Chapter 4: University Perspective: Formulating, Operating, and Sustaining an ERC

Most of this History was based on the direct experiences and observations of its two authors—particularly those of Lynn Preston, a co-founder and longtime leader of the ERC Program at NSF, and to a much lesser extent those of Courtland Lewis, for many years the communications contractor to the Program. But the authors realized that the view of the ERC experience from the university side was very important and a necessary part of the History. To understand the challenges and lessons learned from formulating, operating, and sustaining ERCs from that perspective, and since neither of the authors was directly involved in those efforts, we turned to a few university administrators closely involved with several ongoing and graduated ERCs. We are grateful for their contributions to this chapter.  

4-A       Stephen W. Director: Thoughts on the ERC Program

The following was provided by Stephen W. Director, a pioneer in the field of electronic design automation and who has a long record of commitment to and innovation in engineering education and university administration. He served as a faculty member of the Engineering Design Research Center (EDRC) at Carnegie Mellon University (CMU) (Class of 1986). In 1982, he founded the SRC-CMU Research Center for Computer-Aided Design and served as its Director from 1982 to 1989. He also served as the Head of the CMU Department of Electrical and Computer Engineering, when the Data Storage Systems Center (DSSC) (Class of 1990) was formed with many leaders and faculty from that department. Later he served as the Dean of Engineering at CMU until his departure to serve as the Dean of Engineering at the University of Michigan (1996-2005).  There, he oversaw the operation of one of the two ERCs awarded to UM, (the ERC for Reconfigurable Manufacturing Systems (Class of 1996)) and proposal preparation and operation of the second one (the Wireless Integrated Microsystems Center (Class of 2000). In 2005 he moved on to Drexel University where he served as the Provost and Senior Vice President for Academic Affairs (2005–2008). Director then moved to Northeastern University where he served as the Provost and Senior Vice President for Academic Affairs from 2008 to 2015 until his retirement. At Northeastern, he provided overall guidance to the Dean of Engineering, Allen Soyster, and Michael Silevitch, the Director of the Center for Subsurface Sensing and Imaging Systems (CenSSIS).

4-A(a) Preamble

Hardly a college of engineering today would not claim that interdisciplinary research, team work, outreach, university/industry partnerships are central to their mission. But when the Engineering Research Center (ERC) Program began at NSF, there was a serious disconnect between academe and industry—i.e., between what academe produced and what the nation’s industries needed. At the graduate level, universities were increasingly producing “engineering scientists” in the mold of physicists and chemists, people who were best prepared to be faculty members. Undergraduate education had a strong “engineering science” bias and generally did not produce people who understood engineering systems or how to make things in industry.

4-A(b) What Changed All This?

In the early 1980s, for the first time in many decades, American companies were losing their economic dominance to foreign competitors. We excelled in scientific discovery but had difficulty in converting basic knowledge to competitive products. Our engineering students graduated with profound mastery of a narrowly specialized body of knowledge in a traditional discipline but could not always see the larger picture that was necessary for our industries to compete successfully.

The ERC Program was established to counter this status quo. However, some aspects of the NSF plan for ERCs clashed with the way university research was carried out at the time.

  • To begin with, the ERC concept ran counter to the traditional “Single-Investigator Research Project” culture of academic research.
  • The ERCs also challenged the traditional disciplinary structure of academic engineering, both intellectually and organizationally. Intellectually, their cross-disciplinary focus was a sharp departure from the specialized, compartmentalized disciplines around which engineering research and education were organized.
  • The ERCs’ close involvement with industry was a new phenomenon on many campuses, and it encountered stiff resistance. Engineering faculties had spent decades thinking that in order to “elevate” their stature on campus they needed to emulate the style employed by the physical science faculties, and there was a good deal of anxiety that this progress would be lost, with engineering faculties (in the extreme view) becoming potentially little more than industrial job shops.
  • There was also a belief that industrially oriented, cross-disciplinary research was inherently more superficial and less fundamental than single-investigator research in the traditional disciplines. This perception could lead to difficulties over the promotion and tenure of faculty who engaged in ERC research.
  • Consequently, the success of the ERC Program was not guaranteed. So why did the ERC Program succeed? What were the characteristics of the most successful centers? How did the ERC Program change universities?

4-A(c) Observations at Three ERC Host Universities

Here are my observations from the three universities that I have been associated with—as a PI, Dean, and Provost—that had ERCs.

i.      At Carnegie Mellon University

The ERCs that were developed and awarded to Carnegie Mellon University (CMU) did not just pop up out of nowhere. They were grounded in work that had already begun. The Engineering Design Research Center (EDRC) grew out of the Design Research Center (DRC), which had been founded 12 years before the first proposal to NSF’s ERC Program. Creation of the EDRC was driven by Herb Simon’s vision of design as a “science” and a team of faculty who shared that vision.

The Data Storage Systems Center (DSSC) grew out of a successful preexisting interdisciplinary multi-investigator activity, the Magnetics Technology Center (MTC), formed in 1983. The MTC was driven by a vision of what was needed to help the critically important data storage industry, which was feeling a very real threat from Japan.

Unlike many universities at the time, CMU had an established a culture of interdisciplinarity. Joint departmental appointments were in place long before they were instituted in other universities. The promotion and tenure process was broad and flexible enough to accommodate collaborative efforts. There was good alignment of university, college and departmental visions of research. There was also a history of strong industrial funding for research. Examples included:

CMU was willing to undertake research aimed at increasing competitiveness of U.S. industry in areas of great economic importance.

ii.            At the University of Michigan

The ERC for Reconfigurable Manufacturing Systems (RMS) grew out of a far-reaching vision that the Center’s founding director had about how to improve the efficiency of the manufacturing of consumer goods—a critical need in the US. The RMS ERC established a whole new field. By the time this ERC was created, the concept of second-generation centers had emerged, which meant that in addition to strong research, education, and industrial collaboration/technology transfer programs, centers were to become leaders in addressing two other important issues facing colleges of engineering: diversity and outreach.

