Convergent engineering research centers (CERCs) need not be one-size-fits-all in their structure or in their operations, because each grand-challenge-like initiative requires its own unique team, partnerships, and methodologies for achieving success. At the same time, all CERCs should share certain fundamental characteristics, including agility, effective collaboration across institutional boundaries, flexibility, and the option to recompete for federal funding after 10 years, providing that rigorous performance criteria are met.
The continuing acceleration of research, development, and innovation in the global economy has been well documented,1 as have the impacts of this pace of change on research universities and industry. In order to reach their original milestones as well as strive toward aspirational goals, CERCs will need the freedom to thoughtfully adapt to changing circumstances, without the fear of being punished for taking calculated risks. Agility implies making modifications to the research agenda,2 staffing, and partnerships in order to take advantage of new opportunities; constant monitoring of and adaptation to the competition; and phasing out unproductive or outdated activities.
Agility has implications for budgeting, funding, and reporting. Agile management of CERCs will require dynamic budgeting to flexibly adjust to a center’s need to modify its program. Dedicated funds, included in or in addition to those provided in the main center award, may be necessary to permit centers to explore potential new research directions. One existing mechanism for providing such flexibility can be seen in the seed-funding opportunities offered to NSF’s Materials Research Science and Engineering Centers and the Energy Frontier Research Centers of the Department of Energy. Some multidisciplinary research center initiatives in other parts of the world, including Germany’s Collaborative Research Centers program and the Spokes program of Science Foundation Ireland, provide supplementary grants to help centers engage new industrial partners.3 Foreign centers have recognized the importance of being agile and adaptable in terms of addressing opportunities and barriers to translation, scale-up, and industrialization.4
1 See, for example, R. Kurzweil, 2001, “The Law of Accelerating Returns,” March 7, http://www.kurzweilai.net/the-law-of-accelerating-returns.
2 This must be done carefully, because graduate students are on a 4- to 5-year schedule where changes in research direction can be damaging.
3 E. O’Sullivan, 2016, “A Review of International Approaches to Center-based Multidisciplinary Engineering Research,” a paper commissioned for this study, available at https://www.nae.edu/Projects/147474.aspx.
While accountability to funding organizations is important, time spent by key personnel filling out forms and writing lengthy annual reports curtails their efforts on breakthrough research and can inhibit center agility.
FINDING 5-1: The fast pace of innovation and the production of new science knowledge can create new opportunities not necessarily envisioned at the time a center is established.
RECOMMENDATION 5-1: To keep pace with a rapidly evolving technological and economic landscape, National Science Foundation program managers should give the convergent engineering research centers director and leadership team the flexibility and authority to adapt their research plans and add and subtract partners as required, so long as they remain accountable for these changes.
The CERC’s vision will evolve over its NSF funding lifetime, as will the stakeholders and key players. Accordingly, center teams should consist of members who are complementary, synergistic, and adaptive, with a strong sense that attacking a problem from many (often orthogonal) perspectives can often unlock solutions that would not be available to a more homogeneous team.
Collaboration at multiple levels is necessary for successful convergence-based research, development, and innovation. The committee believes that most complex, societally relevant problems demand strong multidisciplinary teams spanning academia and industry. Increasingly, these teams will need to be global in scale, such as Singapore’s Campus for Research Excellence and Technological Enterprise (CREATE), which engages university partners in the United States, Europe, China, and Israel. Other new models of global collaboration are being developed—for example, within international teams pursuing the X-Prize. These and similar efforts aspire to leverage the entire global innovation economy to accomplish their missions.
A defining characteristic of CERCs is that they benefit from the fundamentals of team research, which means establishing structures and methods for deep and meaningful collaboration. Formal collaboration plans in proposals can help to ensure that needed infrastructure and processes are in place.5
Addressing grand challenge opportunities successfully requires deep, real-time collaboration to refine common research goals and strategies. This means almost continuous interaction, not just yearly meetings where participants give presentations. David Kelley, director of the Stanford University d.school, has made “radical collaboration” one of its foundational principles.6 Communication among partners is obviously critical, but regular in-person collaboration may be impossible for teams that are widely distributed geographically. Fortunately, the limitations of remote collaboration are being addressed by social media, as well as synchronous communication platforms such as those available with virtual reality,7 and others.8 Experts can be added to solve specific problems as needed. CERCs will need to optimize such platforms to accelerate research and development (R&D) and innovation from their global contributors.
The following question arises about how distributed a CERC should be: Is a fully virtual “center” consisting of a network of geographically distributed researchers collaborating electronically even feasible? Evidence suggests that innovation frequently comes when a core group of researchers with diverse backgrounds have frequent
5 National Research Council (NRC), 2015, Enhancing the Effectiveness of Team Science, The National Academies Press, Washington, D.C., p. 208.
