In today's fast-paced competitive economy, research projects that enable cutting-edge technology and innovative thinking are the name of the game. One company that has benefited from university-industry research and development projects is the Motorola Advanced Technology Center (MATC), a part of Motorola Labs that undertakes long-term research and short-term development.
The following practical tips are based on nearly 200 projects from 1991 to 2001 sponsored by MATC. The projects involved 25 universities in the United States, Europe, Asia and South America and spanned topics in manufacturing engineering, materials science, electronics engineering, mechanical engineering, computer science, optics and industrial design.
Ideally, the R&D project should involve direct participation of end-users of the technology. When this is not possible, a mentor in a corporate R&D lab can convey feedback from end users. Even when a university's research program does not exactly fit a company's current projects, there still is room for collaboration.
Project Initiation
Finding a suitable partner is the first step to initiate a university R&D project. To make this important contact, an industry sponsor can look to:
- ■ Former research advisers for graduate studies
- ■ Faculty at a university where a company funds a research center
- ■ University faculty or students at an alma mater
- ■ Professional friendships
- ■ People seeking internships or those on sabbaticals
- ■ Recommendations from National Science Foundation research centers
- ■ Papers in research journals, conference proceedings, other publications.
Two types of partnerships have led to the biggest “wins” at Motorola:
■ Collaboration with someone experienced in an area directly applicable to the company. The benefit is a shorter learning curve.
■ Established researchers seeking new applications for their technology—requiring lower infrastructure costs and the opportunity to co-author publications in new, interdisciplinary areas. In choosing a research topic, address the interests of senior management sponsors, the anticipated financial impact and the technology receiver within the company. Ideally, the R&D project should involve direct participation of end-users of the technology. When this is not possible, a mentor in a corporate R&D lab can convey feedback from end users.
Even when a university's research program does not exactly fit a company's current projects, there still is room for collaboration. If a proposed project matches the personal interests of R&D staff, they will have the incentive to provide guidance, find potential end users for the project deliverables and identify resources for the collaboration. Regardless of the potential impact of the research, without a technology receiver, a company is likely to put a project on the back burner and forget it.
Exhibit 1. The “Project in a Box” model highlights interfaces that determine if project scope will be manageable for the university and if project deliverables will be usable by the sponsor company.
Exhibit 2. University projects require a significant and consistent time commitment, but their net effect can be a large increase in the productivity of a company's R&D team.
Scope Variables
The scope of an industry-university R&D project can vary greatly depending on its duration, available resources, and project management (PM) formality. Project types include:
- ■ Direct grant to a professor
- ■ Company-designated funds provided through a research center
- ■ Center-designated funds from member companies in a research center
- ■ Shared grant from a government agency
- ■ Internship/sabbatical by professor
- ■ Internship/co-op assignment by a university student
- ■ Sponsorship of a graduate or undergraduate class project
- ■ Participation on a graduate student's thesis committee
- ■ Joint work on a paper without a formal funding commitment.
In defining the scope of an R&D project, consider the maturity of the technology. The Research phase includes preliminary concept exploration and work in fundamentally new areas. The Prove Application phase addresses prototyping and demonstrating a new concept's applicability to a company's products and/or processes. The Develop phase involves the design and creation of processes, hardware, software and materials to implement a new technology within the company. The Implement phase involves transferring technology to end users and documenting results.
Considering that most university grants are reviewed every one to two years, focus on one or two of the four phases. New research partnerships should start during the middle two phases.
The majority of professors, students and industry mentors work conscientiously to ensure their projects succeed. Nevertheless, because of the individual, creative nature of R&D work, a project may “fail” even before it starts. Human resources are a crucial factor in the early stages.
Communication in Execution
Managing communications is the most important part of managing university-industry R&D projects. Partnerships between groups that can visit each other in a one-day trip are the easiest to initiate and sustain, but phone calls and electronic collaboration (e-mail and shared workspaces) enable partnerships from any point on the globe. Building personal relationships through initial face-to-face visits is vital—in some cultures, it will take several visits before a working relationship can be established.
Differences in time zones can either help or hinder communication, depending on how a project is managed. To increase productivity and decrease cycle time, it is possible to coordinate hand-offs between teams in different time zones and achieve around-the-clock collaboration.
