Each ERC develops an overall strategic plan that encompasses all areas of its operation. This level of planning is discussed briefly in Chapter 2, "Center Leadership and Strategic Direction." However, the core of the ERC's strategic plan is its research strategy. Strategic research planning is generally a systematic, stepwise process, as described below.

3.2.1 State-of-the-Art Analysis

The first step in developing an ERC research strategy should be to conduct an analysis of the state of the art in the field. An understanding of the state-of-the-art underlies the development of a center's vision of achieving major advances in a next-generation engineered system. This understanding aids in identifying the barriers that stand in the way of further progress, culminating in a set of research objectives that, if met, move the center toward achieving its vision. Thus, the state-of-the-art analysis gives the current situation, the vision statement describes the desired future state, and identified technical barriers elucidate what needs to be done so that the future outlined in the vision statement can be achieved. As the center progresses, the state-of-the-art analysis should be continually updated, tracing the development of the work in fields closely aligned with the center's vision and documenting the contributions made by the ERC's faculty and others around the world. This documentation provides the center with a history of its contributions, and those of others, to the field.

3.2.2 Vision

Just as a well-thought-out research strategy is essential for effective research management, a compelling and well-articulated vision of the center's purpose and advances to be achieved is the key to a compelling research strategy. The vision should motivate and encourage participation and interest in the center and its activities, aid in recruiting faculty and industry to become involved in the center's research agenda, and attract industry and government support. It is important to have the broadest possible understanding and acceptance of the vision among ERC participants. Without this, it will be difficult to keep the research integrated and focused on achieving the goals of the center. It is essential that all participants understand the vision, why it is important, and the means the center plans to employ to achieve the vision. The steps required to achieve the vision and how these steps will be undertaken should then be articulated in the strategic plan.

All ERC participants responding to a questionnaire distributed by the authors of this chapter agreed that it is the Center Director who is responsible for initially articulating the center's vision. Usually it is the Director who recruits research faculty and staff who have expertise in a particular area considered to be essential for achieving the center's vision. The challenge is in formulating a shared vision that balances "top down" management with "bottom up" faculty expertise. The Director is responsible for putting in place a process for formulating the center's shared vision. If this is skillfully done, faculty should take part in contributing to an achievable vision, developed through a logical, participatory process, not one mandated from above.

3.2.3 Identification of Technical Barriers

As in engineering practice, where definition of the problem is one of the most important steps in the process of developing a solution, in academic research management articulation of the technical barriers is essential in carrying out a successful research strategy. Otherwise, the center faculty could easily work on solving problems that may be of interest to them but that would not lead the center toward achieving its vision.

In some cases, there may be a number of technical barriers to achieving the ERC's next-generation systems goals and a center may choose to focus on only a few of them. It may be that the barriers represent different approaches that could be taken to achieve the same vision, and by choosing a subset of them the center is making a statement about what it believes to be the optimum method for achieving progress. Or the center may choose certain barriers on which to focus based on the core competency or expertise of the majority of the center participants. No matter the reason for the choice of problems to tackle, the center should make it clear to all participants why the technical barriers it chose to address were selected. This will help in further elucidating, to those inside and also outside of the center, the strategy the center plans to follow in achieving its vision.

A center must also identify broader barriers to success that are not strictly technical, such as costs, policies, and regulations. In some fields, industry is in an excellent position to thoroughly appreciate these issues. They develop products for various applications and understand the forces that may prevent their products from being accepted in the marketplace. One forum for identifying barrier issues is at a strategic planning meeting that includes industrial representatives, where technical discussion sessions focused on barrier issue identification may be held among faculty, students, and industrial colleagues.

The research areas on which the center has chosen to focus in order to achieve its vision should follow from its understanding of the state of the art and identification of the technical and system-level barriers yet to be overcome. The heart of a center's strategy is how its research is brought to bear on these barriers, which cannot be overcome without the integration of interrelated research activities.

Each ERC focuses its research efforts at three levels: fundamental research, development of enabling technology/tools, and the engineered systems level. The relative distribution of effort across these three levels will differ from center to center, and across time at a given center. The research strategy should identify what breakthroughs or developments in fundamental science and/or enabling technology are required initially, how they are interconnected, how further progress will build on these achievements and contribute to a convergence on the systems level, and which projects can and should proceed in parallel.

