Members of the ASA Committee on Funded Research (CFR), whose charge includes facilitating communication and interaction between the statistical community and funding organizations, sought to learn about how university departments engage industry in research collaborations. They found many departments willing to share their experiences by answering the following questions. CRF members hope this information furthers industry collaborations by the statistical community.

IOWA STATE UNIVERSITY

Max Morris is professor and chair of the department of statistics at Iowa State University and a statistical consultant affiliated with Los Alamos National Laboratory.

Please describe your department’s industry and/or government engagements/collaborations.

We have a set of agreements with various companies that provide support and professional experience for some of our more advanced graduate students. In brief, we sign a contract with the company to provide statistical support for 20 hours per week (the typical research assistant load) for one or more semesters. The contract is signed by the company and university and is processed through our grants and contracts office. The company pays the university, and the funds are used to support a graduate assistantship for the student (tuition and stipend at our usual rate) for the prescribed length of time.

With this setup, the student is an employee of the university, rather than the company—an arrangement especially helpful for some international students. For the most part, department faculty are not directly involved. The work done by the student is generally what we would call “consulting” or “applied research” and most often not related to their dissertation research. The experience is comparable to that of many industrial summer internships, but may last more than one term and is such that the student can do the work from campus.

How did the engagement/collaborations come about? How are the collaborations sustained in a way that is mutually beneficial to students, the university, industry, and the community-at-large?

These agreements are usually initiated by the companies, with communication to our department asking about opportunities to work with our students. Beyond financial support, the benefit to the student is primarily in the form of professional experience. Although the immediate benefit to the company is the work done by the student, there is often an interest in getting an “early look” at a potential employee, much as with many summer internship programs.

What arrangements about the collaborations might be helpful to share?

The arrangements are formalized through a contract between the company and university. I (as department chair) handle setting these up and, along with the associate chair, identify students who are good candidates. In some cases, the company also participates by interviewing students we identify as candidates. Before it is finalized, a contract must be approved by the university grants and contracts office; invoicing and fund transfer is handled by our central administration. There is usually little direct faculty oversight of the students (and so they must be relatively mature and independent), but as the university contact for these arrangements, I receive feedback from the companies about the students’ performance. In most cases, the students involved in these arrangements participate with a company for multiple semesters.

What benefits for the department have you seen come out of these collaborations?

For the department, these arrangements supplement our state-funded teaching assistantships and grant-funded research assistantships, giving us a bit more flexibility in how we arrange support for our students.

SOUTHERN METHODIST UNIVERSITY

Lynne Stokes is chair of the department of statistical science at Southern Methodist University. Work on problems for the US Census Bureau, National Center for Education Statistics, National Oceanic and Atmospheric Administration, US Energy Information Administration, and other federal agencies has been some of the most interesting of her career.

Please describe your department’s industry and/or government engagements/collaborations.

We have several types of engagements with business/government that pay our graduate students. Besides the usual grants, we currently have the following two activities:

• Local business (management consulting) provides stipend/overhead for a PhD student (just completed second year of this) in exchange for his time. This is in lieu of a teaching or research assistantship in the department. This has been mutually beneficial, as company has extended full-time offer to student on graduation. Student has accepted. We have had similar arrangements with hospital systems in the past.
• We have a long-term consultancy with a federal agency. The funding goes through our department’s statistical consulting center. Three students have received funding (stipends, summer funding) for work on various projects as needed. Some projects are short term (a few weeks) and some are long term (several years).

Other kinds of relationships:

• We have relationships in the area with companies who come to our department to interview students for summer internships. We facilitate the interviewing process by setting up appointments and giving them space. This is for grad students at both the master’s and PhD levels.
• We provide a forum for local companies to discuss their company’s projects (i.e., give a work-related seminar). This is for our master’s students. This is mostly to facilitate their recruiting of our students, but also provides these students with information about what statistical methods are used in the workplace.
• Local companies provided people (a lot of them the same as mentioned in the previous paragraphs) to serve as judges at the DataFest contest we held this year.
• We provide free consulting services to companies/nonprofits willing to be clients for our graduate students in their required consulting class.

How did the engagement/collaborations come about? How are the collaborations sustained in a way that is mutually beneficial to students, the university, industry, and community-at-large?

Some are cold calls that companies make to our department looking for help either on consulting problems or for recruiting.

