Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
4 Report of the Panel on the Aerospace Industry This report was prepared by the Panel on the Aerospace Industry, one of five panels formed by the Committee on the Impact of Academic Research on Indus- trial Performance. The panel of five included three members of the NAB (one from academia and two from industry), one other member from academia, and one from industry. Two of the NAE members were also members of the parent committee. The charge to the panel was to evaluate the past impact of academic research on the performance of the aerospace industry and identify ways to in- crease the impact in light of recent and ongoing changes in the structure and economic situation of the industry. The report is intended for policy makers in industry, government, and academia. Industry performance was defined as share- holder value. This metric differs from the traditional measure of success in the aerospace industry, which was its contribution to national security or to the space program. The aerospace industry was selected for study as an example of an industry, now relatively mature, that developed with extensive funding by government in research and technology and that is dependent on advanced technologies for its present and future economic competitiveness. Therefore, the industry might pro- vide a baseline for comparison with other less mature industry sectors. The aerospace industry has been the beneficiary of more than 50 years of government-subsidized research conducted by industry, universities, and govern- ment laboratories. Subsidies have taken the form of direct funding by the Na- tional Aeronautics and Space Administration (NASA) and the U.S. Air Force, other defense-related funding, incentives in government contracts, and tax incen- tives; research was focused primarily on improving performance to meet the 115
116 THE IMPACT OF ACADEMIC RESEARCH ON INDUSTRIAL PERFORMANCE needs of national defense and the space race. In recent years, large cuts in federal support, combined with other competitive and financial pressures, have resulted in major changes in the industry. Dramatic consolidation, largely the result of huge cuts in defense spending and greater emphasis on the use of commercial, off-the-shelf technologies in the 1990s, as well as increasing global competition, has changed the scope, priorities, and practice of aerospace R&D. Spending on R&D in the industry declined throughout the 1990s to less than half its peak in 1987. In 2000, total R&D spending (by government, industry, and other institutions) in aerospace totaled about $10.3 billion, accounting for roughly 9 percent of R&D among manufac- turing industries (NSF, 2001a). Employment is down 40 percent from its peak in 1989, and the number of scientists and engineers in 1999 was less than half the number employed in 1986 (AIA, 2001~. (It is interesting to note, however, that over the last two decades scientists and engineers in the aerospace industry have earned more than 25 percent more than their counterparts in other industrial sectors [AIA, 20011.) As technologies have matured, margins have shrunk, cost reductions have taken precedence over improvements in performance, and elec- tronics and information technology now account for a large percentage of aero- space product value and technical emphasis. Priorities in R&D have changed accordingly. Historic patterns of industry-university interaction, which were based on significant government funding of R&D, have been broken; new models will certainly emerge that encompass not only R&D funding, practice, and expecta- tions, but also engineering education. But first, significant cultural and practical barriers will have to be overcome. Indeed, for academic research to have the maximum beneficial effect on the new aerospace industry, the entire structure of academic research in aerospace will have to change. SCOPE OF THE STUDY According to government classifications, the aerospace industry includes aircraft (NAICS 336411), aircraft engines and engine parts (NAICS 336412), aircraft equipment and parts (NAICS 336413), missiles and space vehicles and parts (NAICS 336414), guided missile and space vehicle propulsion units and parts (NAICS 336415), and guided missile and space vehicle parts not classified elsewhere (NAICS 336419~. The panel has expanded this definition to include space-based information systems, a burgeoning segment of space commerce. The academic community that supports the industry was also defined broadly to in- clude departments of mechanical engineering, materials science and engineering, and computer science, as well as the relatively few departments of aero- space englneenng. Overall industry sales for 2000 exceeded $146 billion (U.S. Bureau of the Census, 2002~. The panel considered it essential to limit discussion to five sectors