Each ERC should include in its plan the components discussed here. However, creativity in the development of individual center plans is valued and expected. Various approaches to strategic planning will be found to be preferable in individual centers, depending on their field and their specific circumstances. Attached for illustration are actual strategic plans of three ERCs (Example 1, Example 2, Example 3). All are examples of effective plans, yet they cover a spectrum from loose/conceptual to tight/detailed and highly defined.

3.2.4.1 Goals and Objectives

In organizing its research strategy, a center should formulate major goals that must be met as a demonstration of progress toward achieving its vision. The formulation of clear-cut goals will aid in defining a pathway toward major accomplishments, along with the outcomes expected from specific projects or research areas.

As in any academic research activity, projects within a center should have objectives and an estimated time required to meet those objectives. As a center strives to meet its longer-term goals, objectives should be specified in the strategic plan that bring to a culmination parts of many individual projects. For example, consider the following related but separate projects: (1) developing the analytical hardware to take atomic-scale images of chemistry differences on a surface, (2) deriving techniques to engineer controlled patterned chemistry at that surface, and (3) refining tools to measure the resultant patterns of biological cells that adhere at different locations on the patterned surface. These projects all have objectives and tangible deliverables. The goal that their combined results achieves is the ability, for the first time, to affect biological cell adhesion in a quantitative way by controlling surface chemistry.

The case study below illustrates the way that objectives support the achievement of goals in the context of a center that strives for large improvements in the performance of electronic systems.

CASE STUDY:
The strategic planning process utilized at Georgia Tech's Packaging Research Center (PRC) begins with the identification of system-level goals, such as a next-generation pager or PC or camcorder. The next-generation pager can be used as an example: Today's pager has some 500 components, weighs 6 ounces, has certain functions, operates at a microprocessor speed of 900 MHz, and costs about $120 to manufacture. The goal might be to improve all these features by a factor of 10 (e.g., in the case of the pager, to achieve 9 GHz operation).

The question is then asked, what packaging technologies are needed? For any electronic device, the answer involves 11 areas: electrical design, electrical testing, high-density wiring, large-area intelligent manufacturing, integral passives, optoelectronics, RF electronics, flipchip assembly, thermal management, reliability, and system prototyping. Each of these represents an enabling technology, and constitutes a research thrust toward the system-level goal. Working with their industrial partners, the PRC then develops a roadmap for each enabling technology. Next, they define the fundamental research projects. A testbed is then formed (this is termed a "test vehicle"), which is a generic prototype for a given electronic device or product such as the pager. The faculty are asked to focus on testing their research results in the context of the test vehicle. This requirement keeps the researchers disciplined in the directions their projects take.

Finally, systems manufacturing companies from among the industrial members use portions of the prototype to build a prototype specific to their industry. For example, Motorola might build a prototype pager, and Boeing might build the power supply for an F-22 fighter.

Thus, the strategic plan provides a structure for the PRC's research and systematizes it in a way that results in a wide range of useful applications.

Just as the PRC's research is drawn from specified systems goals, similarly in the Data Storage Systems Center at Carnegie-Mellon University all strategic planning is focused on attaining a specified system specification (e.g., hard drive or optical drive performance) that is very advanced and ambitious. An example might be to attain 100 Gbit/sq. in. recording density, when the current industry standard is 2 Gbit/sq. in., along with the entire configuration for such a system. Such an approach makes it easier to implement a strategic plan; the goal is finite and concrete. Some ERCs operate in a field that is amenable to this approach; not all do.

In identifying goals for the center, a useful question to ask the faculty during strategic planning is: "What should we expect to see in a year (or two, or three) if you were to be successful?" One ERC research manager who asked this question found that most people replied fairly specifically -- enough so that objectives could be derived from the responses. The idea to do this came from the center's industrial partners; it was the kind of information they themselves asked within the company.

