BLOGS
StatsMadeEasy Blog
(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 “Greg's DOE Adventure - Factorial Design, Part 2”. Take a look!
|
|
SOFTWARE ALERT
Version 12.0.9 of Design‑Expert® Software Released
Current Design‑Expert v12 (DX12) users will do well by updating to 12.0.9 via Help from within your installed program. View the changelog for details on the changes—mainly maintenance items. 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.)
|
|
|
CONFERENCE ALERT
Free webinars by world experts on experiment design, plus inspiring case studies by leading practitioners
Kicked off by my presentation on “Know the SCOR for Multifactor Strategy of Experimentation”, the 2020 Online DOE Summit provides an unprecedented opportunity to sharpen up your DOE skills. Our Founder Pat Whitcomb will lead off this virtual conference by relating “My Lifelong Journey with DOE”. Next comes Marcus Perry: Editor in Chief, Quality Engineering; Professor of Statistics, The University of Alabama, with “Some Experiences in Modern Experimental Design”. The final keynoter is Geoff Vining: Professor of Statistics, Virginia Tech, who speaks on “Innovative Mixture-Process Models”. Stat‑Ease Consultant and Director of R&D, Martin Bezener, follows with a tutorial on “Strategies for Sequential Experimentation”.
The Summit then shifts to a series of sessions by users who will demonstrate how Design‑Expert makes the most from every experiment.
|
|
FAQ
How do you use the perturbation plot to help visualize response surfaces?
Original question from a DOE Trainer:
“I would appreciate it very much if you could help me a bit with understanding the perturbation plot provided in Design‑Expert as one of its outstanding selection of Model Graphs for factorial and response surface designs. To keep things simple, I have deliberately excluded perturbation plots from my training. However, with your help I may add a slide on it. I appreciate how this shows each factor alone affects the response. This comes through very clearly as seen from at the center of all factor ranges—the default. However, the plot becomes more difficult to interpret when one moves the reference point from the default. For example, see this graph where I moved all three active factors B, D and E off center to 1, 0.25 and -0.6 coded units; respectively. Why does the line for D only extend up to 0.75?”
 
|
|
Answer from Stat‑Ease Consultant Shari Kraber:
“At the initial (default) reference point each of the three active factors (the ones in your model) could change by plus or minus 1 coded unit, thus filling out the whole range on the perturbation plot (see graph at left). However, after you moved the reference point, factor D starts at 0.25 (the new 0 point on the bottom axis) and thus can only go up another 0.75 in coded units before it hits its limit at plus 1. On the other hand, factor D can now go down by 1.25 units to its limit at minus 1. Design-Expert indicates this by the arrow seen at the left end of the D line. Similar issues come into play for factors B and E.”
P.S. For a primer on perturbation plots, see pages 86-87 in the section on “Model Graphs” in RSM Simplified: Optimizing Processes Using Response Surface Methods for Design of Experiments, 2nd Edition. To keep things simple, I prefer not to change the reference point from its default at the center of the design space and use the perturbation plot for these two special cases:
- The model is linear, that is, only main effects. Then you can get a feel for the relative effects for magnitude and direction at a glance.
- The model is nonlinear and includes more than two factors. Then you can select the two most curved factors to put on your initial contour and 3D plots, thus making them most ‘dramatic’.
If you do change the reference point on the perturbation plot, it is best to do so to the setup recommended by numeric optimization. Design-Expert will automatically do so. Then the perturbation displays sensitivity of each factor being moved from the peak of desirability.
-Mark
|
|
VIDEO ALERT
See how optimal design makes it easy for formulators to set up experiments on mixtures
A newly posted video on the Stat‑Ease YouTube Channel demonstrates Optimal Mixture Design made easy by Design‑Expert software. Check it out!
(Learn more about optimal design by attending the next distance-learning presentation of Mixture Design for Optimal Formulations.)
|
|
INFO ALERT
“Design of Experiments Pinpoints Better Products” and “Response Surface Methods Point the Way to Higher rAAV Yields”
Master chemist Jason Pandolfo demonstrates the power of DOE via a one-two punch of factorial and mixture design in this article on Design of Experiments Pinpoints Better Products published by Lubes 'N' Greases in their April 2020 edition. Check out the wonderful graphic of experimenters in action! The “impact of this study was immediate and long lasting”. For another view of this inspiring case study, see this stand-alone newsletter article.
The May issue of Genetic Engineering & Biotechnology News (GEN) features this stirring story of how Response Surface Methods Point the Way to Higher rAAV Yields. You can see by the compelling 3d graphic in Figure 4 how researchers at Adverum Biotechnologies achieved peak results from advanced DOE methods made possible by Design‑Expert. |
|
EVENTS ALERT
“Designing High Impact Experiments” webinar for American Society of Quality Statistics Division
Stat-Ease Consultant Shari Kraber will brief the American Society of Quality Statistics Division on “Designing High Impact Experiments” via her webinar at 12 noon USA-CDT on June 23. See the abstract and register for her talk here.
PS. 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.
|
|
WORKSHOP ALERT
New distance-learning options to master DOE from the comfort of your own desk—Introductory-pricing and early-bird deals end soon, so sign up now!
You can do no better for quickly advancing your DOE skills than attending a Stat‑Ease workshop. Our expert instructors provide you with a lively and extremely informative series of lectures interspersed by valuable hands-on exercises.
See this web page for complete schedule and site information on all Stat‑Ease distance-learning workshops. To enroll, scroll down to the workshop of your choice and click on it, or email our Lead Client Specialist Rachel Poleke 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 a private workshop, which is most convenient and effective for your staff. For a quote, email [email protected]. |
|
|
“Using statistics to understand how the world worked was to understand the mind of God.”
— David Spiegelhalter, University of Cambridge statistician, speaking about Florence Nightingale’s belief in the power of data for healthcare (Source: “Thou Shalt Not Underestimate Florence Nightingale”, Smithsonian magazine, March 2020.).
|
|
|
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 programmers: Hank Anderson, Joe Carriere, and Mike Brownson
Stat-Ease business staff: Cathy Hickman, Greg Campbell, and Rachel Poleke
—who provide such supreme support!
Circulation: Over 7500 worldwide
|
|
|
|