AEROSPACE INDUSTRY 117 of the industry that made significantly different contributions to total industry shipments of 1999-2000.2 · gas turbine propulsion systems: $10 billion · civil transport aircraft: $30 billion . launch vehicles: $11 billion · unmanned aerial vehicles: less than $1 billion · space-based information systems: $12 billion Because the first three are mature sectors, cost and reliability have replaced performance as their principle criteria for success. The last two are relatively immature sectors that are undergoing rapid development and are, therefore, more dependent on new technologies. Gas-Turbine Propulsion Systems The gas-turbine industry developed in the 1950s and 1960s with the rapid conversion of both military and commercial aircraft from reciprocating to jet propulsion engines. Initially, several companies entered the arena, but in a rather short time all but about a half-dozen had dropped out, either because of the large financial investment required or because of technical difficulties. Currently, three large manufacturers of jet engines General Electric Aircraft Engines, Pratt & Whitney, and Rolls-Royce are engaged in intense competition for both military and commercial business. Smaller firms supply niche markets, such as general . · · ·. aviation or missiles. Driven by competition for higher thrust/weight ratios and lower fuel con- sumption, the providers of jet engines are pressing materials and fluid mechanical and solid mechanical design procedures to their limits. In this high-stakes busi- ness, the development of a new engine costs up to $1 billion, and companies are eager to take advantage of the improved understanding and new techniques that academia can provide. The knowledge base for academic researchers is arcane, however, and the community of researchers is small. Presently, only about a half- dozen universities in the United States are making significant contributions. Civil Transport Aircraft Prior to World War II, the commercial aircraft industry was robust. With the introduction of the jet transport in the 1950s, the industry entered a period of rapid growth. Since the late 1970s, sales of civil aircraft have more than doubled, from $14.3 billion to more than $38 billion (AIA, 2001~. Like the engine indus- try, the commercial aircraft industry initially comprised several companies, all but one of which have now been absorbed into the Boeing Company, which has
118 THE IMPACT OF ACADEMIC RESEARCH ON INDUSTRIAL PERFORMANCE dominated the field for the last decade. In fact, because some aerospace firms that are not formally part of Boeing manufacture major components of Boeing air- craft, its dominance is actually greater than it appears. Boeing's only substantial competition in commercial aircraft is Airbus Industry, a European consortium. The commercial transport market is fiercely contested, and market share can be gained or lost by small differences in performance or cost. Technologies to improve performance and manufacturing are both critical to success, which creates a fertile field for academic research. A significant contribution of academic re- search to the industry has been the development of design techniques based in computational fluid dynamics. For example, the Aerospace Design Program at the Georgia Institute of Technology receives financial support from a number of aero- space companies. In addition, the Lean Aerospace Initiative at the Massachusetts Institute of Technology (see Box 4-1) and the Automation and Robotics Research Institute at the University of Texas-Arlington are addressing manufacturing tech- nology and management issues and have attracted significant industry interest.
AEROSPACE INDUSTRY 119 Launch Vehicles The launch-vehicle industry grew out of the ballistic missile industry with the dawn of the space age in 1958. The new industry was entirely dependent on government funding until the birth of the communications satellite industry in the 1970s. In the last 10 years, the commercial market for satellites has become comparable to the government market for satellites, missiles, and NASA mission hardware, and with the proliferation of communications satellite constellations, the demand for launch services has grown rapidly.3 There are currently three major U.S. suppliers of launch services (Boeing, Lockheed Martin, and Orbital) and four foreign suppliers of either launch vehicles or launch services. All but the two or three most recently developed launch vehicles have been derived from ballistic missile technology, which was heavily funded by the federal government from the 1950s through the 1970s. Initially, academic re- search made substantial contributions in key technical areas, such as reentry,
120 THE IMPACT OF ACADEMIC RESEARCH ON INDUSTRIAL PERFORMANCE the combustion of rocket engines, and the development of resin matrix compos- ites, which are widely used in solid-rocket motor cases, nozzles, and core vehicle components.4 In the past few years, launch vehicles have not been a very fertile field for academic research, partly because of the lack of new developments and partly because the major problems have been developmental rather than basic. Unmanned Aerial Vehicles Although drones have been used since the beginning of aviation as targets and as research vehicles, unpiloted aircraft became important militarily with the introduction of cruise missiles enabled by the development of very small gas- turbine engines and terrain-following guidance systems, which extended their range and gave them the capability of attacking designated targets. Unmanned aerial vehicles (UAVs) have been operational for about two decades. Recently, however, there has been a good deal of interest in both the military and scientific communities in the development of very small, autonomous, aerial vehicles (microair vehicles [MAVs] so small that they are essentially covert) equipped with miniaturized imaging systems and guidance systems enabled by micro- circuit technology (see Box 4-2~.