At the time the RMS ERC was established in 1996, the University of Michigan’s (UM) College of Engineering still had a strong single-investigator bias. There was widespread skepticism about a center that could only succeed if faculty collaborated. Many engineering faculty were concerned about the “applied” nature of the Center’s research and the close association it had with industry, in spite of the fact that the College had a very successful Industry/University Research Center and previous funding from the SRC. In this regard, the impact of the ERC was immediate. A culture of interdisciplinary and collaborative research began to take hold as faculty saw the success of the ERC Program at large and the RMS in particular. There was also a marked increase in competition among faculty groups to submit proposals for a second ERC. By the time the second UM ERC (the Wireless Integrated Microsystems Center, or WIMS) came into existence in 2000, NSF had upped the ante again; now they wanted ERCs to take a lead in demonstrating that multiple universities could collaborate. Consequently, Michigan State and Michigan Tech became partners with UM in this endeavor. Multi-university collaboration exposed the College to a whole new level of complexity, requiring administrators from more than one university to figure out how to work together toward a common goal in spite of having very different administrative structures and cultures.

ERCs had a great impact on UM’s College of Engineering and the University. Interdisciplinary work became recognized as a “legitimate” role for faculty and Diversity and educational outreach activities became a college-wide activity. A more thoughtful approach to technology transfer evolved.

iii.          At Northeastern University

In 2000, Northeastern (NU) received its ERC, focused on sub-surface imaging, much earlier in its development as a “research university” than most other ERC host institutions had. Northeastern in general, and the College of Engineering in particular, had primarily been focused on undergraduate education but had started on a path to establish a significant graduate research program; the ERC helped lead the way in this regard. As with CMU and UM, a key to successfully winning—and keeping—an ERC was the determination, strong vision, patience and perseverance of the PIs and the commitment of the Dean. Northeastern took a different route to multi-university collaboration by involving the University of Puerto Rico–Mayagüez.

Winning an ERC had a huge impact on Northeastern:

  • It clearly demonstrated to the faculty that Northeastern could play in the “big leagues.”
  • It caused Northeastern to realize that a successful research program needed a real infrastructure to manage the research.
  • Northeastern’s relationship with industry changed. Previously, industry was looked to primarily as a provider of co-op jobs for Northeastern students. Industry now started to see NU as a research partner as well and began to send research dollars to Northeastern to support the ERC and other research programs.
  • Faculty became much more team oriented.
  • The ERC had a great impact on engineering education. For example, the Gordon Foundation donated $20M to Northeastern to establish the Gordon Engineering Leadership Program, which offered a new master’s degree aimed at increasing the pool of engineering leaders capable of initiating and driving large-scale projects and managing multi-disciplinary teams.

4-A(d) Overall Observations

ERC awards have had a substantial impact on the colleges of engineering that have hosted them:

  • These colleges have seen a significant increase in number of undergraduates engaged in research.
  • There has been an increase in the inclusion of industrial representatives on PhD committees.
  • They have legitimized “use-inspired fundamental research.”
  • They have stimulated a broader culture of interdisciplinary research.

Companies that have participated in ERC programs have also been impacted in important ways:

  • Specifically, these companies learn to appreciate the fundamental research efforts of a center as part of the longer-term systems perspective.
  • ERCs have provided well-trained employees for companies.

The ERC Program has been a training ground for academic leadership as well. A number of center directors have gone on to become department heads, deans, provosts and university presidents. I think this was an unanticipated outcome. ERCs impacted the broader university community as well, especially in terms of diversity and outreach programs.

Prerequisites for winning an ERC include:

  • A grand, system-level vision that is easy to explain and can excite others to rally around —an ERC is not a collection of disjointed projects
  • Strong leader(s) who is able build a team and but at the same time can share the credit for the vision and success
  • Having the potential of creating a new field, driving the field forward, and/or transforming industrial practice
  • A history of successful interdisciplinary and collaborative research.

What does it take to make an ERC work?

  • Alignment at all levels—department., college, university administrations
  • Ability to successfully collaborate across the boundaries of discipline-focused colleges—this is oftentimes a culture shock that sometimes takes strong leadership from the administration to make happen
  • Ability to successfully collaborate between universities—a definite culture change
  • Adequate space—collocated facilities are a big plus to facilitate collaboration
  • Good group chemistry—very important, as it can make or break a center
  • Understanding of the large administrative effort that is required to coordinate research programs across different departments, colleges, and universities
  • A solid, well-thought-out IP policy helps avoid problems down the line.

Benefits of an ERC to the University:

  • Stable funding source for long-term fundamental research vs. piecemeal project-based contracts/grants
    • Ten $300K/yr grants are not equal to one $3M/yr center grant
  • Support for a world-class infrastructure to support cutting-edge research activities
  • Attraction for industrial sponsors
  • Prestige to the college and university
  • Great visibility for student recruitment
  • Expanded undergraduate research opportunities
  • Engagement of students with industry

Benefits of an ERC to Industrial Partners:

  • Provides the opportunity to network with competitors and suppliers on neutral ground.
  • In case of federal funding, offers a way to leverage their research contribution
  • Provides a good source of new hires
  • Opportunity to try out new codes/equipment
  • Early look at new technologies and intellectual property.

