Stat-Ease Blog

Thanksgiving is fast approaching—time to begin the meal planning. With this in mind, the NBC Today show’s October 22nd tips for "75 Thanksgiving desserts for the sweetest end to your feast" caught my eye, in particular the Donut Loaf pound cake. My 11 grandkids would love this “giant powdered sugar donut” (and their Poppa, too!).

I became a big fan of pound cake in the early 1990s while teaching DOE to food scientists at Sara Lee Corporation. Their ready-made pound cakes really hit the spot. However, it is hard to beat starting from scratch and baking your own pound cake. The recipe goes backs hundreds of years to a time when many people could not read, thus it simply called for a pound each of flour, butter, sugar and eggs. Not having a strong interest in baking and wanting to minimize ingredients and complexity (other than adding milk for moisture and baking powder for tenderness), I made this formulation my starting point for a mixture DOE, using the Sara Lee classic pound cake as the standard for comparison.

As I always advise Stat-Ease clients, before designing an experiment, begin with the first principles. I took advantage of my work with Sara Lee to gain insights on the food science of pound cake. Then I checked out Rose Levy Beranbaum’s The Cake Bible from my local library. I was a bit dismayed to learn from this research that the experts recommended cake flour, which costs about four times more than the all-purpose (AP) variety. Having worked in a flour mill during my time at General Mills as a process engineer, I was skeptical. Therefore, I developed a way to ‘have my cake and eat it too’: via a multicomponent constraint (MCC), my experiment design incorporated both varieties of flour. Figure 1 shows how to enter this in Stat-Ease software.

By the way, as you can see in the screen shot, I scaled back the total weight of each experimental cake to 1 pound (16 ounces by weight), keeping each of the four ingredients in a specified range with the MCC preventing the combined amount of flour from going out of bounds.

The trace plot shown in Figure 2 provides the ingredient directions for a pound cake that pleases kids (based on tastes of my young family of 5 at the time) are straight-forward: more sugar, less eggs and go with the cheap AP flour (its track not appreciably different than the cake flour.)

For all the details on my pound cake experiment, refer to "Mixing it up with Computer-Aided Design"—the manuscript for a publication by Today's Chemist at Work in their November 1997 issue. This DOE is also featured in “MCCs Made as Easy as Making a Pound Cake” in Chapter 6 of Formulation Simplified: Finding the Sweet Spot through Design and Analysis of Experiments with Mixtures.

The only thing I would do different nowadays is pour a lot of powdered sugar over the top a la the Today show recipe. One thing that I will not do, despite it being so popular during the Halloween/Thanksgiving season, is add pumpkin spice. But go ahead if you like—do your own thing while experimenting on pound cake for your family’s feast. Happy holidays! Enjoy!

To learn more about MCCs and master DOE for food, chemical, pharmaceutical, cosmetic or any other recipe improvement projects, enroll in a Stat-Ease “Mixture Design for Optimal Formulations” public workshop or arrange for a private presentation to your R&D team.

Consider Screening Down Components to a Vital Few Before Studying Them In-Depth

At the outset of my chemical engineering career, I spent 2 years working with various R&D groups for a petroleum company in Southern California. One of my rotations brought me to their tertiary oil-recovery lab, which featured a wall of shelves filled to the brim with hundreds of surfactants. It amazed me how the chemist would seemingly know just the right combination of anionic, nonionic, cationic and amphoteric varieties to blend for the desired performance. I often wondered, though, whether empirical screening might have paid off by revealing a few surprisingly better ingredients. Then after settling in on the vital few components doing an in-depth experiment may very well have led to discovery of previously unknown synergisms. However, this was before the advent of personal computers and software for mixture design of experiments (DOE), and, thus, extremely daunting for non-statisticians.

Nowadays I help many formulators make the most from mixture DOE via Stat-Ease softwares’ easy-to-use statistical tools. I was very encouraged to see this 2021 meta-analysis that found 200 or so recent publications (2016-2020) demonstrating the successful application of mixture DOE for food, beverage and pharmaceutical formulation development. I believe that this number can be multiplied many-fold to extrapolate these findings to other process industries—chemicals, coatings, cosmetics, plastics, and so forth. Also, keep in mind that most successes never get published—kept confidential until patented.

However, though I am very heartened by the widespread adoption of mixture DOE, screening remains underutilized based on my experience and a very meager yield of publications from 2016 to present from a Google-Scholar search. I believe the main reasons to be:

• Formulators prefer to rely on their profound knowledge of the chemistry for selection of ingredients (refer to my story about surfactants for tertiary oil recovery)
• The number of possibilities get overwhelming; for example, this 2016 Nature publication reports that experimenters on a pear cell suspension culture got thrown off by the 65 blends they believed were required for simplex screening of 20 components (too bad, as shown in the Stat-Ease software screenshot below, by cutting out the optional check blends and constraint-plane-centroids, this could be cut back to substantially.)

