Delegation and staffing during the life cycle of an ERC is an issue of fundamental importance. The related questions of how much to delegate, what management and operations functions to delegate, and how best to accomplish this distribution of responsibilities should be addressed in the planning stages of the ERC and on into the initial stages of funding and implementation of the ERC proposal.

The structure of the center also bears directly on how well its participants and its various research thrusts will interact to achieve the cross-disciplinary flavor inherent in a successful ERC.

2.3.1 Director's Choice: How Much To Delegate?

When an ERC is funded by NSF it is probable that there will have been a personal moving force who has initiated the application and gathered the research team. It is equally probable that the initiator will have a large and well-funded research group, or it is unlikely that the application would have been successful. But it is apparent that the duties of the Director of an ERC are sufficiently challenging that they are very difficult to combine with those of a successful and busy PI or research group leader unless the person concerned is adept at delegation. For this reason the first, and one of the most important, choices that a founding Director will make will be the extent to which s/he delegates responsibilities within the center as it begins its progress towards its first date with destiny at the end of Year 3. The results of our survey clearly indicate that Directors of ERCs form a spectrum in this matter, ranging from the "one-man band" to the relaxed "master delegator." Some have gradually increased the delegation of responsibilities as the activities have become more routinized and the ERC more "settled." If the Director delegates too little, s/he risks eventual "burn-out" and the loss of his/her own research program and even the center itself. If s/he delegates too much, s/he is likely to lose control of the center and jeopardize its ultimate success. There clearly are some instances in which a strong and well-balanced group of researchers persuades a faculty member with experience in administration to assume the directorship of a nascent center. These centers start with an advantage because the new Director will probably already know how to delegate effectively, but it is only advantageous if the Director is or can become the intellectual leader as well.

The founding Director should assess the importance of all the potential roles within the center and decide which to delegate and which to retain. Three major factors must govern the Director's choice of areas in which to concentrate attention: the peculiar strengths of his/her center team, the overall interests of the center and, finally, the meshing of his/her own research interests with the welfare of the center. The results of our survey indicate that 7 of 12 Directors had effectively sacrificed their personal research program in favor of the interests of the center, while 3 had modified their research programs and 2 had maintained them virtually intact. One reported that he had actually increased slightly the number of graduate students he supervises and had broadened the scope of his research due to exposure to the center's broad range of research. Some Directors said that they have been able to maintain the momentum of their personal programs by an increased dependence on postdoctoral fellows. Several reported that their research, not surprisingly, had become more cross-disciplinary, which helped sustain it. In any case, the Director's research must fit integrally within the scope of the center's research or it may be seen as a conflict of interest and a threat to the cohesiveness of the center. Such conflicts are viewed as serious by NSF, and must be resolved quickly.

2.3.2 Director's Choice: What To Delegate?

If the Director chooses to retain control of administrative and routine personnel matters, s/he will drown in details as the center grows to contain about 100 people. If the Director retains direct, personal control of financial matters, s/he can use this control to steer the center in detail, but s/he will be held responsible for every fiduciary ripple and s/he will encounter resentment when support is withheld or withdrawn. Experience teaches that, given the efficiency of sole autocratic command, a researcher can control and steer a research group of about 40 with some help from experienced staff and postdoctoral fellows; a few ERCs are "Mom & Pop" centers with a Director, a capable Assistant Director or Administrative Director with broad operational responsibilities, and several thrust leaders. But an integrated center with numerous faculty members and dozens of students is simply too large for one person to manage effectively. The sole exception occurs when a center becomes an "on-campus granting agency" and parcels out NSF funds and industrial money to fully independent PIs who run their own labs in the time-honored way, while the Director runs his/her own operation and acts as the reporting center. However, this approach is contrary to NSF policy and subverts the purpose of the ERC Program; such a center will not last long.

Over time, it has become apparent that the ERC Director can be aided in this complex role by a deputy executive officer who shares some of the leadership and management responsibilities in the ERC, in a manner that complements the leadership style of the Director. In recognition of this fact, NSF now requires all new ERCs to propose a Deputy Director. The Directors of existing ERCs without a Deputy Director or the equivalent should give careful consideration to hiring a Deputy.

2.3.2.1 Administrative Areas Within the ERC

Because an ERC with NSF funding and average industrial, state, and university support constitutes a roughly $10 million enterprise, at least a basic administrative structure is mandatory for effective organizational and financial responsibility. (It is worth noting that two ERCs have encountered serious problems of financial mismanagement that could have been avoided by sound, professional fiscal control; see Chapter 6, Section 6.4.) Since the Director has, by necessity or by choice, probably delegated much of the administrative responsibility to trained specialists, the center will be shaped by the choices that are made in setting up this structure.

All ERCs have an Administrative Director (AD), or the equivalent, who is responsible for general management of the day-to-day operations of the center. The Director and the AD typically work very closely together; the smaller the administrative staff, the more this tends to be the case. The position of AD requires a strong generalist, and selection of the right person is critical. These individuals play a key role in the overall success of the centers. It is essential that the AD understand fully the vision of the center, its ideals, and its intended impact, and that s/he be treated as a partner in bringing them to fruition. The AD accepts the responsibility of implementing the center's vision in a manner acceptable to the university's bureaucracy. There must be a mutual respect, with the Director articulating the concepts and ensuring buy-in and the AD providing a reality check on what is possible and identifying ways to implement the concepts. (See Chapter 6 for a full discussion of this function.)

It must be noted, however, that the Director is ultimately responsible for any administrative lapses that may occur; consequently, it is important to maintain supervisory oversight and control of office management functions. One potential danger inherent in the small "Mom & Pop" center structure is that the Director may delegate too much authority to the AD. The Director and AD may then mutually reinforce each other's belief that "everything's fine," even if administrative problems have begun to appear. Such a situation actually did arise at two well-established ERCs, leading to the departure of the Directors.