There can be downsides to an ERC:

  • Not every faculty member likes to work in a center mode, or has a working style amenable to center-mode research.
  • Junior faculty can use an ERC as a basis to blaze new trails for which they can gain major credit, but they must exercise care not to use center support as a substitute for getting independent outside support.
  • NSF requirements for reports, site visits, K-12 outreach, and technology transfer activities place a heavy administrative burden on the leadership team and staff.
  • Industrial funding has drawbacks:
    • Start-up and maintenance efforts are large
    • Could require disproportionately large amounts of faculty and student time to maintain good relations with the company
    • Companies often encourage/favor applied, short-term results over basic research.
  • Heavy emphasis on sustainability of center after “graduation,” especially in the early years of a new center, can be a distraction and distort priorities. Strong centers will survive as a matter of course, even if the emphasis changes somewhat.” [1]

4-B       Allen Soyster Interview: A Dean’s Perspective

Additional insights were gained from an interview in 2019 with Allen Soyster, the Dean of Engineering at Northeastern University, regarding the submission, awarding, operating, and sustaining of the Center for Subsurface Sensing and Imaging Systems (CenSSIS). Soyster retired from Northeastern and the interview serves as a first-person summary of the challenges and rewards of preparing and submitting an ERC Proposal, and operating and sustaining an awarded ERC.

Dr. Allen Soyster arrived at Northeastern University (NEU) from Penn State University (where he had been a department head) as the Dean of Engineering in January 1997, at the same time as a new President, Richard Freeland. During his job interview, Dr. Soyster brought up the idea of applying for an NSF ERC, partly as a way to increase the visibility and stature of NEU, which is located in Boston and surrounded by several highly-ranked, world-class institutions. He believed that as a smaller, private university, NEU could provide better support, in relative terms, than a large public university like Penn State could.

Upon his return to Penn State, Soyster was contacted by NEU Prof. Michael Silevitch, who had been on the interview team, to support the ERC idea and propose an ERC application on “high-resolution sensing and imaging.” After Soyster came to NEU to serve as the Dean of Engineering, this application was submitted in 1998 (as a single-institution ERC, with no partners) and was unsuccessful.

Following that experience, Soyster took a strong role in planning a second attempt—this time, at Silevitch’s suggestion, adding subsurface sensing to give the proposal more of an engineered-systems focus: i.e., finding hidden objects. That new focus had a variety of potential applications in defense (e.g., land mines), medicine (e.g., tumors), homeland security (e.g., bombs in luggage), etc. Almost immediately, Raytheon Corp.’s CEO became engaged as a champion for the proposal; and the company, through its research staff, assisted greatly in shaping the research thrusts of the proposed center and in serving as a beacon for other industry partners.[2]

Soyster and Freeland together had strong connections with potential university partners for the NEU-based ERC at Boston University (BU); Rensselaer Polytechnic Institute (RPI), and the University of Puerto Rico-Mayaguez (UPRM). Industry partners were attracted from a wide variety of industrial sectors, and Silevitch was instrumental in establishing a hierarchy of membership levels.

President Freeland committed $500K for planning—a significant amount for NEU—out of which some funds were allocated to each of the proposed academic partner institutions. The planning was well-organized and intense. The pre-proposal was submitted and, among the 90 pre-proposals submitted, NEU’s proposal was one of the 15 invited to submit a full proposal.  To the delight of the team, the proposal was selected as one of seven for a site visit and a collective effort began to prepare for the visit. A total of five “red-team” rehearsal presentations were run before NSF arrived for its first site visit. Soyster was heavily involved with Silevitch in making sure the best possible effort was mounted in pursuit of what was now named the Center for Subsurface Sensing and Imaging Systems (CenSSIS). The effort was a success and CenSSIS was established as one of only two ERCs in the Class of 2000.

A large new research building had been built on campus just before Soyster’s arrival, and some space was still not officially allocated; so NEU allocated space in this building and others for CenSSIS’ administrative offices and research facilities.

Once CenSSIS was in operation, Soyster stayed closely involved with the Center and with ensuring its success and its smooth integration into the campus administratively. For example, Electrical Engineering (EE) was by far the largest department in the College of Engineering. The presence of the ERC brought prestige to NEU and helped attract several new high-caliber faculty, which benefitted the EE Department. But CenSSIS, largely an EE center, had become almost a separate department. A new head of EE arriving on campus felt his authority to be diminished by this situation, but Soyster reassured him so that it did not become a serious issue. In particular, Soyster always distributed new salary funds to the departments and let the department heads allocate it as they saw fit. He kept President Freeland abreast of the progress of the ERC and “went to bat” for Center faculty in the university-level promotion and tenure committees. Even in the early 2000s, this was necessary as team research and multi-authored papers often were not accorded the same weight as traditional single-authored work and papers.

Soyster instituted monthly telephone conference calls with the deans of engineering at the partner institutions, BU, RPI, and UPRM. Meeting agendas were sent out in advance, specifying what each school would be expected to report on. This practice continued for the first 4 to 5 years, to make sure that “they weren’t just walking off with the money and doing what they wanted.”

The relationship between Soyster and CenSSIS Director Silevitch remained essentially the same until Soyster left NEU in 2006 to come to NSF as the Director of the Engineering Education and Centers Division, in which the ERC Program is housed. That is, it involved frequent but not heavy-handed communication and oversight, and constructive advice. Silevitch’s forte was hiring the best people, both in research and staff. He had a strong interest in engineering education and was encouraged to emphasize that in the Center’s programs. Upper administration never intervened in Center affairs.

When asked, “Was there a conscious effort on your part to create and manage the university culture needed to support the Center? ”Soyster replied, “Very much so.” The awarding of the ERC was considered to be a great coup. President Freeland and Soyster convened a special meeting of the Board of Trustees to showcase it and subsequently held additional meetings to keep the ERC uppermost in the minds of Board members. The goal of NEU administration from Soyster on up was to ensure that the Center had what it needed to succeed. Those who did things to make NEU stand out academically were always supported. The Center’s presence in and of itself helped elevate NEU; when Soyster arrived, it was ranked 155th overall nationally; now (2019) it is ranked 44th by US News.[3]

Soyster concludes that a center cannot be successful unless the top leadership takes it on as “their” project. As an ERC proposal reviewer, he said, he would make sure that the President of the lead university is strongly in support of the effort.”[4]

4-C       Center Director Perspective

Dr. Brent Shanks, the Director of the Center for Biorenewable Chemicals (CBiRC) an ERC established in 2008, contributed his insights as follows.