7 See, for example, C. Zakrzewski, 2016, Virtual reality takes on the videoconference, Wall Street Journal, September 18, http://www.wsj.com/articles/virtual-reality-takes-on-the-videoconference-1474250761.
8 ETEC 510 contributors, “Synchronous and Asynchronous Communication:Tools for Collaboration,” ETEC 510, http://etec.ctlt.ubc.ca/510wiki/index.php?title=Synchronous_and_Asynchronous_Communication:Tools_for_Collaboration&oldid=57264, accessed August 7, 2016.
face-to-face interactions, both formal and informal.9 Collaboration over long distances can be successful when it is already established that the parties can work together, but the committee’s vision is that CERCs would have a common physical location where researchers can interact to promote team building and team “maintenance.” Thus, while frequent face-to-face interactions for geographically distributed centers may not be practical, centers should plan and budget for appropriate in-person interactions. CERC kick-off events and annual review meetings will benefit particularly from in-person meetings.
FINDING 5-2: Problems of global scale and convergence require continuous and deep collaboration.
RECOMMENDATION 5-2: The National Science Foundation should only support new convergent engineering research centers whose members exhibit the collaboration skills, experiences, and personal attitudes required to be successful. Each member must have a unique, complementary role to play and be excited about the opportunity to work collaboratively with their other teammates. The management processes used must support deep, real-time collaboration.
Center plans need to create faculty reward systems that support teams as they engage in deep collaboration. Future centers should facilitate frequent virtual interactions as well as frequent face-to-face meetings of team members at central facilities to promote communication and collaboration and avoid duplication of effort.
Collaboration with Industry
Collaboration with industry and venture capitalists will be essential to the success of CERCs. Current engineering research centers (ERCs) are advised by industry advisory boards (IABs), and these IABs are used to critique internally solicited research proposals for center initiatives. Engaging with potential industrial partners early in the proposal process would be optimal, but, realistically, at this stage companies will not be able to assess a center’s offerings in terms of knowledge, technology, and intellectual property (IP).10 Successful relationships with industry require engagement, trust, and an incubation process to develop meaningful commercial value.11,12 One way to encourage these outcomes is to involve industrial partners early in discussions about a CERC’s strategic direction and in the ongoing creation of commercial value propositions. The committee believes that formulation and presentation of well-constructed value propositions, as described in Box 2.5, will enhance industry involvement. Experienced industrial partners can serve as mentors for the center’s students and faculty to become future innovation leaders.
Different universities typically have their own IP policies and contracting processes. Before a CERC forms, the contracting officers of the participating universities will have to agree on a common policy. Different industries also have different IP policies. In the case of IP, each CERC will likely have to work out an approach that fits with its relevant industry and community partners. This can be difficult and often complicates collaboration. Nevertheless, other government agencies, such as the Defense Advanced Research Projects Agency (DARPA) and I-ARPA (Intelligence Advanced Research Projects Activity), have succeeded in facilitating the conditions required for deep collaboration.
FINDING 5-3: There are barriers related to issues such as IP rights and academic appointments that can hinder or prevent the meaningful collaboration among institutions required for center success.
9 L.R. Blenke, 2013, “The Role of Face-to-Face Interactions in the Success of Virtual Project Teams,” doctoral dissertation, Missouri University of Science and Technology, http://scholarsmine.mst.edu/cgi/viewcontent.cgi?article=3306&context=doctoral_dissertations.
10 Exceptions may arise when centers grow out of pre-existing collaborative university-industry research efforts.
11 S.C. Betts and M.D. Santoro, 2011, Somewhere between markets and hierarchies: Controlling industry university relationships for success, Academy of Strategic Management Journal 10(1):19-44.
12 M.D. Santoro and P.A. Saparito, 2003, The firm’s trust in its university partner as a key mediator in advancing knowledge and new technologies, IEEE Transactions on Engineering Management 50(3):362-373.
RECOMMENDATION 5-3: To receive funding, convergent engineering research centers should demonstrate clear plans that enable potential center partners in universities, industry, or other sectors to engage in open and transparent collaboration. Anyone, from any institution—international, government, university, nonprofit, or industry—who can contribute to the center’s mission should be able to join the team in a seamless way. Mechanisms should be in place from the outset to remove or minimize institutional, budget, intellectual property, export control, attribution of credit, or other barriers.
NSF’s insistence on submission of formal collaboration plans in CERC proposals will help to ensure that cross-institutional barriers are addressed from the outset.13
CERCs are an integral part of the overall U.S. science and technology enterprise. In today’s environment, however, NSF alone cannot be expected to meet either the financial or the intellectual requirements for this critical initiative. Cross-agency collaboration and cross-agency fertilization can help address this challenge.