The decision to hire specialists for other functions will affect the center in various ways. If an assertive accountant is hired, the finances of the center will be well managed; but at an extreme, account management may not be as flexible as the Director needs it to be to accommodate frequent changes in the strategic plan and new developments that demand shifts in resources. If a computer specialist is hired, the center will have excellent data handling, good communications, and an attractive image on its home page. If the industrial interface is handled by administrative staff on a part-time basis, the Director will be the de facto salesman for the center. On the other hand, if an aggressive industrial liaison officer and technology transfer specialist is hired who can convince the PIs to buy into the industrial interface, the industrial interface will burgeon and there will likely be a strong technical connection with industry through the PIs. If the details and the policy of interdisciplinary education in the center are managed by a part-time education committee, students will be trained in their home departments and assembled in the center for occasional seminars and NSF site visits. But if an imaginative and personable education specialist is appointed, the students will make their home and form their friendships and loyalties in the center--and may be able to fund their own education with competitive scholarships.

The administrative infrastructures of centers thus range from a few people gathered tightly around the Director to small armies of specialists working for the benefit of the center, and each choice that is made will affect the kind of center that will emerge at the critical third- and sixth-year review milestones. Also critical is the decision of what structure to use for reporting and supervision (discussed in Section 2.3.2.3, "Mechanisms of Delegation"). The choices made in setting up the infrastructure of the center are matters of policy, and not of financial expediency, because at least two of the key responsibilities (technology transfer and education) may become largely self-funding. The administrative structure of the center must be set up thoughtfully by the Director, who must ensure that all major policy matters remain firmly under the control of center leadership and are complementary to the primary objectives of the center--industrially relevant cutting edge research and team-based interdisciplinary education.

The size of the administrative teams that run the daily affairs of ERCs varies between 3 and 19,² approximately. The Director who chooses the smaller, "tighter" option will retain more control over the operations of the center, because s/he can readily meet with the key people and set policy through simple conversation. The expanded team will offer greater possibilities for growth for the center, and for participation by more of its members, but the Director will probably have to relinquish some of his/her personal control and may be forced to appoint an overall manager from among this large group of center employees--or, preferably, to bring in a Deputy Director or Executive Officer. Administrative teams are easy to expand but difficult to contract; like many well-established units in industry and government, they often have an effective life of their own and a strong instinct for self-perpetuation. For this reason, the establishment of a large team constitutes a commitment to the center, on the part of the Director and the university administration. This commitment itself may drive the conversion of the ERC to an ongoing research/education entity after NSF support is terminated.

2.3.2.2 Program Areas Within the ERC

The three pillars of the ERC Program are research, education, and technology transfer. However, it is clear that the first (research) is a sine qua non, in that there are no educational or technological advantages to be gained from research if it is not outstanding. Also, all ERC Directors quickly come to understand that NSF site visit teams have been instructed by NSF to view research as the first preference "gate" when assessing the extent to which an individual ERC has succeeded in its mission.

If an ERC is to be a "better idea" center, rather than a "where do we get the money?" center, there must be a clear linkage between the strategic plan and research management and another linkage between research management and the management of finance and of education. In other words, decisions about the investment of center resources in specific research projects and in the support of students in specific research areas must be guided by a strategic plan in which the center is united. When that critical time of each year rolls around when decisions have to be made about how resources should be allocated to the various thrust areas, the Director will find himself in a situation that will be dictated by choices s/he has made at the outset. Either there will be a clearly stated strategic plan that makes the finance committee's job possible, or there will be a struggle for funds and the Director will have to make all of the final decisions. If there is a clearly stated strategic plan, the Director should be vigilant to discern the real authorship of key inputs to that plan. The strategic plan of a center can be manipulated by a small group of faculty with preconceived notions of what direction they want the center's research to take or, at an extreme, by a single strong personality--often the Director--who simply tells the troops that this is what s/he has decided. The smaller the coterie of influential insiders, the more NSF money there is for each individual in that group. But NSF Program Directors and site visitors can detect such a situation fairly easily and will not tolerate it. ERCs have failed to win renewal because their research program lost its cohesiveness and collapsed into a collection of loosely connected single-investigator projects. Effective strategic planning can prevent this tendency toward centralized self-interest.

2.3.2.2.1 Research Management

Chapter 3 deals specifically and in detail with research management in an ERC. However, management of this activity is central to the overall management and direction of an ERC and impinges on the success of every other area of center activity. The discussion here addresses research management in this broad context.

Research in an ERC is inspired and directed by the center's vision, as articulated by the Director and as supported by its members. The practical vehicle for the realization of this vision is the strategic plan, and the mechanism for its execution is the structure of thrust areas and testbeds found in most ERCs. As was described in Section 2.2, the Directors of most ERCs maintain firm control of the center-level strategic planning process; most decisions in research management are made by these Directors with the advice of a small inner circle of senior center researchers who comprise an executive committee. Several respondents regretted the lack of involvement of junior faculty in both the strategic planning and the research management processes, but cited difficulties in involving large numbers of people in these decisions. In a few cases, periodic retreats or cybersessions for the input of ideas have been employed as a means of involving more center members, and these have been proven highly effective in facilitating the development of a "center" perspective. But generally the responsibility for the planning and management of research remains centralized. At the detail level, however, junior faculty and students are more often involved in setting goals and milestones.

The most common and efficient apparatus for research management appears to be the appointment, by the Director, of a limited number (no more than five) of thrust area leaders who then join a few senior faculty colleagues to form a research steering committee. This "closed" structure is inherently efficient, in terms of both reporting and decision-making, but it can stifle the growth and renewal of the center because it can be very difficult for new people and new research ideas to break into this exclusive club.

