Stat-Ease
If you are having trouble viewing this email view it online.
 
Vol: 16 | No: 4 | Jul/Aug '16
Stat-Ease
The DOE FAQ Alert
     
 

Stat-Ease Statistical Group

Dear Experimenter,
Here’s another set of frequently asked questions (FAQs) from me and the rest of our StatHelp team about design of experiments (DOE), plus alerts to timely information and free software updates. If you missed the previous DOE FAQ Alert click here.

To open another avenue of communication with fellow DOE and Stat-Ease fans, sign up for The Stat-Ease Professional Network on LinkedIn.

 

 
Stats Made Easy Blog

 

 

 

 

 

 

 

 

 

 

-  

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

1:  Webinar alert: Taking advantage of automated selection tools for response surface modeling
2:  Publication alert: Second edition of RSM Simplified published
3:  Software alert: Version 10.03.1 of Design-Expert® software released (free update for licensed users of v10)
4: FAQ: Dealing with significant curvature from a two-level factorial experiment with center points
5: Expert-FAQ: How to analyze results from a definitive screening design (DSD)
6: Events alert: Stat-Ease on exhibit and on stage with informative talks in the UK and USA—Do not miss this opportunity to network on DOE and charge up your know-how.
7: Workshop alert: See when and where to learn about DOE—Sign up now before classes fill.
 
 


P.S. Quote for the month:
Do not bother putting your hypothesis to the test if you are bound and determined. (Page down to the end of this e-zine to enjoy the actual quote.)


- Back to top -

1: Webinar alert: Taking advantage of automated selection tools for response surface modeling

Via two world-wide webinars, Consultant Martin Bezener will spell out how to put Design-Expert software to good use by “Taking Advantage of Automated Selection Tools for Response Surface Modeling”. Although this is an advanced topic, it will be kept to an intermediate level that suits experimenters, i.e., you need not be a statistician to benefit from attending.

As Martin will explain, after a response surface method (RSM) experiment is performed, one of the critical issues becomes selection of a good empirical model. Recently, automatic model selection tools have become very popular, especially in situations where there are a large number of factors. While these tools are fast and easy to use, some caution must be exercised, especially if the design space is constrained (e.g.; a mixture design) or irregularly shaped.

In this webinar, Martin will provide a brief overview to model selection in RSM. Then, while demonstrating tools in Design-Expert version 10, he will discuss the pros and cons of several automatic model selection techniques.

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

  1. Tuesday, August 9 at 6:30 am USA-CDT*
  2. Wednesday, August 10 at 11:00 am USA-CDT*

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].

*(To determine the time in your zone of the world, try using this link. We are based in Minneapolis, which appears on the city list that you must manipulate to calculate the time correctly.) 


- Back to top -


2: Publication alert: Second edition of RSM Simplified published

Together with my co-author Pat Whitcomb, I am pleased to announce publication of the second edition of RSM Simplified, Optimizing Processes Using Response Surface Methods for Design of Experiments by Taylor&Francis/CRC/Productivity Press, New York, NY. It picks up where our third edition of DOE Simplified leaves off—this two-book series covering the full range of screening, characterization and optimization (SCO). The revised RSM Simplified adds details on split plot tools that handle hard-to-change (HTC) factors as well as practical aspects for the putting these powerful optimization methods to good use. See more about this new book and order it here.


- Back to top -


3: Software alert: Version 10.03.1 of Design-Expert software released (free update for licensed users of v10)

Newly-released version 10.03.1 of Design-Expert software is posted at this download site for free trial evaluation. To update older licensed versions of 10.0, simply download the update from within the program, or download the full installation and reinstall it. The release primarily provides maintenance of existing features. View the Read Me file for details on this update, installation tips, known ‘bugs,’ change history, and FAQs.

PS. Reminder: 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.


- Back to top -


4: FAQ: Dealing with significant curvature from a two-level factorial experiment with center points

Original Question from a Pharmaceutical Product Developer:
“I ran a two-level design with center points on three factors. One of my critical attributes went out of limits on the center point runs only. Design-Expert’s analysis of variance (ANOVA) for the adjusted model reveals significant curvature. Now what should I do?”