AEROSPACE INDUSTRY 121 Another area of growing interest is unpiloted vehicles capable of long endur- ance flight at very high altitudes for atmospheric sampling or surveillance. This type of UAV, marketed by innovative companies such as Aurora Flight Systems, was facilitated by advances in lightweight materials, control technology, and modeling and simulation. Several firms that emerged directly from academic research continue to rely on research by academics. Space-Based Information Systems Space-based information systems include all systems that use orbital assets to acquire or transmit information, such as observational satellites for military surveillance, weather satellites, navigation and positioning satellites, and com- mercial communications satellites. The first space-based information systems were military surveillance satellites that were launched in the 1960s with great secrecy. These systems have been systematically upgraded since then and are still an essential component of our national defense. Weather satellites and early geosynchronous communications satellites came next. The number and capacity of geosynchronous satellites have increased steadily. The largest potential in- crease in communication satellites, however, will be the launch of large constel- lations of satellites, some in low-Earth orbit ("little" and "big" LEOs) and some in specialized orbits for particular markets. Although several companies address- ing this market failed in the late 1990s (e.g., Iridium, ICO Global Communica- tions), others (e.g., Teledesic and Satellite LLC) plan to launch extensive satellite networks to provide broadband data communications.5 The increase in the number of satellites has been enabled by the explosive development of microcircuits, which made information processing possible, in- cluding information buffering and the passing of information between satellites in an array and ground stations. Electronic miniaturization has also made it pos- sible to build satellites with considerable capabilities that weigh as little as 90 pounds. The key technologies for these satellites are microcircuitry, antennae, photovoltaic power supplies, lightweight structures, and small propulsion sys- tems for orbital positioning and maintenance. High-volume, low-cost manufac- turing of standardized parts has been essential to the emergence of this segment of the industry. INNOVATION The most obvious innovations in aerospace have come from industry through the development of new products and systems. This pattern goes back to the Wright brothers, who were motivated to conduct their research on airfoils and control by a desire to build and market a useful aircraft. With few excep- tions (e.g., Robert H. Goddard, who was a professor, and C.S. Draper, who invented and developed inertial guidance), key innovations have not arisen
122 THE IMPACT OF ACADEMIC RESEARCH ON INDUSTRIAL PERFORMANCE from academic research. The pioneers of aerospace were engineers and entre- preneurs who used available technologies to create new capabilities for flight in the atmosphere or beyond. Several government laboratories were created to stimulate the growth of the industry. The first of these was the National Advisory Committee for Aero- nautics (NACA), which was created in 1915 when decision makers in the government became aware that the United States was far behind European nations in the development of aeronautics. The establishment of the Langley Aeronautical Laboratory at Hampton, Virginia, followed; research there was initially focused on aerodynamics, structures, control, and propulsion for mili- tary aircraft. The Langley laboratory had very little connection with universi- ties. During World War II, some universities established very large and effec- tive R&D programs. Examples include the Radiation Laboratory at MIT, which played a major role in the development of radar in the United States, and the nuclear laboratories of the University of California. It is important to realize that these were essentially industry laboratories embedded in the academic environment for the duration of the war. They did not pursue academic research agendas as we think of them now. After World War II, all of the military services established laboratories to work on the technologies most important to them. These laboratories were staffed by a mix of civil servants and military personnel. One of the functions of the military-service laboratories was to maintain contact with universities by providing financial support and encouraging faculty to address issues of concern to the services. At about the same time, the services established organizations devoted to funding basic research. The first of these, the Office of Naval Research, was followed by the Air Force Office of Scientific Re- search (AFOSR) and the Office of Army Research. In their heyday, these offices commanded sizeable research budgets and funded much of the aca- demic research in aerospace, a good deal of it only tenuously connected to the needs of the services. In 1958, NASA was established with the U.S. commitment to the Apollo Program; NASA followed a similar pattern and funded a great deal of academic research. Whereas NACA had been almost entirely an in-house research organi- zation devoted to aeronautical technology and facilities, NASA issued grants and contracts to industry and universities, using its civil service workforce to manage these activities. Of course, these research activities were subordinate to NASA' s main mission, which was to go to the Moon. Today, government agencies command far fewer resources for research and have focused more tightly on their missions. Some are also reexamining their relationships with academic researchers. For instance, in 1994 NASA's Office of Aeronautics appointed a University Strategy Task Force to review the agency's support of academic research and recommend policy and other changes to ensure the long-term health of aerospace research in academia.