Several centers cite difficulties in closing down existing projects. Most report that they depend ultimately on the Director to make these hard decisions, but such ERCs have closed research management structures that may require that the research committee vote against one of its own members in order to accomplish this essential revitalization--a process that is inherently conservative. A number of centers take industrial input (usually obtained at IAB meetings) on the relevance of projects heavily into account in deciding whether to continue them. Site visit recommendations are another source of input. Involvement of the IAB is said to be critical, along with the objective judgment of the Director, in overcoming the tendency of the well-entrenched "old boy's club" to maintain the status quo. One Director notes that these decisions are not made overnight; first, the need for action becomes generally recognized and then action is taken. It is easier to terminate unsuccessful lines of research if there is a detailed strategic plan with milestones; this makes it apparent when a project is going nowhere and/or no longer fits within the strategic plan. Another Director stresses that, "To maintain a high level of morale and cooperation, every attempt is made to discontinue projects with fairness and dignity." Open channels of communication, with emphasis on the ERC research as a team effort, help to soften the blow. Most ERCs provide support to the graduate student(s) involved for at least one semester after termination and try to accommodate them within other ongoing projects. Some centers use a scientific advisory committee to provide overall input regarding the quality of the projects, their relevance to the strategic plan, and the value they add to the state of the art in the field.

It is important to begin with the right number of research thrust areas. The "right" number may differ from center to center and field to field, and may also change across time. However, in general the fewer the research thrust areas, the easier it is to manage the research program (see case study).

CASE STUDY: One of the early challenges faced by the Center for Advanced Electronic Materials Processing, at North Carolina State University, was the organizational structure of the research program. Because there were so many research areas of interest to the faculty, all of which could be justifiably incorporated into the center, the Center began with ten thrust areas. However, experience indicated that, in an ERC carrying out collaborative, focused research, the Director and thrust area leaders should meet weekly to monitor the progress of the Center and to make strategic decisions as to required changes and new directions. It rapidly became evident that this was impossible to do because there were too many thrust area leaders and the program was too fragmented. Yet the various faculty members who were serving as thrust area leaders were reluctant to give up their participation in that capacity. Consequently, the Director was faced with the challenge of restructuring the Center and reducing the number of thrust areas (and thus reducing the number of thrust area leaders) without alienating the faculty who were needed to carry out the research agenda.

The first step was to talk with each faculty participant and convince them of the need to reduce the number of thrust areas. The alternative was to continue to struggle with too many areas and face the seemingly impossible task of giving the program focus. After the groundwork was laid for reducing the number of thrust areas, the Director next solicited the help of the Industrial Advisory Board by having them endorse and strongly support the proposal to restructure. With the help of the IAB and with the insight gained from individual discussions with the ten original thrust area leaders, the Director was able to restructure the Center in a rational way that had the support of essentially all of the participating faculty. The result was a much more manageable thrust area organizational structure which soon demonstrated its advantages for carrying out a well-focused, successful research program. The success of the new organizational structure gave faculty in the Center confidence in the Center's management and the effectiveness of the decision-making process. That confidence, in turn, made it relatively easy in subsequent years to again restructure the program as needed to enhance the Center's ability to achieve its goal. It also made it possible for the Director to terminate ineffective or non-productive projects with a minimum of turmoil.

One of the keys to the ongoing success of the ERC Program is the commitment of faculty to work together as a team to fulfill the research and educational mission of the center. This is more likely to happen when the faculty recognize the importance of collaboration and if they have confidence in the Director and in the process by which decisions are reached. Thus, it is essential that center management have credibility in the eyes of the faculty.

To facilitate the growth of an ERC and the realization of its vision, the Director should set up effective mechanisms for the intake of new people with new ideas. The survey of Directors clearly shows that center-run grant competitions are a poor mechanism for this intake; for the most part, such projects end up not aligning very well with the overall mission of the center. Because most Center Directors have actually sacrificed or heavily modified their personal research programs in favor of the center's interests, these people have both the motivation and the detachment necessary to welcome new talent into their ERCs. In most cases the Director is well advised to make the intake of new people, who are necessary for the realization of the center's vision, a very personal affair. They can be scouted, on campus or by recruitment, and their work can be steered toward the center's interests even before they are introduced to the ERC. They can then be invited to give seminars in the center's programs and/or invited to ERC retreats, to see if their fresh ideas strike a responsive note with the center as a whole. If so, they can then be brought into the center, where their presence and their new ideas allow the center to breathe fresh air. (See Section 2.4, "Recruiting.")

An ERC is an excellent power base because it represents a large amount of research money, and it will attract those who are interested in wielding financial power. The Director of an ERC must make a choice as to whether s/he will become the sole power broker, the leader of a small and select coterie of power brokers, or the arbiter of power who balances the process for the good of the center. A lesson learned from the management of research centers is that even the most promising center, founded on the most talented team of researchers, needs a constant flux of new people and new ideas to keep its edge. All centers try to stay ahead of the curve by recruiting excellent graduate students and postdocs, but very few give newly recruited faculty members senior positions with real access to center resources (especially if they are from outside the university). It is clearly NSF's intent that each individual ERC should continue as a research/ education/ technology transfer entity long after its support is terminated. The ERCs that we build must be built to last. For this reason, the Director of an ERC must resolve to build an effective faculty intake mechanism into the center (including the possibility of recruiting beyond the home university), select the new team members with exquisite care, and choose research management structures that allow the newcomers to share power and resources on an equal footing with all other participants.