• Misapplying factorial screening to mixtures, which, unfortunately happens a lot due to these process-focused experiments being simpler and more commonly used. This is really a shame as pointed out in this Stat-Ease blog post

I feel sure that it pays to screen down many components to a vital few before doing an in-depth optimization study. Stat-Ease software provides some great options for doing so. Give screening a try!!

For more details on mixture screening designs and a solid strategy of experiments for optimizing formulations, see my webinar on Strategy of Experiments for Optimal Formulation. If you would like to speak with our team about putting mixture DOE to good use for your R&D, please contact us.

Thank you to our presenters and all the attendees who showed up to our 2022 Online DOE Summit! We're proud to host this annual, premier DOE conference to help connect practitioners of design of experiments and spread best practices & tips throughout the global research community. Nearly 300 scientists from around the world were able to make it to the live sessions, and many more will be able to view the recordings on the Stat-Ease YouTube channel in the coming months.

Due to a scheduling conflict, we had to move Martin Bezener's talk on "The Latest and Greatest in Design-Expert and Stat-Ease 360." This presentation will provide a briefing on the major innovations now available with our advanced software product, Stat-Ease 360, and a bit of what's in store for the future. Attend the whole talk to be entered into a drawing for a free copy of the book DOE Simplified: Practical Tools for Effective Experimentation, 3rd Edition. New date and time: Wednesday, October 12, 2022 at 10 am US Central time.

Even if you registered for the Summit already, you'll need to register for the new time on October 12. Click this link to head to the registration page. If you are not able to attend the live session, go to the Stat-Ease YouTube channel for the recording.

Want to be notified about our upcoming live webinars throughout the year, or about other educational opportunities? Think you'll be ready to speak on your own DOE experiences next year? Sign up for our mailing list! We send emails every month to let you know what's happening at Stat-Ease. If you just want the highlights, sign up for the DOE FAQ Alert to receive a newsletter from Engineering Consultant Mark Anderson every other month.

Thank you again for helping to make the 2022 Online DOE Summit a huge success, and we'll see you again in 2023!

We recently published the July-August edition of The DOE FAQ Alert. One of the items in that publication was the question below, and it's too interesting not to share here as well.

Original question from a Research Scientist:

"Empowered by the Stat-Ease class on mixture DOE and the use of Design-Expert, I have put these tools to good use for the past couple of years. However, I am having to more and more defend why a mixture design is more appropriate than factorials or response surface methods when experimenting on formulations. Do you have any resources, blogs posts, or real-world data that would better articulate why trying to use a full factorial or central composite design on mixture components is not the most effective option?"

Answer from Stat-Ease Consultant Martin Bezener:

“First, I assume you are talking about factorials or response surface method (RSM) designs involving the proportions of the components. It makes no sense to use a factorial or RSM if you are dealing with amounts, since doubling the amount of everything should not affect the response, but it will in a factorial or response-surface model.

"There are some major issues with factorial designs. For one thing, the upper bounds of all the components need to sum to less than 1. For example, let’s say you experimented on three components with the following ranges:

A. X1: 10 - 20%
B. X2: 5 - 6%
C. X3: 10 - 90%

then the full-factorial design would lay out a run at all-maximum levels, which makes no sense as that gives a total of 116% (20+6+90). Oftentimes people get away with this because there is a filler component (like water) that takes the formulation to a fixed total of 100%, but this doesn't always happen.

"Also, a factorial design will only consider the extreme combinations (lows/highs) of the mixture. So, you'll get tons of vertices but no points in the interior of the space. This is a waste of resources, since a factorial design doesn't allow fitting anything beyond an interaction model.

"An RSM design can be ‘crammed’ into mixture space to allow curvature fits, but this is generally a very poor design choice. Using ratios of components provides a work-around, but that has its own problems.

"Whenever you try to make the problem fit the design (rather than the other way around), you lose valuable information. A very nice illustration of this was provided in the by Mark Anderson in his article on the “Peril of Parts & the Failure of Fillers as Excuses to Dodge Mixture Design” in the May 2013 Stat-Teaser.”

An addendum from Mark Anderson, Principal of Stat-Ease and author of The DOE FAQ Alert:

"The 'problems' Martin refers to for using ratios (tedious math!) are detailed in RSM Simplified Chapter 11: 'Applying RSM to Mixtures'. You can learn more about this book and the others in the Simplified series ('DOE' and 'Formulation') on our website."

Links for additional information:

• Back issues of The DOE FAQ Alert
• Signup to receive the bimonthly DOE FAQ Alert via email
• Learn more about mixture design by attending our computer-intensive workshop Mixture Design for Optimal Formulations