3.2.4.2 Deliverables

The ability of the ERC to achieve industrial use of its knowledge and technology is a key element of a successful ERC, one that requires a strong partnership with industry. Deliverables can take many forms, including fundamental knowledge advances and discoveries, new theories or models describing physical properties or behavior, access to students, experimental techniques, software, and new materials and processes. The opportunities to develop these deliverables, and their possible utility to the industrial partners, should be recognized and considered during the development of the research strategy. Deliverables may stem from research at all levels of investigation, ranging from basic science to engineering testbeds. It should be noted that how a deliverable can best be understood and incorporated into a company's workflow is not necessarily straightforward, as students and faculty may have different expectations from those in industry regarding the utility and user-friendliness of the tools they develop.

Experienced ERC research managers find that a steady flow of deliverables is important for keeping industry engaged and satisfied with the center. One recommends that the ERC always try to give industry something every year out of each project. Some centers have hired one or more professional engineering/technical staff members whose job is to turn research results into products that industry can readily understand and use.

See Section 3.3.5 for a discussion of industrial collaboration and technology transfer.

3.2.4.3 Testbeds

The key to determining if the center's research agenda will successfully lead to a next-generation engineered system and achievement of the center's vision is to periodically pull together the research outcomes and test the system. Thus, the research strategy must involve the development of proof-of-concept testbeds, which integrate elements of the system to determine if all components work together as planned. These testbeds not only ensure that the research outcomes are integrated and tested, they also serve to drive the research. They act as focal points for the researchers when planning their research projects. They also supply a framework for faculty, students, and industry representatives to work together and to gain a better understanding of the system they are working with. In addition, and perhaps most importantly, testbeds highlight which areas of the center's research require further investigation, ensuring that the research is focusing on the critical problems that must be addressed in order to achieve the center's vision. Thus, testbeds facilitate modification of the center's research strategy based on what is being learned as the center works toward achieving its vision.

Many ERCs have had highly successful testbeds. An early one, at the Purdue University ERC, was an integrated manufacturing software system called the Quick Turnaround Cell (QTC). The QTC combines computer-aided design and computer-aided manufacturing in a single software package that automates the several operations and processes involved in designing and machining limited numbers of prismatic parts. The QTC testbed was the Research, Development, and Engineering Laboratory of the U.S. Army Missile Command at Redstone Arsenal, near Huntsville, Alabama. In the early 1990s this project broke new ground in inter-institutional collaboration and research, eventually moving from Huntsville to application in industry.

Testbed development relies on research focusing on both fundamental scientific knowledge and the development of enabling technology. Although the balance between fundamental research and technology development may vary from center to center, all centers will have some aspects of both, as both are needed to make significant progress toward the development of the next generation of engineered systems.

3.2.5.1 Faculty Involvement

When putting together a research strategy and agenda, both the research required to achieve the vision and the research interests of the faculty must be taken into account. If there is not a good intersection of the two, it will be difficult to recruit enthusiastic researchers to the center. Also, in the academic setting, a directive "command approach" generally does not succeed in motivating faculty. A center is most likely to reach its goals if all participants feel a sense of ownership of the research strategy. ERC research managers have found that it is very important to obtain agreement to the research strategy of the whole ERC team up front. Trying to achieve consensus later on is impossible unless there is agreement at the outset.

Obtaining and maintaining faculty buy-in to the strategic research plan are not the same thing. Centers often encounter the problem that, once they have agreed to a strategic plan, faculty members are obliged to modify their own personal research objectives to an extent. Some will try to weave their own research into the center's agenda whenever possible. One Center Director commented that it is important to make it obvious to the faculty fairly quickly (i.e., within 6 months) that they are gaining value from their participation. In his center this issue was "a major continuing challenge, requiring endless discussion." Another Director noted that it is the Director's job to determine which of the faculty has a broad enough intellectual portfolio to make changes that the center needs, at the expense of their personal intellectual pursuits. The faculty cannot do this as objectively as the Director can, and some will resist making major changes in the direction of their research. There must be at least a critical core of faculty who are flexible enough to adapt to changing needs.

