson, Joe Carriere, and Mike Brownson
  • Stat-Ease business staff—Cathy Hickman, Greg Campbell, and Rachel Pollack—who provide such supreme support!
Circulation: Over 6000 worldwide
Subscribe
Copyright ©2020 Stat-Ease, Inc., All rights reserved.

Our mailing address is:
Stat-Ease, Inc.
Broadway Place West
1300 Godward Street NE
Suite 6400
Minneapolis, MN 55413


Want to change how you receive these emails?
You can update your preferences or unsubscribe from this list.