The federal agency relationship resulted from a National Academies panel I served on. I met people at the agency who were trying to implement changes. This relationship has now been ongoing for more than 10 years.

A fair number of these relationships were from local companies hiring our students and having them work out well. They are interested in a pipeline to students. Then, over time, our students continue the recruitment cycle.

Some of the relationships are from the local ASA chapter meetings. We hold them on our campus, and so we get to know industry people who come to the meetings.

What arrangements about the collaborations might be helpful to share?

The consulting center is a good vehicle for supporting nondomestic students in the summer. The immigration laws allow people holding student visas to work on campus, but it is more difficult for them to get authorization to work off campus. If a company is willing to give up “control” and let the student be supervised by us and work on our campus, this allows them access to scarce talent.

What benefits for the department have you seen come out of these collaborations?

We have had a variety of benefits, including the following:

• We have been able to place our master’s students well locally, which helps us in recruitment.
• We have been able to have more PhD students (larger number of stipends) and provide more of them support in the summer on a part-time (flexible) basis, so they can work on dissertations and still earn enough to survive.
• We have generated research topics. Two current dissertation topics are from the federal agency consultancy.
• We have not yet hired adjuncts from our industry contacts, but we have discussed this with our industry friends, and will use them if we find ourselves short-handed.
• We have had the opportunity to train our students in consulting.

What benefits for the broader community have you seen come out of these collaborations?

Companies have been very happy with our students as employees. We are now working with two local companies to facilitate their employees’ participation in our master’s program. To do this, we are moving some of our courses to the evening, allowing them to receive credit in some courses (e.g., consulting) from their work experience. This is providing a better workforce for the companies and good (tuition-paying) students for us.

As mentioned above, we do provide free consulting to companies willing to work with our students. This has been especially useful for nonprofits without funds to hire statistical help.

What are the challenges or barriers to such collaborations? What have you learned since your department began these collaborations to deal with these challenges?

Challenges are locating people within the industry to partner with. Then, even after you have a working relationship with someone in a company, they may move on and you lose your contact.

We have not found an effective way to locate people in industry with whom to partner. We rely on them finding us. Because we are in a large metropolitan area, this has worked for us, but I can see it would take a lot more resources than a faculty member/chair would have to cultivate relationships if you were in a smaller area.

Most problems we get from local companies do not rise to the level of research problems for PhD students. They are still beneficial to train students in consulting. However, occasionally, there is a reluctance on the part of the company to pay the full cost of consultation, thinking the problem is helping supply our PhD students with dissertation problems. This has led to some awkward conversations and wasted time in scoping potential problems.

VIRGINIA TECH

Jennifer H. Van Mullekom is the director of the Statistical Applications and Innovations Group (SAIG) at Virginia Tech. She has collaborated on both sides of the fence as an industrial statistician for more than 18 years and now as an associate professor of practice.
 
 
 
 
Anne Ryan Driscoll is an assistant collegiate professor at Virginia Tech. Her research interests include statistical process monitoring, design of experiments, and statistics education. She has also collaborated on projects for the Department of Defense and NASA.

 

Please describe your department’s industry and/or government engagements/collaborations.

The Virginia Tech Department of Statistics collaboration projects span grants from various government agencies, partners in the Virginia Tech Corporate Research Center, department of statistics corporate partners, and private companies that contact the department independently of our other programs.

We currently have a large number of collaborations not related to large federal granting agencies such as the National Institutes of Health or National Science Foundation. Our longstanding Corporate Partners Program creates an opportunity to exchange ideas between corporate statistics leadership and our department. Corporate Partners creates a forum to receive input on our program, place our students in internships and permanent positions, and facilitate collaboration.

We now have a new way to collaborate on short-term and long-term projects through the Statistical Applications and Innovations Group (SAIG), which recently became an external service center. SAIG can provide quotes for design, analysis, and reporting, which also may lead to research collaborations.

How did the engagement/collaborations come about? How are the collaborations sustained in a way that is mutually beneficial to students, the university, industry, and the community-at-large?

Collaborations originate in a variety of ways. Many of our government and industry collaborations have come through alumni employed at various agencies. They often encounter research questions in their daily work that are “important, but not urgent.” Their organizations lack the personnel to devote to these projects, so they turn to the university for both expertise and cost-effective flexible staffing to devote to the projects.