It may be useful for the Director of an ERC to remain somewhat removed from matters of research management through the appointment of an Associate Director for Research and/or a research steering committee. This delegation of responsibility for the day-to-day management of research, and for the research reporting function that looms each year in the ERC Program, can leave the Director free to serve as custodian of the vision and as the chief arbitrator of the strategic plan. The research apparatus of the ERC is directed by the strategic plan. For this reason the strategic planning process is the ideal level at which new ideas should be introduced into the center's research. If the full membership of the center, including new faculty recruits and peripheral members who would like to become more integrated, have the opportunity to make meaningful contributions at the strategic planning level, the center can become a living, evolving community.

Rather than making all research management decisions personally, an ERC Director may find it more useful to maintain the vitality of the center by making sure that all ideas that serve the vision have an equal chance of implementation. It is certainly not the intent of the ERC Program to provide 11 years of high-level funding to a Director and an unchanging group of researchers, however capable and even brilliant they may be. Thus, a new Director must decide whether s/he will retain complete personal control over research management, set up a closed system of research management involving a select group of insiders, or augment the closed system with strategic planning and revitalization mechanisms that involve the whole center. Perhaps there is no choice to be made by a new Director, in the initial stages of the organization of an ERC, that will affect the center more than this pivotal decision.

2.3.2.2.2 Education

One of the three pillars of the ERC Program, education is the element with which most centers feel that they have had the greatest success. This success may reflect the national need for education of interdisciplinary team-oriented PhDs, more than the effective policies and programs implemented by individual ERC Directors; but in any case it is a very fertile area that may come to dominate the future of individual ERCs and even of the ERC Program itself. In recent years, the National Academies of Science and Engineering and NSF have joined with other professional groups to rethink engineering education at all levels.³ The impetus for this initiative is their collective alarm at the continued production of graduates at all levels who are totally untrained in team research and often openly antagonistic to industrially relevant research. The solution to this pressing problem is inherent in the ERC Program. The Program has stressed industrially oriented, interdisciplinary team research from its inception, and it has a great deal to teach the engineering science community about this urgently needed revolution in university education. Because educational policies take so long to come to fruition, it is probably wise for the Director of a new ERC to study the history of education in the Program and to make choices early in the life of the ERC that will allow this natural process to happen.

Chapter 4 presents a full treatment of ERC education programs and issues across the board. The discussion here will be limited to education issues from the standpoint of overall center management and direction.

With one partial exception, ERCs are not degree-granting academic entities. The basic responsibility for education rests with the traditional university departments; the most that an ERC can offer to the undergraduate or graduate student is an enhancement of their education. Traditional university departments are much like centrifuges, in that they apply very small force vectors to faculty and students whose interests lie near the center of the discipline that they are studying. The traditional departmental culture tends to favor students who are prepared to study classical subjects. Conversely, very large force vectors often are applied to students and faculty who venture too close to the boundaries between their nominal field and a neighboring area. The ERC culture is a boon and a refuge for these peripheral folk because it offers facilities for interdisciplinary research and the opportunity to work on research teams with people with widely different backgrounds. However, the fact remains that the ERC does not grant degrees, so that both students and faculty must be capable practitioners of their specific engineering or scientific field in order to find an intellectual base and to attain membership in a supportive community. An ERC Director must decide, very early in the life of the center, how much to coordinate with the departmental graduate programs. This coordination may be detailed, such as scheduling seminars to avoid department/ERC conflicts, or broad, such as establishing the proper balance in competing for the department's pool of graduate students; but in any case the spirit is more important than the details, and good relations are important.

Education at all levels is a lot like gardening, in that it is labor intensive and requires great patience and commitment. Hence, it is an area in which an ERC Director must delegate. There are decisions that must be made early in the life of the center that will have a considerable influence on the education program. The first of these issues is how hard the Director will press the university for contiguous space, which is discussed in more detail in Section 2.5.1.1, but which is an issue of significance for the center's education program. All Directors agree that contiguous space is highly desirable, but they disagree on the extent to which the difficulty of an ERC's being distributed throughout a university campus poses a problem. Whether or not the ERC Director becomes engaged in the university "space wars," the success of the ERC education program depends upon collegiality among ERC students at all levels of their university experience. Whether the Director oils this process with large quantities of pizza, with a scintillating seminar program, or with nights and weekends spent in contiguous office and common lab space, the process itself is a vital catalyst. For example, when laboratory space is not contiguous with the center, the students on an integrated project should be located in the same laboratory. In education the Director can facilitate, with a large array of imaginative policies ranging from student internships in industry to special funds for travel to conferences on relevant subjects; but s/he cannot control the basic educational process.

It is here that the ultramodern ERC structure comes up against a very successful but much more traditional institution, the university. In the Middle Ages, university students literally came to the universities/monasteries to sit at the feet of the monks. Although today's universities are obviously advanced in ideas and practices, the fundamental model is little changed in our traditional departments, and in many universities the ERC Director who feels that s/he can interfere with this traditional educational process is in for some rude shocks. An ERC can offer large stipends and a very heady team research atmosphere, but it cannot offer the student a degree.

For this reason the Director of an ERC must realize that the center's vital crop of interdisciplinary, team-oriented, industry-friendly graduates is dependent on many factors outside of his/her direct control. If the Director chooses to be unfriendly toward the departments who have students affiliated with the center, s/he can expect the graduate committee of each of those departments to raise major difficulties about the balance their students must strike between the department and the center. If the Director is fair-minded about departmental imperatives (like obligatory teaching assistant assignments), and if both the department and the center hold the student's interests to be paramount, a mutual trust will develop that will let the ERC "thing" happen. The ERC "thing" is unique in most universities, and it involves an enriched and very challenging interdisciplinary education for a limited number of excellent students who reflect credit on both their department and the center.

