Design of experiments (DOE or sometimes DoE), also known as “planned experimentation,” is an experimental method backed by statistics to help you make breakthrough product and process improvements while reducing your operations budget. See below for in-depth explanations on multifactor testing and how it works:

• Trimming the FAT out of Experimental Methods— a primer on DOE versus the traditional OFAT method of experimentation.
• Trimming the FAT: Part II—This follow-up article offers a case study that illustrates how a two-level factorial DOE can reveal a breakthrough interaction.
• Rethink Experiment Design—This is another article on DOE versus the traditional OFAT method.
• Magic of Multifactor Testing Revealed by Fun Physics Experiment—a fun intro to design of experiments with a bouncy ball experiment.
• Eight Keys to Successful DOE—Using DOE successfully depends on understanding eight fundamental concepts which are explained in this article.

If you’re doing Six Sigma or Lean, DOE is a very important element in these methodologies. Learn more about how they interact with these articles:

• Design of Experiments—A Primer—This article looks at some of the basic DOE concepts that are necessary for practitioners to understand.
• The Six Sigma Method and Design of Experiments—This article by Peter Peterka (www.6sigma.us) explains what Six Sigma is and DOE's role in it.
• Cost-Effective and Information-Efficient Robust Design for Optimizing Processes and Accomplishing Six Sigma Objectives—Standard factorial designs (one array) offer a cost-effective and information-efficient robust design alternative to parameter designs (two -array) made popular by Taguchi. This paper compares these two methods (one-array versus two-array) in depth via an industrial case study. It then discusses advanced tools for robust design that involve application of response surface methods (RSM) and measurement of propagation of error (POE). Also see this postscript to the article.
• Six Sigma for the Road—Read this Stat-Teaser newsletter article from Mark Anderson on using propagation of error (POE) and multiple response optimization to analyze a DOE on driving time.

Ready to learn more? Take our free Stat-Ease Academy eLearning courses for hands-on exploration of basic DOE topics, or grab one of the Simplified books at your local book retailer:

The Statistics Made Easy by Stat-Ease YouTube channel contains hundreds of hours of DOE educational content. Start with the DOE Crash Course for Experimenters, then browse the rest of the "New to DOE?" playlist or explore further topics of interest at your leisure. We hold occasional live webinars as well: see what's next on this page.

You can also sign up for the every-other-month DOE FAQ Alert newsletter, which contains real DOE questions from Stat-Ease clients and real answers from our staff.

If you need further instruction, or if you’re looking for a deeper dive into DOE topics, register for one of our public training courses or organize a private training session for your team of 6+.

We periodically offer a live New-User Intro to Design-Expert software webinar, but if you don’t see one scheduled, you can view the recording on our YouTube channel. Learn about factorial design, the core tool for DOE, followed by a peek at response surface methods (RSM) for process optimization and last, but not least, a look into mixture design for optimal formulation.

Use the software’s built-in tutorials to explore different functions of the software, from simply navigating it (using the One-Factor Categoric tutorial), to performing response surface methodology (RSM), to applying Python scripting to your experiments in Stat-Ease 360.

Sign up for our monthly event & software feature newsletters to be notified when we’re hosting new webinars and for handy software tips & tricks.

If you need further instruction, or if you’re looking for a deeper dive into DOE topics, register for one of our public training courses or organize a private training session for your team of 6+.

Formulation work requires a specific type of DOE known as “mixture designs.” Factorial and RSM methods simply won’t work when you’re handling components of a mixture as opposed to independent factor variables. Use these resources to learn about mixture design:

• A Primer on Mixture Design: What’s In It for Formulators?—a "what's in it for me" introduction to encourage formulators to work through the more intense aspects of mixture design, response modeling, statistical analysis and numerical optimization.
• A Crash Course in Mixture Design of Experiments—this on-demand, recorded webinar by Stat-Ease President Martin Bezener, Ph.D. provides a compelling demo that lays out what makes mixture design of experiments (DOE) so effective for accelerating your formulation efforts. The fuel provided in this 1-hour webinar will kick-start your first mixture designed experiment.

Ready to learn more? Grab the book Formulation Simplified: Finding the Sweet Spot through Design and Analysis of Experiments with Mixtures, by Anderson, Whitcomb, & Bezener. Formulation Simplified deviates significantly from all other statistical texts on design of experiments (DOE) for mixtures by making these powerful methods easy to study and fun to read. The authors provide a rich array of insightful case studies that illustrate the essentials of mixture DOE. Real-world examples are offered as problems at the end of many chapters for those who are serious about trying these new tools on their own.

You can also sign up for the every-other-month DOE FAQ Alert newsletter, which contains real DOE questions from Stat-Ease clients and real answers from our staff.

If you need further instruction, or if you’re looking for a deeper dive into DOE topics, register for one of our public training courses or organize a private training session for your team of 6+.

Great! If your question is “how can I buy Stat-Ease software?” or similar, please use this form to send our team a message. If your question is about DOE or using Stat-Ease software, you can send those questions to our experts using this form instead.