AEROSPACE INDUSTRY 123 Because of the combination of substantial government funding and the emer- gence of government laboratories dedicated to aerospace research, universities assumed a role of supporting the R&D activities of the services and NASA by addressing issues that emerged during the technology development process and improving the base technologies for later applications. Universities also played a large role in the development and improvement of techniques for analyzing fluid flows and structures that were enabled by advances in digital computation. But even in the area of computational fluid dynamics, much of the innovation origi- nated in government laboratories and industry, with academic research playing a supporting role. FINDINGS Responses to questionnaires and discussions at a workshop convened as part of this study revealed consistent concerns about the impact of academia on the aerospace industry and the future of the university-industry relationship (see Ad- dendum). These concerns can be divided into five subject areas: (1) the implica- tions of changes in the federal research support structure; (2) the value placed on academic research and education by industry and the implications for industry support of academic research; (3) the impact of changes in the industry on the research and educational capabilities of universities; (4) intellectual property rights and how they affect university-industry collaborations; and (5) arrangements to promote cooperative research between academia and industry. Changes in the Federal Research Support Structure For three decades after World War II, the mission-oriented agencies of the federal government had charters from Congress and successive administrations to support a broad range of research without having to demonstrate the applica- bility of the research to their missions. The first agency with this kind of flexibil- ity was the Office of Naval Research, but eventually the Air Force, Army, NASA, U.S. Department of Energy, and other smaller agencies were given the same support and flexibility. Later, the National Science Foundation (NSF) and the National Institutes of Health (NIH) were created with specific responsibilities for funding so-called basic research (meaning research without specific, known ap- plications). NSF has not played a very significant role in funding for aerospace- oriented research, although some research on manufacturing funded through the NSF Engineering Directorate may be applicable to aerospace. The strong support of the federal government for aerospace research has diminished significantly in the last decade. Federal funding has fallen dramati- cally, and available funding is being managed much more carefully to achieve specific results at lower cost. Mission-oriented agencies now insist on demon- strated relevance to their missions as a precondition for funding research. This
124 THE IMPACT OF ACADEMIC RESEARCH ON INDUSTRIAL PERFORMANCE represents a significant change from the earlier mind-set when ensuring the health of academic research was considered an important part of an agency's mission. Beginning in 1990, the AFOSR began to assess academic research in terms of improvements in the performance of military aircraft. Universities were no longer considered AFOSR customers, but means to an end. AFOSR focused particularly on how the results of academic research could be transitioned into applications that would improve Air Force systems. For example, in fluid dynam- ics, a mature technology, the transition has taken 20 or more years from basic research to application. AFOSR is working toward shortening the transition time using a new model of technological innovation based on a work published by Stanford University, Conceptual Foundations of Multi-Disciplinary Thinking (Kline, 1995~. Today, when the Air Force needs a new product, the reserve of knowledge is first reviewed, and research is funded only if the necessary knowl- edge is not available. AFOSR has found that the knowledge base created by years of support for research is extensive and that the need for new basic research is not as great as it once was. This reflects the maturity of the technology/product/ industry. Using this approach, AFOSR program managers have become brokers between industry and academic researchers, facilitating networking in the re- search community by communicating the needs of industry, the sources of knowl- edge from past research results, and the creation of new knowledge through new research, when necessary. These changes have been made in other agencies as well. NASA, for ex- ample, used to have a generic space technology research program. Today, re- search decisions have been distributed to offices with mission responsibilities, and only technologies with near-term mission applicability are supported. Because of these changes, significantly less advanced research in aerospace is being done or even contemplated by the U.S. government, potentially shrink- ing the future pool of technology available in the field. Arguably, this situation may be a correct response to the maturing process; technological progress in the industry is progressing slower than in the past; the development of new aerospace systems is now measured in decades rather than years. Value Placed on Academic Research and Education Industry's Viewpoint The panel's research strongly suggests that mature sectors of the industry value academia principally for its graduates at all levels. Therefore, industry places more value on researchers than on research itself. Masters-level students are especially attractive to industry because they have a broad knowledge of their fields and have not become specialized in a narrow area of interest required for a doctoral thesis. Although numerous technical contributions from academia can be identified (see Box 4-3), the research results per se are not highly valued,