In an area such as education, in which the ERC can facilitate but not dictate, the Director must exercise great care in setting up program structures. The appointment of either a part-time faculty member or a full-time professional as Education Coordinator (see Chapter 4) will greatly facilitate the recruitment and integration of both undergraduate and graduate students into the center. This person may also handle the laborious logistics of the center's seminar and internship program(s) and the educational reporting requirements of sponsors. An experienced Education Coordinator should know the students well enough to flag cases in which the student is confused and/or troubled by conflicting demands of the center and of his/her home department, or by any of the myriad problems that beset the engineering acolyte today. Because a mature ERC may involve 40-60 undergraduate and 60-80 graduate students, the Director cannot begin to involve himself in either their personal or collective supervision. For this reason it is advisable to appoint an education committee whose chairperson works closely with the Education Coordinator. In this way each student knows that the center provides a professional and a faculty member that they can contact with any problems, while the Education Coordinator and the rest of the committee serve the center by advising the Director on policies that affect center students. Because of the patience required, and because of the ambiguity of the relationships between the center and its allied departments, the Director may be well advised to look for a light touch when appointing members to the education committee.

While the ERC pattern does not fit precisely into the educational structure of the modern university, it may comprise its salvation. The ERC Program has pioneered interdisciplinary, team-oriented research, and it has a lot of experience in both the pitfalls and the benefits of the "new" paradigm currently being endorsed for both science and engineering education at the highest levels of national policy. This very important paradigm shift is of considerable interest to the NSF and, as the most successful of its programs in this respect, the ERC Program may see fit to base a large element of its continuity on its accomplishments in the area of education. If this assessment is correct, as the ERC moves to self-suffiency after 11 years, individual ERCs may find a very successful interdisciplinary education program to be a large factor in their continuity within their own university environments.

2.3.2.2.3 Technology Transfer

Technology transfer is the third pillar of the ERC Program. There are indications, based on personal contacts and our survey, that this is the element of the Program that is most variable in the extent to which Center Directors really buy in. Perhaps for this reason, there is also a wide variety of performance across the ERCs in this program element, which was so pivotal in the original funding of this program by Congress and by the National Science Foundation. Virtually all ERC Directors are academic engineers and, occasionally, academic scientists. It is axiomatic that most academics feel that technology transfer is something that they could easily do, without any special training or experience, if they could just spare time from their well-understood responsibilities in research and/or education. Perhaps it seems to many of them to be simply a matter of a few simple phone calls and of taking a little time with the industrial people who regularly beat a path to the doors of their institutions.

This perception may be the key to the inability of American science and engineering to make a real impact on American industry, and this perceived problem was a large part of the rationale for the creation of the ERC Program in 1984. The subject of Industrial Collaboration and Technology Transfer is covered at length in Chapter 5. The treatment of this topic here is from the standpoint of overall center management and direction.

The Director of a new ERC is likely to be committed to research, or s/he would not have been successful in being funded, and s/he probably has the abiding concern of the academic with educational process. However, there is no guarantee that the new Director will have the same level of commitment to technology transfer. For this reason there is an especially wide range of choices that the Director will make in the early stages of the center's development that will affect the degree to which the center will be able to present its vision to industry for practical exploitation. At a minimum, the Director may use the academic workshop mechanism to collect a group of companies to form a paying audience in the applicable field, and then simply allocate a couple of articulate graduate students to give tours of the center in their spare time (although this approach will almost certainly fail to produce a successful industrial program.) Alternatively, the Director may designate one or more of the center PIs who are already well-connected to industry to handle company recruitment and meetings of the Industrial Advisory Board (IAB) or its analogue--although this approach, too, is not ideal. The technology transfer area does not seem to lend itself to management by a committee, and none of the respondents to our survey reported the formation of a committee for this purpose. If the incoming Director is personally committed to the process of technology transfer--as befits an ERC Director--s/he will take a very active role in company recruitment (also vital to center funding), in interacting with the IAB members, and in developing opportunities for joint research with sponsoring companies. And if s/he is serious about technology transfer, s/he will also hire a Director of Technology Transfer (often called an Industrial Liaison Specialist) from outside of the academic framework, and give this employee the freedom to build meaningful relationships on an ongoing basis with companies interested in the center's technology.

The technology transfer "portfolio" really cannot be separated from the research and education elements of the center, because ERC research should be cooperative with industry and industry should play a very active role in the education of our students (e.g., through internships). Ideally, the Director should be fully committed to technology transfer, and should consider activity in this area to be integral to other activities in research and education. S/he should ensure that the directors of education and technology transfer work well together and strive to coordinate their program activities where appropriate. S/he should work very closely with the center's Director of Technology Transfer, and with people in similar positions in the university, to move center technologies directly into that part of American industry that will benefit the most. The incoming Director will have telegraphed the extent of his/her commitment to technology transfer to the new center's member companies within the first 18 months of the center's operation by the choices that s/he has made to staff and activate this vital element of the center.

2.3.2.3 Mechanisms of Delegation

Sometimes just as important as the degree to which the Director delegates responsibility within the center are the mechanism(s) of delegation. In the administrative area (Section 2.3.2.1), economy dictates that individuals be allocated specific responsibilities, such as computer networking or accounting, and then be reinforced with less trained personnel when their workloads become intolerable. An important early decision must be the type of supervision and reporting to utilize for the center administrative staff. There is clearly a limit, which is dependent on the personality and policies of the Center Director, to the number of center employees who can take detailed direction from this one source. The alternative is to have all administrative staff report through another designated individual (a Deputy Director, Executive Director, or Administrative Director).

This delegation should be done very carefully because its consequences for the smooth operation of the center are likely to be quite significant. Everyone involved in an ERC must realize that the whole exercise is driven by the center's vision and strategic plan. The administrative function is only an "engine" (albeit an essential one) to facilitate the realization of the vision and, as such, it will always be secondary to the program activities of the center. Whether the engine is a large 12-cylinder or a small 4-cylinder really doesn't matter, as long as the vision is realized. A simple rule of thumb is that the administrative structure of the center should always be elastic (but structured). Flexibility is also important when establishing a reporting structure, because it may be difficult to assess realistically in advance how large a staff might ultimately be in place.

