Topics in the body text of this DOE FAQ Alert are headlined below (the "Expert" ones, if any, delve into statistical details):

P.S. Quote for the month: A gem from Karl Pearson.

(Page down to the end of this e-zine to enjoy the actual quote.)

1: E-learning alert: The Stat-Ease Academy is now open for stimulating interactive education on DOE. Try a class for free!

We are pleased to announce the launch of the Stat-Ease Academy. Applying the same delightful recipe that makes our workshops so effective (as noted in the testimonial below), the Stat-Ease Academy e-learning delivers what you need to tool up on design of experiments, all on your schedule at your desk or on your mobile device.
_____________________________________________________________________________________________________

“A more balanced presentation of the practical and theoretical aspects of DOE would be hard to imagine...”
—Ed Warner, Merck & Company
_____________________________________________________________________________________________________

Choose from a variety of courses:

• PreDOE Web-Based Training (Free!): Start here to brush up on the fundamental statistics used in DOE. The class is ideal for technicians, managers and professionals who need the basics or a refresher on them.
• 4 Easy Steps to Effective Factorial Design (Free!): Get a fast start on how to set up experiments that are sure to advance your cause no matter what the outcome. (I recommend this be the class that you sample, it featuring pizzazz to the max. You will not be bored! More importantly, this presentation is chock full of really valuable advice. —Mark)
• Finding the Vital Settings via Factorial Analysis: In a fast-paced and fun course that keeps you engaged all along the way, learn to select effects, interpret the statistics, recognize diagnostic patterns, and showcase your results.
• Easier Experimenting with Factorial Split Plots (EEFSP): Make your experiment far easier by gaining a handle on hard-to-change factors via split-plot designs, which incorporate convenient restrictions on randomization. Real-world systems never make it easy to change every factor—some are bound to hard. Therefore, the content of EEFSP e-learning is a “need to know” for all experimenters.
• DOE Simplified Launchpad: Bring the DOE Simplified, 3rd Edition book by Mark Anderson and Pat Whitcomb to life with this narrated presentation of the first three chapters of the book, including hands-on exercises.

2: FAQ: The peril in sorting by a hard-to-change factor without analyzing it properly as a split plot (easily done with Design-Expert® software!)

Original question from a Senior Quality Engineer:“First of all, thank you for the really interesting and helpful Split-Plot Pros and Cons webinar (download slides and/video here). This came just in time for me to design a 4-factor DOE with one hard-to-change factor (materials) at 6 levels and the other 3 factors at two levels. The materials, which are my main interest, will be tested one at a time in a small oven. I can do only one run per day.
If I simply run a completely randomized design (CRD) with the hard-to-change factor sorted for convenience, will the results come out the same? If so, that would avoid complications in the statistical analysis.”

Answer from Stat-Ease Consultant Brooks Henderson:“If you run the CRD and artificially sort the factor levels, your p-values will be incorrect.* Many people sort the factor levels as you suggest, unknowingly creating what is really a split-plot design. Then due to incorrect statistics they end up selecting too many hard-to-change (HTC) factors and too few easy-to-change (ETC) factors. However, the magnitude of the effects come out the same either way (whether analyzed as CRD or split plot). Therefore, by overlooking the statistical tests and picking effects by their importance, i.e, the ones that you deem as important on the basis of their size, then useful results might be salvaged.
So, the two choices are: let it be fully randomized (no sorting) with a normal design and ANOVA analysis.  Or, run a split-plot design where Design-Expert will sort it for you and do the proper REML analysis.”

*P.S. To see an illustration of p values right and wrong , see tables 3 and 4 in this article on How To Analyze A Split-Plot Experiment. In this case it makes a really big difference in the end results.

—Mark

(Learn more about factorial split plots by attending the two-day computer-intensive workshop Experiment Design Made Easy. Click on the title for a description of this class and link from this page to the course outline and schedule. Then, if you like, enroll online.)

3: FAQ: Peculiar differences in designs for screening mixtures

Answer:
The simplex screening designs originally developed by Snee & Marquardt in 1976 (“Screening concepts and designs for experiments with mixtures,” Technometrics, 18, 19–29) lay out 3q+n points, where q is the number of components and n is the number of centroids. They are over built for their purpose of fitting a linear model. That’s why we make the axial check blends, constraint plane centroids and overall centroids optional. Going with none of these optional points leaves the q vertices, i.e., 12 points for your case.

The optimal design also establishes q points at a minimum (equaling the number of linear coefficients in Scheffè polynomial) with optional additional model points (default zero) lack-of-fit (5) replicate (5) and additional center points—aka centroids (default zero). So in your case Design-Expert (DX) recommends 22 runs (=12+5+5).

If runs are dear, I recommend as a compromise going with simplex screening (being built to a template not so mysterious as optimal and thus less daunting to chemists) with the q constraint plane centroids (a bit over the top for those who are geometry-challenged) turned off.  In your case of 12 components that would lead to a 29 component design.

—Mark

More detail from Stat-Ease Consultant Wayne Adams:“The screening design isn’t limited to a simplex. If the experimenter doesn’t change anything, and leaves the total set to 1, and the range for all components 0 to 1, then a simplex is the default. The type of design you get depends on the component entries. This is different behavior than optimal designs. For optimal designs you get the number of runs from the model and nothing else.”

Final word:Wayne makes a good point. DX will suggest up to 2q vertices if that many are available from the geometry based on inputted constraints. So for 12 components it defaults to 29 points, including 5 additional centroid reps.

—Mark

(Learn more about component screening by attending the computer-intensive two-day workshop Mixture Design for Optimal Formulations. Click on the title for a complete description of this class. Link from this page to the course outline and schedule. Then, if you like, enroll online.)