  1. Proposal Choice:  As a PI, how did you settle on a vision and ensure that it was systems-focused, transformational, and not incremental?

There were two key conditions that were the basis for us developing a vision for an Engineering Research Center in biobased chemicals. The first was that biomass conversion was receiving a great deal of attention at Iowa State University (ISU), which was not surprising given our location and long history in agricultural product processing. Perhaps more important was the second condition in which we saw a key technological dislocation in the biomass conversion area. This dislocation had two components; 1) biofuels was the main focus of attention so researchers working on biobased chemicals were taking the same approach as that being used for biofuels and 2) the biological and chemical conversion communities were working independently to develop conversion technology. Therefore, we had a requisite combination of a technological need intersecting with key expertise residing at ISU, which created the conditions for a compelling vision.

The vision for CBiRC ultimately came from discussions between biological and chemical conversion researchers. Those discussions made us recast the technological paradigm for biobased chemicals. Prior to CBiRC, researchers were looking at how their technology could be used to produce biobased chemicals. The CBiRC paradigm asked what technologies were required to efficiently produce biobased chemicals, e.g., we went from a technology-led approach to systems approach derived from the products we were trying to produce.  This paradigm shift created a clear transformational vision, which now organized the technologies that needed to be developed for the biobased chemical industry. 

2. What did it take to form a faculty team across disciplines and universities and get their commitment to the ERC?

For CBiRC, the formation of the faculty team was driven by our vision for the center. Once the vision was in place, we could begin to see the technological expertise we needed to develop that vision. We reached out to specific individuals that we felt could bring complementary expertise to the effort. Importantly, we built out the initial team based on filling the gaps we had on the team coupled with personal connections with the researchers. We targeted researchers that we had reason to believe would be productive collaborators if they were part of a center. This approach meant that we recruited faculty by offering them a vision of what we were trying to accomplish and explaining how they fit into that vision. The basis for our initial researchers was faculty that worked in areas aligned with our vision so that they were not asked to take their research into a completely new direction. Instead they were asked to think of their research expertise through a different lens and slightly modify their approach.

3. Given your role, how many layers up through the university administration were involved in the following and what issues arose at those levels regarding:

  • Planning funds

We received a total of about $70k in planning funds, which were primarily used for administrative staff support and teaching buyout for the Principal Investigator (PI). The funds came from two colleges at ISU, Engineering and Agriculture, and an interdisciplinary program in biomass utilization. The funding was pieced together by the PI rather than centrally by the institution.

  • Proposal approval

Two ERC proposal went forward from ISU during the cycle that CBiRC was funded and there was no other internal competition.

  • Negotiating commitments for space and cost sharing

As mentioned in the previous response, two ERC proposals went forward and the PIs met together with the College of Engineering and the Office of the Provost to negotiate the cost sharing. Each of the proposing teams were promised $600k/y if they were selected by NSF. We were fortunate that at the time we were developing the proposal, the State of Iowa committed to fund a new interdisciplinary building called the Biorenewables Research Laboratory in which we were promised space if the ERC was funded.

  • Assuring support for staff (administrative manager, industrial liaison officer, education programs, etc.)

Given the many activities required of an ERC, we targeted to have sufficient funding from sources other than the base NSF ERC award to cover the cost of the support staff. We used the cost match from the university, industrial funds, and additionally grants for funding our support staff. Only our Administrative Director and one support staff were wholly funded from our university cost match. All other support staff was funded from multiple sources. Examples of other external funding include money from the State of Iowa to support our ILO teaching a graduate course in technology-led entrepreneurship, industry sponsored projects, SBIR/STTR funding to support our staff research engineer, and GK-12 and STEM-C grants to support our Precollege Education Director.

4. At the proposal stage, what was entailed in forming the industrial partnership and gaining their commitments to fund and participate if the award were to be made?

We leaned heavily on industrial relationships that were already in place prior to embarking on the ERC proposal. As mentioned above, ISU was already very active in the biomass utilization area, so we had pre-existing relationships with a number of companies. In the proposal process we met individually with the companies and presented our vision to them.

5. What was involved in supporting the writing of the proposal, the review process, and the site visit?

As mentioned above, we received planning funds that provided administrative support for the proposal preparation as well as release time for the PI. We also had a consultative phone call with the Director of an active NSF ERC. Prior to the site visit, we had a practice site visit that included several administrators and faculty members from ISU. We did not engage any other consultants either in the formulation of the proposal or the writing of the proposal.

6. Funded ERCs were instructed to have a budgetary account, into which flowed the NSF funds, cost sharing, industry cash support, other cash support, etc.  How did this requirement impact your university?

All funds that were used to support CBiRC activities were administered by the center, but these funds were not co-mingled into a single account. We had an overarching account for the NSF ERC funds that then had sub-accounts for the research and education activities. Separate accounts were established for all other funds. This approach was necessitated since NSF and sponsored projects were required to have university indirect costs included. The cost share from the university was a separate account that had no indirect charges. For industrial membership fees, we negotiated no indirect costs with the university, so it was a separate account as well. The account structure used by CBiRC was the same as used by other centers in the university.

7. Funded ERCs are expected to involve teams of faculty from different disciplines, combining engineering disciplines with science disciplines. Given these expectations for an interdisciplinary culture, how were the promotion and tenure practices at your university impacted?