In the programmatic area (Section 2.3.2.2), the three pillars of an ERC that are outlined by the Program's charter may each be best served by a different mechanism of delegation. As was noted earlier, most centers have concluded that technology transfer is best handled by a dedicated professional person, with the personal involvement of the Director and other interested PIs, because of the extent to which success in this area depends on both rapid responses and personal contacts and relationships. Most Center Directors have set up "softer" structures to guide the educational programs of their centers. Because the ERCs themselves do not usually grant degrees, a high degree of cooperation is necessary with traditional departments. Education issues are often course-specific and/or student-specific, and many call for wisdom and mature judgment while revolving primarily around the interests of the students concerned. Hundreds of hours can be spent in the resolution of these issues; therefore, in some centers an education committee with an active and very accessible chairperson may constitute the best use of faculty time, provided that most details are handled by the Education Coordinator. In other centers the Education Coordinator has more autonomy in decisionmaking about course-specific and student-specific matters, looking to the Center Director and/or an education committee for broad policy guidance.

The reporting requirements of the ERC Program review process virtually demand that each center must appoint a leader for each research thrust area and that someone, usually the Director, must combine these reports with those of the education and technology transfer programs to produce the annual report. These thrust area leaders also provide a necessary management interface between the Director and the faculty researchers, with responsibility for the detailed planning of research within that thrust. The thrust leaders, perhaps with the appointment of an overall Associate Director for Research, would seem logically to constitute an efficient committee for the planning and execution of the center's strategic research plan. To ensure the periodic introduction of new people, new ideas, and new projects, the Director may choose to create an uptake mechanism whereby the center can be revitalized. This enterprise can take many forms--for example, a committee that emerges from a full meeting of the center (perhaps a retreat) and has the mandate of examining and possibly changing the center's strategic plan may be useful.

2.3.2.4 Principles of Delegation

Delegation of responsibilities is not an art that is peculiar to directing an ERC, and the basic principles that govern this important process are well established. Delegation should serve the personal interests of all parties involved, and it should always be voluntary within the academic context. A Director should consider the career path of all concerned, and should avoid overloading individuals or saddling young researchers with responsibilities before they are ready. Once a specific responsibility has been delegated, the Director should avoid interference or micromanagement. The Director should attempt to balance responsibilities within the ERC so that no single individual acquires excessive power and thus negatively impacts the contributions of other members of the center. People to whom responsibilities are delegated should have a clear idea of what is expected of them and of the length of time for which they are committing themselves. Most of the academics who make up the faculty of ERCs also have a base in their home departments, so that an ERC Director is really running a volunteer operation. The Director has the seductive power of his/her vision of the ERC, with certain financial resources at his/her disposal, but s/he does not have the power implicit in a direct "chain of command," so it is necessary to exercise considerable diplomacy in the process of delegation.

2.3.2.5 Continuity of Leadership

One ticklish area of delegation that should be mentioned concerns perhaps the ultimate delegation--that of the directorship of the center. Succession is an issue that many ambitious executives, in academe as well as business, find difficult to address. If one is performing well and enjoying oneself as a Director, it is perhaps counter-instinctive to make plans to replace oneself. Nevertheless, several ERC Directors have stepped down over the years, and three have passed away--two of them suddenly. As a responsible manager with a major investment of energy and commitment in the center, it is only prudent to provide a viable contingency plan for one's succession and thereby minimize the turbulence that would ensue in the event of the Director's departure.

Many of the centers have no such plan. Of those who have established a plan for continuity of leadership, most have appointed a Deputy Director or an Associate Director (often for research), who will take over the leadership role until a search can be organized to select a new Director (who may or may not be the deputy). This is an individual who routinely takes some of the Director's workload in managing aspects of center function. S/he has a seat on the Executive Council or equivalent and may have a large role in tracking and management of the center's strategic plan and/or interactions with the industrial advisory board. NSF now requires every new ERC to have this position, and some of the existing centers are following suit.

A few centers have purposely allowed the directorship to rotate among senior faculty every three to five years. They believe that this provides strength and balance in a multidisciplinary center, inasmuch as the directorship rotates among the disciplines. The practice also produces a core of seasoned talent within the center who can provide informed advice on complex or difficult issues. However, the centers that have tried this approach have found that it is not workable over the long term. The academic authority structure is already inherently weak enough without further clouding the issue of "who's in charge?"

2.3.3 What Not to Delegate

There is general agreement among ERC Directors that the following responsibilities should not be delegated.

• Major resource allocation and budget decisions, including fiscal oversight.
• Major realignment of center administrative and research structure.
• Final decision on hiring (and termination of) faculty and key staff.
• Final selection of companies to recruit as members.
• Formal contacts with NSF to address major issues.
• Policy interactions with department heads, college deans, and university top administrators.
• Negotiations with university administrators for commitments of resources.
• Integration of the ERC's annual report to NSF.
• Responsibility for the integrity of the ERC's reporting systems.

These clearly are functions and decisions that directly affect the health and continuing success of the ERC. Missteps could be quite destabilizing. As such, responsibility for them should rest with the individual with the lead responsibility for overall center management, the Director.

2.3.4 Facilitating Cross-Disciplinary Interactions

One of the distinguishing features of the ERC Program--an area in which it was a pioneer in contemporary academe--is its emphasis on cross-disciplinary research. Individual ERCs have devised many ways to facilitate cross-disciplinary interactions among the faculty. In a number of centers an absolutist approach is taken: i.e., projects without cross-disciplinary interactions will not be funded through the center's resources; or else the evaluation criteria for project continuation include cross-disciplinary collaboration. In other centers the requirement is not so absolute for all projects, although there is a strong preference for cross-disciplinarity and a strong message is given that collaboration is necessary if the ERC is to be successful.