“When we were beta-testing a decision support system for a factory in Germany, most days finished with a phone conference,” recalls Peter Csaszar, former Ph.D. student at the University of Illinois at Chicago. “It was 1:00 a.m. in Chicago and 8:00 a.m. in Germany. I would go home for the night, and the engineers in the factory would test the latest beta release using the day's production data. Before they left work, they would send me a summary of their results. In this way, it was possible to fix small bugs in the algorithm and prepare some new code releases overnight.”
Contact at least once a week has been a key feature of most successful projects. If students provide a short summary of accomplishments and plan each week, then mentors can provide quick feedback, especially to solve technical problems together. The industry-university team should meet monthly to update the near-term milestones and review progress toward one to two “key deliverables” per student per year. “Each of our key deliverables is practical, not only from a Motorola point of view, but also considering our capability and our need for performing research we can publish,” explains Chi Zhou, Ph.D. student at the University of Illinois at Chicago.
Project Reporting
Twice a year, the mentor should prepare a project summary and brief the management sponsor on the status of the project. This is also a good time to take an inventory of results that could lead to a joint publication or conference presentation. Feedback should be given to the university team on not only the R&D results but also the PM processes. Short, structured performance dialogs give students a taste of the “real world” and help them develop PM skills.
When a project is reviewed and renewed at the end of the academic year, prepare (or update) a high-level four- to 10-page project plan, which should include a statement of work, summary of technologies to be employed, timeline with key milestones, list of personnel, and budget.
A key factor in the success of a project is how well its scope and requirements are defined. It is the responsibility of the industry mentor to clearly define them at the start of the project and quickly resolve questions that arise.
If the scope and requirements are adequately defined, it will be clear what to do, and the university team will have the freedom to creatively explore how to do it. Exhibit 1 shows what to consider when establishing the scope of a “Project in a Box” that can be taken “offline” to the university and, upon completion, be efficiently be brought back “online” within the company.
Exhibit 3. Some areas of PM are more important than others during the Research ®, Prove Application (PA), Develop (D), and Implement (I) phases of a university-industry R&D project. This chart was compiled from a decade of project histories at the MATC. The relative importance of nine areas of A Guide to the Project Management Body of Knowledge (PMBOK® Guide) has been estimated from interviews with Motorola mentors, based on the amount of time they have spent addressing each of these areas for research.
Standard PM practice is to state requirements in an explicit and testable manner. However, a somewhat different approach should be taken for collaborative university-industry R&D projects. The industry sponsor should identify a small set of “must haves” and define the corresponding explicit requirements for the university team. Implied requirements should cover the intended use of project deliverables.
Provide a considerable amount of open space for the university team to innovate and perform publishable research. As Linda Fitzpatrick of the Kellogg Graduate School of Management at Northwestern University, Evanston, Ill., USA, notes, “Motorola didn't give us a lot of information at the start, but by the end things worked fine. We recruited a team with a manufacturing background, and during the project learned a lot about Motorola products and company culture. Key deliverables were stated early, and we had effective teleconferences.”
Easy Integration
To facilitate the integration of project results with the processes and systems inside the sponsor company, system interfaces should be identified as early as possible. One of the most critical and time-consuming tasks for the mentor is to prepare test scenarios. Although a nondisclosure agreement between the University and company typically occurs, test data sets may require “scrubbing,” according to company policies for confidential information.
Expert industry mentors can help the university team survey the publicly-known state of the art and quickly identify applicable techniques. In cases where the company has an internal R&D effort, decide what details to share and how or when to share them to manage risks and reduce project costs. The company sponsor, however, may choose not to share internally developed results. Exhibit 1 highlights the fact that managing project scope plays a major role in managing the flow of technology.
Collaboration requires a significant time commitment from the industry mentor, not only for technical problem-solving but also for PM. According to Dan Gamota, a department manager at the Motorola Advanced Technology Center, “The output is directly proportional to the mentor's input, but sometimes the ‘ball is dropped’ by the mentor…Your commitment is to spend time with the professor and graduate students so that they understand the problem and are intimately aware of the issues.”
Based on the MATC's experience, teams function best when mentors dedicate at least four hours per week to each project. As shown in Exhibit 2, this will reduce the mentor's availability for internal projects, but will increase the overall productivity of his or her group. A full-time project engineer can effectively mentor about three students/projects and, thereby, more than double their overall productivity. In Motorola's experience, some project management areas are more important than others during the four phases of a university-industry R&D project (Exhibit 3). PM