All of the faculty in our center (senior and junior) had both CBiRC support and their own independent support with no extra credit for interdisciplinary work in the center. The operating assumption, which I think was largely validated, was that if the interdisciplinary efforts were valuable then the faculty member could further leverage that for their individual programs. Scholarly output from center-related research was counted the same as scholarly output from the individual faculty member labs. Realistically, funding for a faculty member from the center was not viewed as “competitively” awarded in the same way as if the faculty member was PI of a grant.

8. What university contributions were made to sustain the ERC post-graduation from NSF and for how long?

The university committed to provide base funding to CBiRC after the completion of the NSF ERC funding. The first year post-ERC funding was based on a similar funding level to our time as an ERC and subsequent university support will be tied to CBiRC’s ability to attract external funding. Additionally, biobased chemicals has been identified as a strategic area for the State of Iowa, so ISU is working to have direct support provided by the state in this area.

9. What other issues arose as a consequence of competing for and winning an ERC that would be valuable lessons for other university administrators and/or prospective ERC PIs?

An ERC is a different entity than a typical federal grant with different expectations about the activities of the center. Therefore, it is important that administrators are willing to work creatively with the ERC to knock down some institutional barriers. Our experience was that an ERC is sufficiently prestigious that the university wanted it to be successful. While some things took longer to address than preferred, we typically were able to find a positive path forward.

Examples of barriers that CBiRC needed to have discussed and addressed were unique staffing requirements for the center, the IP relationships between the university, center and center-originated startups, and the procedures and protocols for handling startup companies.

4-D       Operating the Center

4-D(a) Delivering on Commitments

i. Cost Sharing

Cost sharing for ERCs was a financial obligation, included in the proposal and agreed to in a cooperative agreement award. Provision of cost sharing was up to the university and subject to audit. Several center directors noted in their final reports that as the ERC advanced in age and the end of the agreement was in sight, it was more and more difficult to gain the full cost sharing commitment despite the documented agreements. 

ii. Overhead Return to the Center

Most ERCs were able to convince their University administrators to provide some overhead return to the center because of the size and prestige of the ERC award. Sometimes these agreements were renegotiated as the center received renewal funding and worked toward self-sufficiency.

iii. Subcontract payments to Partner Institutions

The Gen-2 and Gen-3 ERCs with partner institutions were more complex to administer than the single university ERCs prevalent in the Gen-1 ERC classes. They were also more complex for NSF to support, in terms of final allocation of funds to the center each year. In the early 2000s, the NSF personnel responsible for ERC cooperative agreements and cost accounting began to observe that some ERCs had substantial unexpended balances in their accounts at the end of an award year. Upon further investigation, it became clear that these ERCs were not paying their partner universities in a timely manner (sometimes due to large equipment delivery delays for which funds were encumbered but not yet paid). The NSF agreements personnel and ERC administrative personnel sent a notice to each ERC inquiring about unpaid balances. Those ERCs with unspent funds in excess of 20 percent of their annual budget were given notice to contact the partner institutions for bill-back invoices and to pay those invoices in a timely manner or their award level would be reduced.  All complied.

4-D(b) Enabling Cultural Change

ERCs are required to foster and enable a cultural change in engineering that manifests itself in greater teaming across faculty, collaborative and cross-disciplinary research, integration of research and education, and collaboration with industry in research and education. ERCs function as an integrating mechanism across the disciplines, as observed by Jim Solberg, the Director of two ERCs focused on manufacturing at Purdue University. He noted that an ERC should not function as a department, rather it should function by “straddling the technical disciplines and the permanent organizational structures of the university, without seeking to become one of them, (which) demands agility. Any ERC needs the enthusiastic cooperation of the departments, and therefore cannot compete with them or draw resources from them.” ..”With good management each of the faculty members saw the ERC as an asset that could contribute to their own objectives and achieve something worthwhile that they could not achieve on their own.[5] Solberg goes on to provide the following guidance on how to enable and protect the cultural change inherent in ERCs:

  • “Put firm policies in place requiring cross-disciplinary teamwork and do not support faculty who want to work alone, no matter how good their ideas. While faculty prefer to work alone, large-scale and complex problems cannot be tackled by faculty working alone.
  • Carry out planned turnover of projects, based upon a system of external review outside the site visit process to ward off a feeling of entitlement or diversion of funds to personal interests outside the ERC’s goals.
  • Actively seek involvement of younger faculty, carefully mentoring them so that ERC support would not be, or appear to be, an easier path than the traditional trajectory. Skepticism voiced by senior faculty takes several years to overcome.
  • Strong and visible support from the Dean of Engineering is essential as there may be envy, revenge generated by termination due to poor performance, and attempts at sabotage by department heads with agendas that don’t support the ERC. Establishing a niche where the ERC is comfortable in the traditional culture can take several years, during which the support of the Dean is essential.
  • Overall a successful change in culture to support the ERC takes patience and determination, do not underestimate the effort and establish firm policies early.”

4-D(c) Enabling Systems-level Research and Testbeds

Ken Wise, Director of the University of Michigan ERC, the Center for Wireless Integrated MicroSystems (WIMS), pointed out that strong administrative support is necessary if ERCs are going to successfully address their systems-level goals. “Academia is oriented strongly toward individual accomplishments, and faculty members accordingly have a strong inclination to work with their graduate students on individual projects. Such projects produce very worthwhile advances in narrowly focused areas  but may not go much beyond that. They do not often rise to the systems level nor do they normally involve real teamwork among a group of faculty members. As a result, (research) thrusts tend to collapse into worthwhile individual (but mostly non-interacting) projects. In contrast, testbeds cut across many projects and thrust areas; they not only are essential in terms of promoting faculty interactions, they force them. They require teamwork in which each project must set aside some of its own work to do the things needed to realize the system.  While even with testbeds doing systems is not easy, without them it would be impossible.”[6]

Preston observed over time that staff engineers are critical to assuring work at the systems level is enabled by testbeds. University administrators have to be ready to establish and even support positions for staff engineers who can build the systems testbeds. Graduate students and post-docs are temporary employees whose careers are dependent upon publications and the work of building a testbed normally does not support that activity. Staff engineers are required to take partially working prototype devices and components and build systems out of them. If the proposing or winning faculty have not planned for funds to support these staff engineers, it is imperative that the university administrators work out plans to support them with ERC funds and/or university funds.