Other mechanisms that are employed to encourage and facilitate these interactions include:

• specific requests for collaborative research to meet an identified need;
• special center funds made available for cross-disciplinary projects and proposals;
• inviting researchers from a range of fields to center research meetings and retreats;
• structuring center research so that project teams cannot complete their assigned projects without obtaining assistance from other teams;
• developing a set of "end-to-end" demonstrations, within and between research clusters, that illustrate how the tools and methodologies work together; and
• engaging in design activities with center industrial affiliates (which commonly involve participants from more than one project or research thrust).

See the case study for additional suggestions.

CASE STUDY: The Director of the Data Storage Systems Center, at Carnegie Mellon University, reports the following: "Most of our projects require cross-disciplinary work to successfully reach their goals. Many of these projects entail weekly group meetings involving faculty, research staff, and students from the various disciplines. We also have regular Center-wide seminars, which draw faculty and students from various disciplines. In the end, from my point of view, it makes no difference what department one comes from. We need a certain skill-set to accomplish our goals. The pool of faculty and students is generally (with the exception of outreach) University-wide and beyond. If I need a certain skill from a particular department, I go to the faculty member and ask if he or she is interested in having a graduate student funded by the Center. Many faculty members also come to me asking if the Center could fund them. I give each one careful consideration. To prepare the students for this cross-disciplinary work, we teach a two-day short course on the system the Center deals with, before the semester starts, to all incoming grad students. Then, in the spring semester we offer (at both the undergraduate and graduate levels) a course that has minimal prerequisites and that covers material from each of the six departments involved in the Center. Faculty from the various departments give lectures on their specialty. Such cross-disciplinary activities allow you to educate better engineers, learn something yourself as a faculty researcher, and solve problems you couldn't otherwise solve."

Directors report that the advantages of cross-disciplinary interactions include: the ability to address technical barriers that could not be overcome in any other way; intellectual challenge and stimulation; and a broader perspective available to faculty and students. The disadvantages are that: cross-disciplinary work requires more time and effort to understand the perspectives of colleagues from other disciplines; departments sometimes do not credit such projects in promotion decisions (this is where input from the ERC Director and the support of the Dean are critical); and faculty may have trouble obtaining individual credit, which they need to obtain promotion. Thus, the disadvantages are systemic, but can be overcome, and in any event they are outweighed by the advantages of this type of research to industry and the Nation.

2.3.5 Locating the Necessary Resources

Operating a well-run ERC involves bringing together the necessary resources, including not only personnel but also facilities and funding. Facilities required to carry out the ERC's research mission include so-called signature space for housing the center administrative offices, conference room, and general space for center-supported activities such as a computer laboratory, student library, and lounge where faculty, students, and staff can gather to discuss their work. Distributed laboratory space is necessary for developing basic materials, device, and system-level competencies. Usually the Dean of the College of Engineering and the chairs of the individual departments make the signature space available to the center. Individual laboratory space is usually made available to faculty on a have-need basis.

Once the space is available, some remodeling is usually required. Generally these resources are obtained from indirect cost recovery (ICR) money. Equipment to outfit the space can be part of the ERC award, or acquired through any number of government equipment grants including those of NSF, or the Department of Defense, industry, and philanthropic foundations.

Research management by necessity involves and revolves around available financial resources. Sources of financial support for ERC programs are many and varied, including: the primary NSF ERC award, university cost-sharing, state and industrial support, and any ERC faculty individual grants that contribute to the common pot of ERC resources. A number of ERCs receive no ICR from their host institution, while other ERCs recover a considerable amount of the indirect costs (IDCs) they generate, to use at their discretion. In the case of those ERCs receiving generous returns of their generated IDCs, some are also expected by their host institutions to subsidize a major portion (5-8 months) of the 9-month academic salary of several center-affiliated faculty members.

Industrial associates in a number of ERCs contribute to the finances of an ERC in a myriad of ways. Membership in an ERC at various levels requires a fee ranging from $1,000 to over $100,000 per year. Funds raised by these fees can be lumped into a common ERC pot; or some ERCs elect to keep industrial member fees separate from thrust area-related research. Also, in most ERCs industry can directly support a specific research endeavor, but this activity will operate independently of the thrust area work and is dependent only on its own budget. In some cases, research activities with industry require utilizing ERC resources to leverage industrial participation.

When considering the center's funding profile, it is important to maintain balance. For example, if most of the center's funding is from NSF, then the relevance to industry is somewhat suspect. Within industry, it is best to develop a diversified portfolio of partners ranging, if possible, horizontally across various industries and vertically from raw materials producers to parts suppliers to system manufacturers. If all the outside funding is from one industry, then there is a certain vulnerability if the particular industry goes through a bad patch. A balance between state and various federal government agency and industry funding is desirable, because no one sponsor or sector has an undue influence over the activities of the ERC.

2.3.6 Avoiding Failure, Pursuing Success

There are many kinds of problems that can lead to failure for an ERC. Difficulties in leadership and management (including financial management), problems in research planning and execution (including disintegration and failure to address its vision), and failure to engage with industry positively and in the proper ways are all sources of serious trouble. These are higher-order issues that are addressed throughout this Manual. However, some problems are less programmatic in nature, and can be ameliorated to some extent by specific actions of the Director. One issue that nearly all ERC Directors cite as a serious problem is the heavy workload placed on the Director. Regardless of the extent to which responsibilities are delegated, the Director is still almost always subject to potential burnout. One Director describes the problem as "a major flaw in the ERC concept;" even after 11 years as an ERC Director, he has found no real solution to the problem. The most prevalent approach to reducing the excessive burden on the Director is to add a Deputy or Associate Director; key staff members such as the Industrial Liaison Specialist and the Education Coordinator also remove some pressure. Several Directors point to the enormous importance of having an Operations Director or Administrative Director who is capable of handling many of the day-to-day operational tasks. Nevertheless, as one Director says, "The specter of burnout is ever-present."