Expertise located within other government agencies, national laboratories, nonprofits, and even citizen scientists can contribute to CERCs. In addition, the government collects a large amount of data (e.g., satellite imagery, economic and trade data, demographic data, environmental data) that can be valuable to specific research projects. However, there are often legal, budgetary, or institutional barriers to accessing this expertise or data in a timely way.
Interagency collaboration allows the federal government to leverage and integrate the skills, expertise, and interests of multiple agencies and their respective innovator communities. Such collaborations are an excellent way for the federal government to focus investments, pool resources, and create public awareness of engineering or technology challenges of societal importance. Recent interagency initiatives that have substantially engaged NSF include the National Nanotechnology Initiative (NNI), the Materials Genome Initiative (MGI), the Advanced Manufacturing Partnership (AMP), the Brain Research through Advancing Innovative Neurotechnologies (BRAIN) Initiative, and the National Robotic Initiative (NRI; see Box 5.1).
Another form of interagency collaboration involves the sharing of unique instruments and facilities available at core facilities, such as Department of Energy’s (DOE’s) national laboratories. Unique technical instrumentation and facilities can accelerate fundamental discoveries leading to new knowledge. Examples are the light and neutron sources located at the national laboratories of DOE’s Office of Science. CERCs will leverage unique instrumentation and facilities at different sites, both experimental and computational, to help achieve major research results. CERCs lacking local capabilities must reach out to entities in those regions that can provide the needed resources.
FINDING 5-4: Unique technical instrumentation and facilities can accelerate fundamental discoveries leading to new knowledge.
RECOMMENDATION 5-4: Whenever possible and appropriate, future centers should leverage unique instrumentation and facilities at different sites, both experimental and computational, to help achieve major research results.
FINDING 5-5: Center-based engineering research focused on grand-challenge-like problems of significant complexity and scale invites collaboration among multiple agencies.
There appears to be a growing consensus in foreign university-industry research centers that they should be networked and aligned; that is, collaborating with and leveraging the complementary capabilities and resources of other national, regional and international innovation actors.14
13 NRC, 2015, Enhancing the Effectiveness of Team Science, The National Academies Press, Washington, D.C. p. 208.
14 E. O’Sullivan, 2016, “A Review of International Approaches to Center-Based, Multidisciplinary Engineering Research,” paper commissioned for this study, available at https://www.nae.edu/Projects/147474.aspx.
RECOMMENDATION 5-5: Cognizant program staff at the National Science Foundation (NSF) should have responsibility for identifying and collaborating with other agencies’ programs that relate to the missions of NSF centers. Where appropriate, multiple agency financial, technical, and human resources should be leveraged to support the work of centers.
Current ERCs receive a maximum of 10 years of funding from NSF; during this time they are encouraged to develop strong relationships with industry, other universities, and government and nonprofit entities that can sustain them after that period. The defined period of support imposes a discipline on operations and helps to ensure that funding is available to a variety of different centers.
A 10-year program in the global innovation economy, where technology improves so rapidly, represents a major opportunity for creating significant advances. It is expected that CERCs will be able to demonstrate dramatic advances by the end of this period. However, given the scope and complexity of the grand-challenge-like problems to be addressed by CERCs, the extra time required for the formation and management of convergent research teams,15 and the comparatively low maturity level of the technologies explored by the ERCs (Figure 1.1), the full impact of a center’s research may not be apparent within 10 years.
Under those circumstances, provided sufficient progress against the goals of the center is being made, CERCs should have the option of competing for a renewal of NSF funding after the initial funding period is over. The
15 K.L. Hall, A.L Vogel, B.A Stipelman, D. Stokols, G. Morgan, and S. Gehlert, 2012, A four-phase model of transdisciplinary team-based research: Goals, team processes, and strategies, Translational Behavioral Medicine 2(4):415-430.
option for competitive renewal is common practice for large, center-based, multidisciplinary research centers at the international level and in many U.S. agencies.16
FINDING 5-6: Current ERCs receive a maximum of 10 years of funding from NSF. Although major advances should be expected of CERCs, the full impact of a CERC’s research may not be apparent after 10 years.
RECOMMENDATION 5-6: “Sun-setting” of National Science Foundation funding for convergent engineering research centers should not be automatic; the centers should be allowed to continue to compete for renewals, assuming rigorous performance criteria are met.
The committee believes that artificially cutting-off funding at a pre-determined date, without regard for consideration of performance or contributions being made, would be wasteful. Although many international center programs have traditionally been funded for lifetimes similar to NSF ERCs—that is, 5 years with the potential for one further 5-year funding period—several programs have recently extended center lifetimes.17