4-D(d) Establishing Policy Boards to Facilitate the ERC

Most centers and their host universities establish policy advisory boards to support the cultural changes inherent in the ERC construct. For example, many ERC lead institutions established University Policy Advisory Boards, comprised of the Vice President for Research, the Dean of the Graduate School, the Dean and Associate Deans for Research and Education in the college of engineering, a senior university faculty member from outside the center and the Center Director. The Board oversees and ensures that the research, education and industrial collaboration and innovation functions of the ERC are carried out in a timely fashion and any needed new policy guidelines are developed and adopted.

Most ERCs have an Executive Committee, comprised of the Director, Associate Directors, and Administrative Manager, that assists the Director in the day-to-day management of the ERC.

4-D(e) Staffing the ERC

An ERC award comes with burdens that the administration should anticipate and support. These include:

  • Staff to deal with the myriad demands arising from the award, such as visitors, local government officials wanting to take credit for the award, people who want to carry out surveys, and NSF and others asking for input and performance data;
  • Early appointment of an Industrial Liaison Officer to serve as an intermediary between the faculty and industry personnel. There is no substitute for face-to-face contact between industry and faculty
  • Staff to support the reporting, data collection, and site visit functions of the ERC
  • Directors must be willing to take on a leadership and management role in the ERC. If he/she thinks that it is OK to turn all that responsibility over to staff so the Director can focus only on research, the ERC will most likely fail.

Overall, the ERC and the university(-ies) have to be flexible to enable staffing positions to grow and evolve with the requirements of the ERC Program and the development of the ERC This includes new types of position descriptions and salaries that reflect the complexity of these positions and the level of responsibility. Janice Brickley, Administrative Director of the ERC for Collaborative Adaptive Sensing of the Atmosphere (CASA) points to this administrative complexity and the need to understand the “big picture”… “administering this diffuse enterprise with multiple academic partners, outreach institutions, industrial partners, faculty, professional and classified staff, graduate and undergraduate students—across multiple time zones, cultures, and geographic areas.” This is both the challenge and reward of the position. (See ERC Best Practices Manual, Chapter 9, Sec. 9.1.[7])

She advises ERC administrative directors that their positions have the following complexity and demands:

  • Administrative coordination of Center activities and office management
  • Program Grant/Contract Administration and Compliance
  • Human Resources Management
  • Communication – primary point of contact internal/external constituencies
  • Information Technology – systems development, database design, and management of data
  • Accounting/Financial Management
  • Production of Annual Report
  • Conference and Events planning and management
  • Infrastructure and Facilities

4-D(e) “Managing” the New Center Director

                                             i.      Coaching/Supporting

Every successful ERC depends on constructive and consistent support from university administration, especially from the Dean of Engineering but continuing through the Provost, Vice President for Research, and the university’s President. Department chairs in Engineering, who in many cases tend to view the new center as a competitor for funds, space, and faculty attention, will, in the best situations, act as supportive allies of this new tide that will raise all the boats in Engineering if it succeeds.

An excellent example is that of Northeastern University. The relationship between the Dean and the new Center Director, both before and after the awarding of the ERC, which “…involved frequent but not heavy-handed communication and oversight, and constructive advice.” (See section 4-B.)

                                          ii.      Replacing Original Center Directors

It can be problematic to put faculty into the position of leading the ERC when there is a suspicion that he/she might leave for a higher position soon after the award is made, or when he/she is a poor team leader but has a strong technical reputation.  This happened with a few awarded ERCs. When these weaknesses prevailed, or the center director left the university or had to be replaced, it was not a situation without internal and external peril. The advice is to select someone who is already a good team leader regardless of age, with a strong vision that drives the ERC’s proposal preparation and team formation, and a strong commitment to leading and growing the ERC for several years. 

                                        iii.      What To Do When a Center Director Unexpectedly Dies?

Unfortunately, in the case of two ERCs, the original Center Director died early in the life of the ERC.  The essay below is an interesting story in how to deal with the situation.

From the perspective of Robert Swenson, then the Vice President for Research at Montana State University (MSU), the essay below chronicles the start of the ERC in 1990, the death of the first Center Director, William Characklis, in 1992, and the decisions made to go on:

“This proposal received outstanding reviews which recommended funding, but there was some controversy within NSF about placing an ERC out in the “sticks.” Partly to address the concerns, the MSU ERC was assigned a Program Officer (PO) to oversee the program who had considerable reservations and ran a tight ship. Unfortunately, Bill, seen as critical to the success of the ERC by the PO, died in June 1992. After several conversations with the PO, it seemed likely that he would recommend a one-year termination grant. However, we were able to get a meeting with the Assistant Director (head) of the Engineering Directorate at NSF to discuss the issues. During a positive discussion, Mike Malone (the President of MSU) and I convinced him to let us carry on—to replace Bill with another outstanding leader and to commit five new faculty lines to the center. He agreed to give us a new PO. On the flight home, Mike said, “Hell, let’s hire two senior people” (to replace Bill). Upon our return we met with Mark Emmert to discuss how we would fund two senior people to head up the ERC and the five positions we would commit to them. A few months later, in early 1993, we hired Bill Costerton as ERC Center Director and James Bryers from Duke as Director of Research. Thereafter, the future of the center was never in doubt.”[8]

