Willis Jensen, WL Gore and Associates, and Pamela McGovern, USDA National Agricultural Statistics Service, on behalf of the ASA SPAIG Committee

The annual ASA Statistical Partnerships in Academe, Industry, and Government (SPAIG) Award highlights outstanding partnerships between academe, industry, and government organizations and promotes new partnerships. This award is distinct from other ASA awards in that it recognizes outstanding collaborations between organizations, while recognizing key individual contributors.

This year’s SPAIG Award honors the collaboration between Procter & Gamble (P&G) and the H. Milton Stewart School of Industrial and Systems Engineering at Georgia Institute of Technology (GT). The award winner was announced as part of JSM 2020 and recognizes their collaborative research leading to significant advancements in quality, reliability, and analysis and design of physical and computer experiments.

We asked William Brenneman from P&G and Roshan Joseph from GT to answer the following questions about their collaboration:

Briefly describe how the collaboration started.

William: P&G and GT started their collaboration through relationships formed at the University of Michigan (1996–2000). Roshan and I were graduate students together in the statistics department and Jeff Wu was the chair. After graduation (2000), I started working at P&G and Roshan and Jeff became faculty at GT (2003). We kept in touch through meeting at conferences. The research started in 2006, when I supervised Lulu Kang as a P&G summer intern. Lulu also did a second internship, and each one resulted in a published paper on research that was of high importance to P&G. The second paper on mixture-of-mixture analysis and design was part of Lulu’s dissertation.

What are the major benefits that have come from the collaboration that would not have otherwise happened?

William: The major benefits for P&G have been that we have solved problems that are of interest and benefit to P&G research and development. Most of the topics are in the design and analysis of experiments, either physical or computer, or the combination of both. Each of the solutions have been utilized at P&G in our innovation process. From a personal perspective, the collaboration has really brought great joy to my work life—to be able to work with Roshan and Jeff, who are incredibly innovative, and their top-notch students has really added to my professional satisfaction. And I would say that working directly with the students and getting to know them on a personal level, especially through the internships or research assistantships, has been very gratifying personally.

Roshan: One of the major benefits of collaborating with P&G was that we got to work on real-life problems. Many of the problems have led to methodological research in statistics and journal publications. But most importantly, seeing the results being used in practice gave great satisfaction for our research. The students also benefitted a lot from the collaboration. Both William Brenneman and William Myers (another collaborator) were actively involved in dissertation committees of our PhD students. Whenever they visited the GT campus, they were keen in talking to the students and sharing their experience of working as a statistician in industry. In fact, William Brenneman has held the position of an adjunct professor at GT since 2013 because of the valuable mentoring he has provided to our students.

What has been the most rewarding and challenging aspects of the collaboration?

William: As I mentioned above, the most rewarding part has been working with Roshan, Jeff, and the many students. I really like to see the students go on to make contributions after graduating in either industry or academia. That is really rewarding—a part of me would like to work full time with students, but at least this allows me to do this on a very part-time basis. The other rewarding part is that finding really good new stats problems can be hard when in academia, and so when I run into problems that I know are new, it is really satisfying to partner with academia to see how the problems enhance the academic process, as well. A truly win-win for both in my mind.

The most challenging aspects have been around the legal part of the research. Working with GT has become more routine, but at first it was difficult working with both P&G and GT legal teams to come up with an agreement that is okay for both sides.

Roshan: Working with P&G statisticians was extremely rewarding because they were all very well-versed in advanced statistical methods. The research problem was already well formulated, or easy to formulate within a few meetings. Their expectations were clear, and the communications were very easy. So, we really enjoyed every bit of our collaboration with P&G. Personally, one of my most rewarding experiences was when my former student, Shan Ba (another contributor), joined P&G’s statistics department as a scientist. He became an integral part of the department and further strengthened our collaboration.

As William mentioned, the legal part was the most challenging aspect of the collaboration, but once it was established, everything else went smoothly.

What advice would you give to individuals and organizations looking to be more collaborative?

William: I would say to individuals from industry to just get started. The easiest way is through internship and co-op experiences. Even if the intern project is just more routine, it can be of great value to the student to experience life as an industrial statistician. But the best experiences are those in which we in industry can determine a problem that requires new research and that can be started during the internship period, but then followed up after the student goes back to the university. The hard part for those in industry is keeping up with as much of the latest research as we can so we can identify new research opportunities. This is easier said than done, but is well worth the effort, as solving these types of problems can lead to much better solutions for the company and much better research problems for students’ dissertations. Win-win.

Roshan: Academia tends to value theoretical research more, but the satisfaction you can gain by doing applied research and the potential to make an impact in real practice is well worth the time and effort. My advice is that when you work on an industry problem, the primary focus should be to solve the problem in the most beneficial way to the industry. Many times, these problems will lead to new methodological research and student dissertations, but that should be considered as a secondary goal. The collaboration has also enriched my teaching. I found my students are better motivated when I share my real experience from industry, rather than when I talk to them about some textbook examples.

Learn more about the SPAIG Committee and award nominations. Nominations for the 2021 SPAIG Award are due March 1, 2021.