Other collaborations result from master agreements negotiated at either the university or college level. When these projects are presented at conferences, we often create interest from other partners. And, of course, networking at national conferences through the American Statistical Association and other professional societies such as the American Society for Quality also lead to these relationships.

The relationships are sustained by a successful track record of meeting deliverables. Our faculty purposely seek opportunities that align with their research interests and the interests of their students. They also scope opportunities properly and negotiate appropriate compensation for supervision and the students’ funding.

The “important, but not urgent” distinction is also made so we set expectations appropriately. Students have course and degree requirements, so they are not capable of dropping everything to devote all their time to solving a highly time-sensitive problem. Regular program reviews and communications are essential to keeping the program on track.

Another, often overlooked, avenue for research is that of a research associate or research professor. These are degreed staff members and faculty within the university who can be devoted full time to a more urgent problem.

The Corporate Partners Program is another way our department fosters collaborations with outside entities. This program was established 19 years ago with the goal of solidifying the informal ties with industry and government built over many years. Most of the partnerships originate because of ties to alumni who are employed at various companies.

Currently, our department benefits from partnerships with five companies. These partnerships are sustained by providing value to both the department and the company. Benefits to the company include an annual symposium and advisory board meeting to learn about initiatives at the department and university, opportunities to recruit students for internships and full-time employment, and easy access to faculty to foster joint research projects. The department benefits from feedback from the partners about refining our curriculum to tune our program to the rapidly changing needs of users of statistics, funding for graduate student scholarships and awards, research projects for faculty and students, and also employment opportunities for students.

What arrangements about the collaborations might be helpful to share?

Our Office of Sponsored Programs and the Virginia Tech Business Engagement Center provide excellent support in working through the logistics of relationships with outside entities. Existing relationships and agreements between VT industry affiliates and Corporate Research Center partners facilitate development of specific collaborations.

A relationship can be initiated by any faculty member. With respect to business relationships, the faculty member then completes a request for appropriate legal agreements on behalf of the university. Once the appropriate documents have been negotiated and signed, the relationship progresses through the appropriate channels in the university depending on its nature. Smaller projects may be handled at the departmental level, while others are handled at the college level.

Collaborations have varied time frames. For example, a short-term analysis collaboration with SAIG could be initiated by me and last a few weeks to a few years. Many of our long-term research relationships with professors and research groups within the department have lasted five years or more.

What benefits for the department have you seen come out of these collaborations?

The publications resulting from our collaborations are quite numerous, including ones in journals such as Journal of Quality Technology and Quality Engineering. A collaboration between Yili Hong and DuPont resulted in multiple publications and a SPES [Section on Physical and Engineering Sciences] award from the American Statistical Association, whereas one between Geoff Vining and NASA resulted in a NASA Engineering Safety Council Engineering Excellence Award. Those collaborations have also led to student dissertation topics with members of the sponsoring organization serving on the student’s committee.

Anne Driscoll also works on the latter collaboration with NASA. Both Vining and Driscoll collaborate with the Department of Defense. An organically evolving collaboration between Baker Hughes, a GE Company, under Robert Gramacy has led to a line of research on computer model calibration of a seal used in oil and gas extraction that is the topic of a student’s dissertation. William (Bill) Woodall has worked with many external collaborators over the years, most recently with the Food and Drug Administration to provide input on their Quality Metrics Program. Additionally, Robertson Evia- has collaborated on K–12 STEM education studies to provide analysis resulting in Science Communication publications for the National Academy for the Advancement of Science.

We have given short courses and webinars for our corporate partners and are continuing to evolve that relationship. We also look forward to forging new collaborations with external companies through SAIG.

What benefits for the broader community have you seen come out of these collaborations?

Our collaborations have certainly established strategic national government and industry partnerships, but our department also focuses on serving the local community. Christian Lucero is working through SAIG with Virginia Cooperative Extension to develop an analysis tool for a multicenter Centers for Disease Control and Prevention–sponsored diabetes management and prevention program. Our work has also resulted in grant applications for the local school systems, analytics supporting the building of a new dormitory for the Virginia Tech Corp of Cadets, and analytics to support policy and staffing in county governments.

What are the challenges or barriers to such collaborations? What have you learned since your department began these collaborations that help with these challenges?