Although pressures are greatest on the Director, burnout of other key personnel within an ERC is also a concern. Suggestions for alleviating this risk include the following:

• rotating standing assignments such as thrust area leadership and seminar planning;
• rotating people and responsibilities in conjunction with the university; and
• planning more thoroughly so as to reduce the number of distractions and unproductive activities.

Like failure, success is also largely a function of leadership and management, planning, and program execution. Many success factors are structural in nature; others are procedural or even attitudinal. Perhaps the best way to address these factors is to allow Center Directors to describe in their own words some of the measures that have helped their centers succeed in research, education, industrial interactions, and outreach:

"We have a clear mission that is closely linked to a well-identified, genuine industrial need."

"Our continual efforts in setting specific goals for each of the research thrusts and assessing milestones has been the most effective way to ensure success for the Center. The strategic planning mechanism is extremely important in this regard because in its full expression it integrates the Center. Goal-directed activities naturally attract industrial interest and also lead to opportunities to improve our educational offerings. From that standpoint, it is important that the Center Director has an accurate understanding of the limitations and potentials of each of the thrust areas."

"The successes of our Center in the areas of research and industrial interaction are closely intertwined--which is, perhaps, one of the reasons for our success in both areas. We have focused our program on well-defined, long-term, systems-oriented goals which the industry initially doubted we could achieve but were certainly willing at least to applaud. As we progressed toward those goals, we transferred component technologies to the industry, and gradually their 'applause' was converted to serious interaction and serious financial support."

"An executive advisory board (EAB) was established to provide national guidance on strategic research direction. The EAB consists of national leaders in the field, the Center Director, thrust leaders, and Industrial Advisory Board officers. The national leaders were selected from government and industry organizations that represent the broad national interests in the specific research area of the ERC."

"Our relations with industry have been key to our success. These include: the active role played by our Technical Advisory Committee, the weekly on-campus seminars by industry personnel; and the fact that we send students to conduct research at industrial sites."

"We have had considerable success recently with a new format for industrial meetings. Meetings in which we just told industry about our research were not received very enthusiastically. Instead, we are having a series of industry workshops on specific topics. We want these to be interactive workshops, so for each one we invite a limited number of key industrial people. Most of the speakers are industrial people, giving us their views of the particular topic, and some of our faculty speak as well. There is considerable time built in for discussion. An important (and time-consuming) element is writing up a report of the workshop afterwards. It is particularly useful to have a workshop on a topic or application in which the ERC has done some initial work, but where this work is not well known to industry; such a workshop can give the ERC instant visibility with industry. It should be noted that this will be positive visibility only if the workshop is run very professionally and the industry people are impressed with both the quality of the ideas and the professional management and operation of the ERC."

"Our most effective educational programs have all been related to enhancing the "systems understanding" of our students. These include short courses to incoming graduate students, summer research programs and seminars for undergraduates, 1-2 day workshops on systems design, and cross-disciplinary systems design courses at both the graduate and undergraduate levels."

"A large outreach program provides introductory contact with science for local disadvantaged school children, and leadership opportunities for the grad student mentors. Through this program we have learned to leverage our outreach activities with those of other research centers on campus to 'make a difference' in our community. The center allowed us to establish productive links with minority institutions and historically black colleges and universities."

"We have been very effective in creating an environment in which the students and faculty can participate in cross-disciplinary, team-oriented research. This kind of education allows our students to enter industry and quickly 'hit the ground running.' It produces a new breed of student, one who has a fundamental appreciation of how industry operates and who recognizes what it takes to be successful in the modern industrial world."

"We also have learned how to collaborate with other universities and industry to share resources and expertise to jointly solve problems that would be impossible for either party to solve alone."

"Finally, we have developed a unique experimental testbed that provides students with an opportunity to participate in research that is at the cutting edge of future technology development. They have the opportunity to work on equipment that is unavailable to students anywhere else in the Nation."

An ERC represents a new type of organization in academe. At each university where a new ERC is established, the members of the ERC faculty and staff generally have to feel their way along in forming a cohesive team. Effective leadership from the Director is indispensable to this process. However, formal training in team-building and organizational interaction in this novel setting can be highly effective in speeding the development of these skills.

Rewarding center participants for strong performance is an excellent morale-booster and an incentive for further success. Many kinds of reward are available for Center Directors to bestow. One of the most prevalent and effective is continued or increased research support, including seed funding; increased compensation is of course one mechanism. Additional travel funds for making presentations at conferences can be provided out of center unrestricted funds, as well as scholarships and fellowships. Increased visibility and support for making presentations at program reviews is appreciated as a career-enhancer. Success should also, of course, lead to promotion and tenure for junior faculty in the center. Several of the centers nominate their deserving staff for university awards and undergraduates for university-sponsored project awards. (For Directors themselves, nomination to membership in the National Academy of Engineering is an appropriate form of recognition; several Directors have achieved NAE membership.) Recognition in the center newsletter and at annual meetings is an intangible but appreciated honor. One center bestows informal "citizenship" awards at center social gatherings. Finally, nothing replaces the personal recognition and appreciation expressed by the Director and other center managers for a job well done.

If you have specific comments and questions on the contents of this site and/or its use, please contact webmaster@erc-assoc.org. To report technical problems, contact techhelp@erc-assoc.org.