From NSF’s perspective, MSU began a search for a new Director of the ERC and decided to appoint Bill Costerton, a microbiologist, but needed NSF’s approval. ERCs are not normally led by non-engineers, however, Costerton was one of the leading researchers in biofilms who also understood the engineering mindset from working with the ERC. Because of these characteristics, Marshall Lih, the Division Director at NSF, and Preston gave their approval. Costerton transferred to MSU in early 1993 to lead the Center and James Bryers, a professor of biochemical engineering from Duke University, was brought on board as Director of Research for the Center to strengthen the engineering leadership. In 1993, the ERC passed its third-year renewal review, its name was changed to the Center for Biofilm Engineering and an even more fruitful collaboration between engineers and microbiologists ensued, the results of which are discussed in Chapter 5-A(e) and the linked file “Role of ERCs in the Development of the Field of Bioengineering.”

4-D(f) Hiring New Faculty

Over time, ERCs impact faculty hiring. At the start, the culture of faculty hiring may be driven by the single investigator/single discipline culture that normally prevails at a university that is not accustomed to interdisciplinary research. However, as the ERC begins to function and the faculty see that they have the possibility to explore new research opportunities with longer-term research support and equipment opportunities they do not have in their own laboratories, cross-disciplinary teaming becomes more prevalent and new center proposals are submitted to other center programs. If the departments reward this type of work, the departments begin to see that if they hire faculty accustomed to or interested in working in teams and across disciplines, the opportunities for their faculty to participate in more grants open up. The ERCs have proven that there is a catalytic effect on this culture change from their creation, operation, and review.

4-D(g) Multi-university ERC Administrative Challenges

Janice Brickley, the Administrative Director of the Center for Collaborative Adaptive Sensing of the Atmosphere (CASA), faced significant challenges in setting up the administrative structure for this multi-university ERC in 2003. CASA was headquartered at the University of Massachusetts and functioned in partnership with Colorado State University, the University of Oklahoma, and the University of Puerto Rico Mayaguez, plus some outreach partnerships. As she trained future administrative directors in new and ongoing ERCs, she referred then to Chapter 9 of the ERC Best Practices Manual:[9]

  • There are additional stakeholders and layers with differing priorities, agendas, and institutional cultures.
  • There is an increased need to manage expectations when there are competing demands for resources, i.e., balancing the Center’s core work of producing research results and educating a diverse future engineering workforce with creating and maintaining the management and administrative infrastructure needed to accomplish both that work and NSF deliverables.
  • Process- and consensus-building takes more time, effort, and shepherding at all levels, but is critical to achieving the collegiality and cohesion needed to think and work as a Center, versus with an institution-specific mindset.
  • The cost of doing business (e.g., administration, operations, marketing expense) is higher and requires a larger percentage of funding. Managing a geographically distributed enterprise with multiple partners requires a more sophisticated administrative structure and additional resources.  For example:
  • Administrative overhead/infrastructure funds for administrative personnel, facilities, and information technology support are needed (at least part-time) at each campus.
  • There are substantial travel-related costs (hotels, airfare, and meals) for Center-wide events such as NSF Annual/Renewal Site Visits and IAB Meetings, the NSF ERC Annual Meeting in Washington, D.C., and periodic Center operating meetings such as Retreats and Executive Committee Meetings.[10]

In 2019, Preston asked 31 graduated ERCs if their multi-university partnerships were sustained post-graduation. Of those who responded, five ERC directors reported that these partnerships were sustained through support from government agencies.  They were:

4-D(h) Collecting Data

For a detailed discussion of data collection at ERCs, see section 9-D(b), “ERC Database of Performance Indicators.”

4-D(i) Dealing with NSF’s Financial Management Guidelines

ERCs are required to have budgets that are separate from the departments of the affiliated faculty. The budgets for direct support have to account for the flow of funds into the center from all sources as separate line items and for the payments flowing out of the center for faculty, students, staff, equipment, etc. to the lead university and its partner universities and outreach partners.  The status of funding must indicate the following:

  • Expenditures: amount paid, recorded in the University accounting system, reported on the Federal Cash Transaction Report (FCTR)
  • Obligations: encumbered, subcontract or purchase order – legally binding
  • Available: funds remaining for the ERC to utilize in future operations (residual).

NSF will monitor any large unexpended balances. The ERC’s program director recommends incremental funding for the next award year, based on technical progress.  The budgets imply dollars required to perform tasks each year to maintain technical progress. If there are funds under-expended in excess of 20 percent of the annual budget, NSF will determine if they are encumbered and if not, will reduce the budget. The budget also has to account for the amount of funds coming to the center from associated projects, where faculty receive the funding directly from industry or another non-ERC source and allocate a specific percent to support work in the center. 

For further details, see Chapter 9, section 9-E, “Program Budgets/Center Funding.”

4-D(j) Planning for Self-sufficiency

Graduated ERCs offer the following advice regarding planning for a self-sustaining ERC: Begin planning for self-sufficiency early. Formulate an effective business plan and get the upper administration to buy-in to the plan, to become a stakeholder and make a commitment to the ERC beyond NSF funding. This takes time and dedication, but it pays off.

4-E       Sustaining the Center Post-Graduation

4-E(a) Providing Transition Funding Support to the ERC

A few university administrations have provided up to $500K transition funding to support the transition from NSF support to self-sufficiency. But generally speaking, this type of financial support is unusual. ERCs are usually expected to “sink or swim” on their own; and with the extensive planning they do in advance of graduation from ERC Program funding, the great majority (more than 80%) succeed in becoming self-sufficient centers.