AEROSPACE INDUSTRY 125 partly because improvements in industry performance (defined as increasing shareholder value) can seldom be traced to them. The indirect benefits of aca- demic research are often not recognized by industry, despite their contributions to the knowledge base that may ultimately contribute to the development of new technologies. The contribution of academic research to the development of very large- scale integration is one example. Academic research directly benefited chip makers (e.g., Intel) and indirectly benefited aerospace companies by enabling the development of enhanced avionics. However, research results become visible to management only when they are applied directly to an aerospace system or im- prove a company's performance. Another example is the development of com- posite materials, the underpinnings for which were developed in academia and the applications of which include turbine engines, solid rocket motors, and other critical and noncritical engine and airframe components. The indirect benefits may not be recognized, although they have tremendous value. In contrast, univer- sity graduates are highly valued because they provide direct visible benefits. If academic research is considered in the broad context of innovation pro- cesses, it plays a very large role, along with government laboratories, in laying the foundations of understanding that lead to the next wave of innovation. This role is reflected in recent studies by Diana Hicks and Francis Narin (2001) of CHI Re- search and others (e.g., Spencer, 2001) showing the predominance of academic research papers cited in patent applications. Of the scientific papers cited on U.S. industrial patents in 1993-1994, academic research was the source of 52.1 percent, roughly twice the number of industry research papers (Narin et al., 1997~. In the current environment, future opportunities for academic research will be much more focused on issues that contribute to market success and will require much more flexibility on the part of academic researchers. For industry, the focus of R&D will be success in the marketplace; R&D should generate technological discriminators, improve affordability, and create new business op- portunities. To receive industry support, academic researchers will have to meet industry's needs for timely and usable results. To achieve these results, industry will have to manage its relationships with universities more effectively. Typically, large aerospace companies define key corporate strategic oppor- tunities and fund them first. Subsequent R&D spending authority is dependent on profit and loss. With a few exceptions, large aerospace companies no longer have central R&D laboratories. Instead, they use contract research capabilities when they are available, and they fund universities philanthropically to support centers and institutes for research in specific areas. At this stage, aerospace companies are especially keen on the development of better methodologies and tools, such as tools in computational fluid dynamics and the research results of the Lean Aero- space Initiative. Small companies maintain a very different relationship with universi- ties for a number of reasons. First, there is a significant state and federal
126 THE IMPACT OF ACADEMIC RESEARCH ON INDUSTRIAL PERFORMANCE infrastructure to support small business research, especially the Small Busi- ness Innovation Research (SBIR) Program and state programs that provide loans and matching grants. Second, many states and universities have tech- nical business "incubators" that provide technical and management exper- tise for new technology companies until their ideas are commercially vi- able. Once these firms graduate from the incubators, they have multiple connections to the university for continued research. Third, small firms typically have little or no research staff; contracts with universities, fac- ulty consultants, student cooperatives, and other resources provide a cost- effective mechanism for securing technical expertise. Fourth, maintaining close relations with universities gives small businesses access to informa- tion and cutting-edge technology that they often have little time to obtain on their own.
AEROSPACE INDUSTRY 127 Finally, university research provides intellectual mixing that leads to con- stant discussions and debates, both formal and informal, about new ideas. Uni- versities serve as brain trusts that provide a long-term perspective on ideas and technology development that is difficult to find elsewhere and is exceedingly valuable to small firms. Small manufacturers of UAVs illustrate many of these points (see Box 4-4~. However, industry must learn how to manage relationships with universi- ties. Small companies tend to want technical expertise from universities, while larger firms tend to want system engineers (with technical expertise). Because of academic schedules, universities are not well suited for research on the short- term critical path. For longer term projects, however, academic research can provide the foundation for new products and services that may be critical to a company's business.