The full range of research expertise required throughout the lifetime of the center as it works to achieve the vision may not be resident at the university at which the center is based, nor at any of its partner universities. The need for additional expertise, when it exists, must be recognized and filled by means of outreach to researchers at other institutions. Normally these would be paid research projects; continued involvement of these faculty in the ERC would depend on the quality of their work and its continued relevance to the ERC's needs.

3.2.5.2 Industrial Buy-in

A key element in building a strong industry partnership is to actively involve industry in the strategic planning process. Both the creation and evolution of an ERC and its strategic plan must involve frequent input from the industrial members. Most ERCs have constructed formal mechanisms through which to continually draw upon industry's advice and experience in technology development. These mechanisms are described in detail in Chapter 5, "Industrial Collaboration and Technology Transfer." However, because they are crucial to obtaining industrial buy-in to the center's research program, they will be discussed briefly here.

Typically, initial industrial participation in an ERC is based upon the prior direct interaction of companies with participating faculty members and/or a long-term association with an academic department. Upon proposing an ERC to NSF, several prospective centers have made the effort to involve industries from the relevant technology sector in an intensive workshop to construct and review a proposed strategic plan.

Once an ERC is established, all ERCs devise ways to engage industrial participation. For example, every ERC has an Industrial Advisory Board (IAB) or the equivalent, often with subcommittees for obtaining more specific technical input at the level of Thrust Areas and projects. Most ERCs hold some form of semiannual IAB meeting. These meetings serve to present research progress to industry and to seek industry's assessment of new research directions. ERCs report that the actual content of specific projects frequently has been "course-corrected" by means of industrial advice. Personnel exchanges in both directions, involving faculty and students as well as industry engineers, are perhaps the best way to increase industrial involvement through technology transfer.

Keeping industry continually involved in the center's strategic planning process can present some difficulties. For example, the Institute for Systems Research at the University of Maryland found it hard to conduct two-way, joint strategic planning with industry because very few companies were willing to open up their own strategic plans. It was much easier to get input from them on the ERC's projects and plans. A couple of observations: Long-term industrial visitors tended to give much more useful input about their company's plans. And individual companies in private meetings with the ERC management gave much more specific input on their plans than they were willing to do in open meetings. This reluctance will vary across centers and industries.

Each center's IAB now carries out analyses of the strengths, weaknesses, opportunities, and threats to survival of the ERC, both for the benefit of the center and for input to the NSF site visit team. (See Chapter 5 for a description of this analysis.)

Such interactions serve to draw the industrial and ERC communities together, but they do not necessarily convince industry to accept or buy into the ERC philosophy. Ultimately, the ERC's responsibility is to develop fundamental discoveries or new technologies that they can then present to industry for further development into competitive, "on-the-shelf" technology. ERCs and their industrial members must resist the tendency to expect the ERC to solve industry's immediate problems. An ERC must not become a research "job shop" for any single company or group of companies. To that end, ERCs should not turn project selection over to their industrial members, as the projects are likely to become skewed in the direction of meeting industry's more immediate needs. Instead, each center must create a mechanism to maintain the collective interest while addressing the specific needs of member companies. In most cases, ERCs promote the funding of specific projects, funded directly by a single company or a small consortium, that address practical industrial needs. Under this mechanism, research results are the property of the company under common licensing agreements and are not shared with all the industrial members.

A center must strive to develop with its industrial supporters a rapport where each recognizes its role in the ERC technology development philosophy. Industry's role is to present the ERC with "model systems" or realistic conditions, so that the ERC's fundamental research has pertinence. A rule of thumb is, "If the ERC cannot pass its discoveries to industry, then it should not do the research." Industry should realize that the time horizon for most technology development in academic basic research is 5-10 years, while in industry it may be measured in months. Again, each ERC must erect infrastructures or mechanisms to draw industry into the active role of taking discoveries and creating technologies. If all an ERC did was to make its industry aware of the basic research without involving industry in the center's pursuit of those advances, the center would be a failure. Mechanisms to develop jointly sponsored research projects or research initiatives that lead to ERC testbeds should be in place at an ERC by Year 3. It is in these joint activities that the ERC and its industry members begin not only to solve problems jointly but also to create new technologies.

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.