Two of the barriers to collaborations with universities include the speed of establishing the legal relationships and data security. Both these problems typically have pre-existing solutions at most large research universities, but it often takes perseverance to execute these aspects of the projects properly. All parties involved must be familiar with the legal process and who to contact to initiate and sign agreements. A basic understanding of intellectual property is also important.

Some of our statistics/data science relationships are more consulting in nature, whereas others may involve proprietary algorithm or software development. The boundaries of these relationships and IP [intellectual property] ownership must be established up front. Statisticians and data scientists also need to be able to educate their lawyers regarding the differences of applications versus IP-containing projects. Members of our profession also need to be aware when a project crosses the line into IP development and how to protect their contribution.

Government and corporations have varying IT [information technology] security requirements. Personal health information creates its own unique challenges. Academicians are not typically “raised in this environment.” There is typically much more freedom in sharing information in academia. There is a learning curve associated with both partner requirements and on-campus resources for creating a secure IT environment. Due diligence on this front ahead of time can pay off in expedited project start-up.

Any other comments?

Professional development courses and leadership training provided by the American Statistical Association are crucial to the development of collaborations. An understanding of vision, mission, project management, emotional intelligence, and execution are essential to building and sustaining collaborations.

ARIZONA STATE UNIVERSITY

Jennifer Broatch, an assistant professor and BS program lead at Arizona State University’s West Campus, assists junior faculty.
 
 
 
 
 
 
Michelle Mancenido works to establish and coordinate industry collaborations and projects for BS statistics majors.

 
 
 

Please describe your department’s industry and/or government engagements/collaborations.

Our primary involvement with industry is a semester-long senior capstone project.

We have three industry partners for this: a health care provider, a professional sports organization, and a global manufacturing company.

The capstone sites present the student team with a sufficiently diverse range of statistical applications to choose from. Students are mentored by faculty from statistics and applied mathematics, but they are ultimately responsible for scoping their projects, collecting data, coordinating with sponsors, formulating statistical models, coding, and—most importantly—presenting the results of their study to their sponsors. Mentors (one per team) are only expected to provide students with advice and help overcome stumbling blocks in the course of the project.

How did the engagement/collaborations come about? How are the collaborations sustained in a way that is mutually beneficial to students, the university, industry, and the community-at-large?

The collaborations are based on faculty network connections. The arrangement is mutually beneficial. For ASU, students are able to work and deal with real-world data that can sometimes be massive and dirty. For the sponsors, the students come in with skills that help them solve business problems they do not have the manpower or resources to tackle. It has also provided a pipeline for qualified job applicants.

The business problem tackled by the student team is selected in collaboration with our industry partners. Students are not simply handed data to analyze. They are active participants in the selection of the research project.

What arrangements about the collaborations might be helpful to share?

Each project is a semester-long contract. The contract is in the form of a project charter, which includes opportunity or problem statements, objectives, scope and limitations, and expected deliverables. The students develop these charters as part of their project management training. The students are also responsible and accountable for ensuring the delivery of quality output and meeting sponsor requirements such as deadlines.

What benefits for the department have you seen come out of these collaborations?

Our partnerships have directly led to jobs and internships within the industry partner. We also use an industry partner as an adjunct faculty to ensure the most current job skills. The capstone offers experience for junior faculty in business problems and the application of statistics (similar to a consulting center).

What benefits for the broader community have you seen come out of these collaborations?

In addition to the capstone, we partner with a local nonprofit conservation society. They have a great need for quality analysis, and—in turn—our students can get experience and help the local community at the same time.

What are the challenges or barriers to such collaborations? What have you learned since your department began these collaborations to deal with these challenges?

Always team dynamics. The students have a very tight timeline (four months), so they are pressured to deliver, which boils over when the dynamics are shaky to begin with. Hence, we add a team-building and leadership component to the course.

Another is the scoping of the project. For most of the project sites, it is a part-time job, so the students sometimes find themselves overwhelmed with meetings, onsite data collection, etc. The faculty mentors are instrumental in ensuring the deliverables and project timeline are feasible.

Company logistics and human resources are another hurdle. For example, our health care partner requires at least two months to get a student started in their system (HIPPA training, vaccinations, etc.)

Finally, we have developed an excellent relationship with our sponsors; as long as the students keep delivering high-quality output, they are willing to accommodate the students year-in and year-out.

Any other comments?

I know these projects can often be a lot of work, but they have been overwhelmingly rewarding for our students.