|
FAQ
Pros and cons of central composite versus Box-Behnken design
Original question from a Metallurgist/Process Engineer:
“There are only three factors we can realistically manipulate in our plant for optimizing the yield. What design do you recommend? By the way, I’ve had a play around with the software, so far it is easy to use and intuitive. I really like the graphs with the slider bars for visualizing data.”
Answer:
Thank you for your compliments on our software!
I recommend you deploy a response surface method (RSM) design. The simplest one will be the three-level Box-Behnken. Another option is the central composite design (CCD), which you can work through via our Response Surface tutorial (via Help). However, that requires more levels and more aggressive ranges. See this Stat-Ease blog on "Choosing the Best Design for Process Optimization" by Shari Kraber for a deeper discussion on these options as well as optimal design.
Follow-up observation from user:
“I’ve had a look at the Box-Behnken and central composite designs as suggested and played around with some designs. Because we need the most aggressive settings to achieve the best result I am tending towards the CCD, with the face-centered option to keep the factors at reasonable levels.”
Further discussion:
Very sensible! As pictured below, the face-centered central composite design (FCD) pushes runs out to the corners of the cube. It is a good alternative to BBD for three factors, which puts points at the centers of the edges and thus does not test the extreme vertices of the cubical factor space.
 
FCD (left) versus BBD (right) layouts for three factors
By the way, as the number of factors increases, the FCD exhibits high variance inflation factors (VIFs) relative to the BBD. For these larger designs, a good compromise between FCD and the standard CCD is provided by the “practical alpha" choice in Stat-Ease software, which sets axial (star) points at a coded distance equal to the 4th root of the number of factors—not nearly so far out as those for the rotatable option, yet nearly as low on VIFs. Thus, the practical CCD preserves sound statistical properties without nearly so extreme on factor levels as the standard (rotatable) one. For more details see my 2003 talk to the Joint Statistical Meeting (JSM) on Practical versus Statistical Aspects of Altering Central Composite Designs.
|
|
EVENTS
Join us on October 24-25 for the premier conference on design and analysis of experiments—the 2023 Online DOE Summit. It is free and available to all online! I will lead it off on with a tasty talk detailing how a “Masterful experiment delivers delectable chocolate chip cookies.” Following me are more meaty presentations by a number of industrial DOE practitioners and trainers. The conference will be capped off with a briefing by Stat-Ease President and CTO Martin Bezener on “The Latest & Greatest in Design-Expert® and Stat-Ease® 360” software. See the times and speakers listed here and follow the links for free registration for one or both days of our 2023 Online DOE Summit.
More DOE Talks
I am presenting at two professional in-person events this coming fall:
- A talk on “Deploying DOE to Identify Vital Factors That Affect Quality” for the American Society for Quality (ASQ), Minnesota Section's 2023 Quality and Regulatory Affairs Conference in Minneapolis on October 10 at 3:00 pm Central in their Continual Improvement & Metrics track.
I look forward to sharing my know-how on multifactor DOE methods with my ASQ and AIChE colleagues.
Click here to view the complete list of Stat-Ease events.
Do you need a speaker on DOE for a learning session within your company or professional society at regional, national, or international levels? If so, please get back to me.
|
|
ONLINE LEARNING
Sharpen up your DOE skills with a mix of free and paid training: whatever fits your business needs.
Comprehensive DOE courses — Online instructor-led learning
See this web page for the complete schedule of upcoming Stat-Ease courses. To enroll in the workshop that suits you best, click Register on that page, or click here to contact us.
If you lead a group of 6 or more colleagues, save money and customize content via a private workshop. For a quote, please contact us.
Free webinars — Sign up now to take advantage
|
|
BLOGS
StatsMadeEasy Blog
My wry look at all things statistical and/or scientific with an engineering perspective.
|
|
Feel free to get back to me via [email protected] with further questions or comments: I would really appreciate hearing from you!
All the best,
Mark J. Anderson, PE, CQE, MBA
Engineering Consultant, Stat-Ease, Inc.
www.linkedin.com/in/markstat/
|
|
|
“Supposing is good, but finding out is better.”
—Mark Twain, Autobiography of Mark Twain, Volume 3, p. 98
|
|
|
Stat-Ease, Design-Expert and Statistics Made Easy are registered trademarks of Stat-Ease, Inc.
Circulation: Over 2500 worldwide
|
|
|
|