Answer from Stat-Ease Consultant Shari Kraber:
“Consider augmenting the factorial design to a central composite design (CCD). This will require you to perform additional runs in order to fit a quadratic model for response surface methodology (RSM). From the design layout screen (the spreadsheet of runs) go to Design Tools, Augment Design, Augment. Choose Central Composite from the pull-down list and accept all the default settings. After completing the new block of runs laid out by Design-Expert, re-analyze the entire set of data (the program will automatically shift to more sophisticated polynomial RSM modeling tools). Before proceeding with this next step, you will do well by working through the Multi-Factor RSM tutorial found here.”

(Learn more about center points 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.)


- Back to top -

5: Expert-FAQ: How to analyze results from definitive screening designs (DSD)

Original Question From an Applied Statistician:
“What is your recommended approach for analysis of definitive screening designs, in particular when second-order effects appear to be significant—given that DSDs confound two-factor interactions with quadratic terms?”

Answer from Stat-Ease Consultant Wayne Adams:
“The tried and true strategy, as with any screening design, is to fit the linear model first to see which factor effects are both significant and substantial; but this sells the DSD a little bit short. Most papers recommend using AICc with a forward selection on the quadratic model only adding terms that improve (make smaller) this criterion.

[AICc stands for Akaike Information Criterion corrected for a small design. Akaike is pronounced (ah kah ee Kay).]

I agree that this approach can work, but only under certain conditions, that are likely to be unprovable a priori. If there are too many significant second-order effects, augment the design to untangle them.

In situations where there are too many factors involved in the interactions or quadratic terms, it goes beyond hope that true models can be discovered. What usually happens is a model can be formed that fits the data very well, but can't predict out into the vertices very well—it doesn't quite fit to the whole cubic region. The remedy here is to limit the predictions by not extrapolating very far beyond the data.”


P.S. We moved DSDs from the Factorial to the Response Surface tab in v10 of Design-Expert, where it can be found under the heading “Supersaturated”. This facilitates analyzing the results using forward selection by AICc, which is now the default for algorithmic model reduction. For an in-depth overview of DSDs, see Definitive Screening Design Characteristics and Analysis Methods.


- Back to top -

6: Events alert: Stat-Ease on exhibit and on stage with informative talks in the UK and USA—Do not miss this opportunity to network on DOE and charge up your know-how.

(Second Notice) Connect with Consultant Pat Whitcomb at ENBIS 16 in Sheffield, UK, September 11-15. See him at the Stat-Ease exhibit there and attend his talk on “A Synergistic Blend of Multivariate Analysis Methods with Design of Experiments Tools”.

(Second Notice) Back in the USA in Minneapolis, the home of Stat-Ease, see a number of our staff at SciX where one of our Consultants will co-teach a one-day workshop on Introduction to DOE and Chemometrics with Camo Software on Tuesday September 20. I plan to give a talk on quality by design (QbD) on “Managing Uncertainty in Design Space” and spend some time at booth 100 organized by Camo, where we will be displaying our programs; Unscrambler X and Design-Expert; respectively. Register for the SciX conference here.

Consultant Martin Bezener will present a talk on “Restricted-Randomization Optimal Design of Experiments Combining Mixture and Non-Mixture Factors” at the Fall Technical Conference (FTC) in Minneapolis which will be held on October 6-7. Register here. The General Chair for FTC is Stat-Ease Consultant Shari Kraber.

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

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.


- Back to top -

7: 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 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.

* Take both EDME and RSM 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 contact the Client Specialist, Rachel, at [email protected] or 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].


- Back to top -

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—Do not bother putting your hypothesis to the test if you are bound and determined:


"If an experiment does not hold out the possibility of causing one to revise one’s views, it is hard to see why it should be done at all.”

—Peter B. Medawar, 1960 Nobel Prize winner for his discovery of acquired immunological tolerance, which became the foundation of tissue and organ transplantation.

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!

Twitter-SmileyFor breaking news from Stat-Ease go to our Twitter feed.

DOE FAQ Alert ©2016 Stat-Ease, Inc.
Circulation: Over 6000 worldwide
All rights reserved.

 
 

Subscribe