4: Webinar alert: Making the most from design evaluation: when, why, how and what to do about it

In this webinar titled “Unleashing Evaluation: Giving Perspective to Power, Precision, and Problems” (repeated three times for your scheduling convenience), Stat-Ease Consultant Wayne Adams looks at the often neglected tools for evaluating the capability of a design. Wayne will show the usefulness of these tools to choose from alternative experiment designs upfront and assess after-the-fact the negative impact caused by unplanned changes. Sign up now to reserve your spot and learn how to ensure that your hard work remains focused on your goal.

Reserve your Webinar seat now at by clicking one of the links below:

1. Tuesday, January 5 at 5:30 am USA-CST* for Europe, Africa, Middle East and western Asia (others welcome!)
2. Wednesday, January 6 at 11 am USA-CST* for the Americas and Caribbean (others welcome!)
3. Thursday, January 7 at 8 pm USA-CT* for eastern Asia and Oceania (others welcome!)

If this is your first Stat-Ease webinar, please review these suggestions on how to be prepared.  If questions remain, direct them to our Client Specialist, Rachel Pollack, via [email protected].

5: Info alert: A couple of compelling case studies on DOE, one that demonstrates a breakthrough pathway for proteins created from nonliving matter, and the other promoting its use for overcoming pilot plant challenges

Scientific Computing magazine published this landmark article detailing how two Guru Nanak Dev University researchers deployed DOE to improve peptide bond yield from 20% to 76%. The experiment revealed a synergistic interaction between temperature and pH. This enabled success for the first time in constructing sequence-specific peptides from a noncatalyzed reaction. Well done!

This post by Chemical Processing, also published in their November print edition, provides a consult by Ron Stites on “Using DoE to Capture Valid Data from Pilot Plants.” Inspiring!

6: Blog alert: Productivity Press "Lean Insider" blog post features discussion on the advantage of DOE over traditional scientific methods

Michael Sinocchi, executive editor with Productivity Press, posted an interview of me in this November 17 “Lean Insider” blog. It addresses the frequently asked question about why the tried-and-true scientific method of changing only one factor at a time (OFAT) does not hold up for effective experimentation.

7: Book Giveaway: Win a free DOE Simplified!

(Sorry, due to the high cost of shipping, this offer applies only to residents of the United States and Canada.)

8: Events alert: 2016 appearances posted—calendar them now so you do not miss your opportunity to discover the latest developments in DOE methodology and software!

Stat-Ease will be represented by CAMO Software at the Annual Meeting of IFPAC in the Washington DC area on January 24 through 27. There you will pick up the latest developments in Process Analytical Technology (PAT) and Quality by Design (QbD) in the process industries. Get more details on IFPAC here.

Save the date for the 2nd Asian DOE User Meeting! It will be held in Udaipur, India March 3-5, 2016. Udaipur has a romantic and historic past and is known for its culture, palaces and scenic areas. It is often called the “Venice of East” and was voted the best city in the world in 2009 by the Travel + Leisure magazine. Don't miss this opportunity to meet with other Design-Expert users and increase your DOE skills in this popular tourist destination. More details to follow.

(Second notice.) Save the date for the Sixth European Design of Experiments (DOE) User Meeting and Workshops in lovely Leuven, Belgium on May 18 through 20. This biannual get-together is co-sponsored by Stat-Ease and the on-site host CQ Consultancy. Watch for more details on this fun and informative conference in future Alerts.

Click here for a list of 2016 appearances by Stat-Ease professionals. This is just the start—more events will be posted as they develop throughout the year. We hope to see you sometime in the coming year.

P.S. Do you need a speaker on DOE for a 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.

9: Workshop alert: Enroll now for Experiment Design Made Easy in sunny San Diego

You can do no better for quickly advancing your DOE skills than attending a Stat-Ease workshop. In these 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. Also, take advantage of a $400 discount when you take two complementary workshops that are offered on consecutive days.

• Experiment Design Made Easy (EDME)
  
  • January 25-26, San Diego, CA
  • March 7-8, 2016*, Edison, NJ
    
    
• Response Surface Methods for Process Optimization (RSM)
  • March 9-10, 2016*, Edison, NJ

*Take both EDME and RSM to earn $400 off the combined tuition!

• Mixture Design for Optimal Formulations (MIX)
• January 21-22**, Oslo, Norway (Click here to register for this class.)
• April 4-5, 2016**, Edison, NJ
  
  
• Advanced Formulations: Combining Mixture & Process Variables (MIX2)
• April 6-7, 2016**, Edison, NJ

** Take both MIX and MIX2, or EDME and MIX, to earn $400 off the combined tuition!

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 call Rachel Pollack at 612-746-2030. 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 a private 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.
2021 East Hennepin Avenue, Suite 480
Minneapolis, Minnesota 55413 USA

P.S. Quote for the month:

 
"Statistics is the grammar of science.”

  —Karl Pearson (for a fascinating biography see Karl Pearson and the Origins of Modern Statistics: An Elastician becomes a Statistician by M. Eileen Magnello (Rutherford Journal, December 2005)

Trademarks: Stat-Ease, Design-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, Shari Kraber, Wayne Adams, Brooks Henderson and Martin Bezener
—Statistical advisor to Stat-Ease: Dr. Gary Oehlert
—Stat-Ease programmers led by Neal Vaughn
—Heidi Hansel Wolfe, Stat-Ease sales and marketing director, and all the remaining staff that provide such supreme support!

For breaking news from Stat-Ease go to this Twitter site.

DOE FAQ Alert ©2015 Stat-Ease, Inc.
Circulation: Over 8900 worldwide
All rights reserved.