4-E(b) Factors Affecting Successful Transition to Self-Sufficiency

The findings of a survey of graduated ERCs found that transition strategies were different depending on the ERC and its core characteristics. Some became part of a broader research group on campus, some focused on their industrial partnership and functioned like an Industry/University Cooperative Research Center (I/UCRC) or were actually funded as a I/URCU, some focused on their core competencies in research and successfully competed for large center-level awards from other federal agencies or industry consortia such as SEMATECH or the SRC. A few chose a more complex strategy with the ERC core competence at the center of a “wheel” of smaller centers or large groups, a confederation of ERCs-type centers—like a sun with orbiting planets feeding off of the core sun.  However, of the ten centers that appear to be the most successful, the transition strategy was to continue as an ERC, and one of those two followed the confederation model.[11]

The following factors were found to have a significant impact on self-sufficiency:

  • “Broad involvement of faculty, staff, industrial partners and university administration in transition planning
  • Institutional factors – degree of university commitment, cross-disciplinary research and education
  • Education program sufficiently valued by faculty and students that it will be maintained
  • Commitment and interests of core group of faculty
  • Active industrial support and continuation of industrial membership and industrial advisory board guidance
  • Effective implementation of a realistic transition strategy that builds on enhances the center’s strengths
  • Quality of leadership of the management team. “[12]

In particular, it was found that successful transition to a fully functioning center on campus has the following characteristics:

  • 96% involved faculty in their transition planning
  • 72% involved the university administration
  • 64% involved their industrial partners, and
  • 60% involved the ERC staff.

These types of centers experienced the least reduction in annual funding, and several of them actually saw an increase in funding. These successful centers received transition funding and continued support from the university in the range of $50K to $1M per year.  Financial assistance ranged from direct cash contributions, to equipment donations, to allocation of research facilities, to administration staff and seed funding support, to graduate student support through waivers and fellowships, to allowing the center to retain all or part of the indirect cost recovery associated with all or some of its research grants, to cost sharing on new proposals put out by the center.[13]

4-E(c) Strengthening Infrastructure as a Catalyst for Further Collaboration

Several graduated ERCs had the following advice: Build an infrastructure that has unique capabilities. State-of-the-art equipment is critical in helping faculty secure extramural funding and can provide an additional revenue stream for the Center when industry pays fees to use the equipment for technical services or directed research.

4-E(d) Sustaining Cross-school Partnerships

Sustaining the partnerships between the ERC’s lead university and its partner institutions can be difficult in the absence of the ERC Program’s supporting funds; but a considerable number of ERCs manage to do it. One of the most productive types of cross-school partnerships is that between engineering and medical schools and partnerships. These partnerships were very strong between the Johns Hopkins School of Computer Science and Engineering and the Johns Hopkins Medical School through the lifetime of the CISST ERC. They continued post-graduation through support from NIH. 

At the CenSSIS ERC, there was a very strong partnership between Northeastern University, RPI, and Boston University with Massachusetts General Hospital; but it was not sustained post-graduation because the focus of the ERC shifted to homeland security.

4-E(e) Sustaining the ERC’s Education Programs

The three “legs” of an ERC are its research programs, industrial collaboration, and its education programs. Of the three, typically the most vulnerable post-graduation are the education programs, particularly pre-college education. This topic is discussed in the Education chapter, section 7-D(e).

One sign of the esteem in which successful ERC Education Directors are held is the frequency with which, after the center’s graduation, they are promoted to higher levels of responsibility in the Dean’s office or higher administration. From those positions they can often assist the center in continuing to fund its education programs. Some examples are:

Dr. Leyla Conrad was the Associate Director for Education and Outreach Programs of the Microsystems Packaging Research Center (PRC), at Georgia Tech. At present, she is the Outreach Director leading education and diversity programs for Georgia Tech’s School of Electrical & Computer Engineering, aiming to increase the number of incoming undergraduate students.

Dr. Anne Donnelly was the Education and Outreach Director of the Particle Engineering Research Center (PERC) at the University of Florida. Currently she is the Director of the Center for Undergraduate Research at UF.

Dr. Gary May was the Education Director of the Packaging Research Center at Georgia Tech. Following its graduation, he became the Dean of Engineering at Georgia Tech, and was later named the Chancellor of the University of California at Davis.

Dr. Claire Duggan has been the K-12 Outreach Director at the Bernard M. Gordon Center for Subsurface Sensing and Imaging Systems (CenSSIS), based at Northeastern University (NU), since its inception in 2000. Since 2008, she has also been the Director for Programs and Operations at NU’s Center for STEM (Science, Technology, Engineering and Mathematics) Education.

[1] Director, Stephen W. (2019). “Thoughts on the ERC Program.” Personal communication contributed, with permission, at the request of Lynn Preston. February 2019.

[2]  Underlying the research planning was the ultimate question of whether there could be a “unified theory” of how to detect hidden objects. That search has remained elusive, although there are commonalities across the different domains of earth, tissue, etc.


[4] Court Lewis interview with Dr. Allen Soyster, January 15, 2019. Used with permission.

[5] Solberg, James J. (1999). Center for Collaborative Manufacturing: Final Report, December 1999. Purdue University, West Lafayette, IN. p. 5

[6] Wise, Kensal (2010). Center for Wireless Integrated MicroSystems (WIMS), an Engineering Research Center: Final Report. The University of Michigan, Ann Arbor, MI. p. 124.


[8], October 14, 2014.

[9] Brickley, Janice (2011). 2011 ERC AD Training, Part IV: AD Perspective, slide 16.


[11] Summarized from: Williams, James E. Jr., and Courtland S. Lewis (2010). Post-Graduation Status of National Science Foundation Engineering Research Centers, Report of a Survey of Graduated ERCs. Melbourne FL, SciTech Communications LLC, January 2010, pp. 14-15.

[12] Ibid., p. 17

[13] Ibid., p. 19