128 THE IMPACT OF ACADEMIC RESEARCH ON INDUSTRIAL PERFORMANCE The reexamination of university-industry relationships should not be con- fined to research. Engineering education must also be adapted to an environment in which the context of engineering, the workplace, and the role of technology have changed dramatically. For instance, the ethics of engineering are becoming more complicated. Mission success may conflict with the emphasis on market success, which demands sticking to business-determined costs and schedules. The required capabilities of engineers, especially their ability to work in multidisciplinary teams, are also changing. Many have argued that engineering colleges have not responded well to these changes (see Box 4-5~. Academic Viewpoint The principal measure of success for a research university is the quality of its faculty and student body. Faculty success is measured principally by standing in
AEROSPACE INDUSTRY 129 research peer groups (in exceptional cases very innovative teaching can convey star status). The most prestigious faculty members seem to attract the brightest and most ambitious students. Thus, universities compete vigorously for the re- search elite who are the recognized intellectual leaders in their fields and are willing to expend significant discretionary resources to attract them for laborato- ries, graduate student support, and faculty control of research funds. Thus, major research universities do not depend entirely on external funding to support the research programs that determine their standing in the academic hierarchy. One measure of independence is that most leading research universities now pay their faculty almost entirely from internal funds, rather than through research con- tracts, even though faculty typically spend about half of their time working on funded research projects. Other costs of research include matching-fund require- ments, costs for bids and proposals, and other costs that are not fully reimbursed
130 THE IMPACT OF ACADEMIC RESEARCH ON INDUSTRIAL PERFORMANCE by contracts. In the final analysis, universities contribute substantially to support for research rather than making money on research, as is often assumed. Some of the brightest and most innovative young faculty may find that furthering their careers in academia often conflicts with focusing on developing strong cooperative relationships with industry. Indeed, too much emphasis on developing industry ties can seriously retard their progress. These same young faculty are often poorly informed about the needs of industry and the differences between industry and academic work modes. Thus, they do not provide role models for students, most of whom seek employment in industry after gradua- tion. To address this problem, Boeing introduced an academic fellows program that brings 10 to 15 academics to work in the company during the summer. The goal of the program is to provide firsthand knowledge of how industry operates and to expose academics to real-world technical problems. Leaders at Boeing argue that the only way to overcome long-standing cultural differences between
AEROSPACE INDUSTRY 13 industry and academia is to provide opportunities for interaction and to maintain a dialogue to identify differences and develop potential solutions. Much of the tension between the needs of the academic research system and industry's needs for personnel and technology can be addressed by better man- agement of university-industry relationships. For example, academic involve- ment in problem definition can help focus research on meeting the needs of both. Industry must also recognize that universities cannot work to an industry sched- ule and should choose research projects accordingly. Projects must be interesting enough to attract graduate students and must be structured in a way that is com- patible with the university reward system. A combination of long-term and short- term funding can provide stability for academic programs that may then be in a better position to address shorter term industry problems. Active industry advi- sory boards can help overcome communication problems.
132 THE IMPACT OF ACADEMIC RESEARCH ON INDUSTRIAL PERFORMANCE Academia also has constraints that industry does not always comprehend. Universities have costs and business pressures just like corporations, although in a different context. Because of reductions in government funding and more fre- quent matching-fund requirements, the dollar amounts of individual government research grants have been reduced. Industry grants also tend to be small, averag- ing about $43,000 for one large aerospace firm. Chasing many small grants can be expensive. Wealthier research universities can absorb these costs more readily than most, but the costs are real. Universities may also face legal restrictions imposed by many states on fee-for-service research contracts that might put the university in direct competition with a private research firm. To industry, these types of constraints can be frustrating and may make universities seem unrespon- sive. Industry must recognize these constraints and modify its expectations. Foreign Students A large fraction of graduate students from around the world who attend U.S. universities remain in the United States, attracted to the innovative, dynamic, research environment in this industry. In 2000, of the 215 doctorates awarded nationwide in aeronautical and aerospace engineering, 85 were awarded to non- U.S. citizens, only 10 of whom had permanent visas (NSF, 2001b). Export laws and other restrictions, however, limit the involvement of foreign students in some types of aerospace research. Typically, for companies that have a substantial fraction of business with the DOD or NASA, employing noncitizens is difficult. Even for purely commercial applications, companies are wary of funding aca- demic research performed by foreign students who may take the results to foreign competitors after they graduate. In some cases, firms are legally required to restrict the involvement of foreign students on contracts with universities. Gener- ally, universities find it difficult to accept these restrictions because they cannot guarantee the future use of information generated in their laboratories. Some universities have restricted some of their laboratories to graduate students who are U.S. citizens, but these arrangements may become increasingly difficult to maintain as the percentage of foreign students (and faculty) in science and engi- neering school populations increases. Therefore, industry and government may have to consider changing their policies to ease restrictions on foreign students and faculty, many of whom opt to remain in this country and thus contribute to U.S. competitiveness (NSF, 2001b). Intellectual Property Rights Another key barrier to collaboration between universities and industry is intellectual property rights. Within universities, intellectual property is usually shared by the faculty member and the university; although the professor may be willing to negotiate ownership away, the institution tends to be very reluctant to
AEROSPACE INDUSTRY 133 do so. For obvious commercial reasons, companies generally insist on propri- etary rights to research results produced by programs they fund. Universities, for a variety of reasons, find it either not in their interest or impractical to grant such rights and, perhaps more importantly, to protect them in the courts. As a result, companies must cope with different policies at different universities and even different projects at the same university. Pursuing the negotiating process on a case-by-case basis can be costly. University faculty and administrators sometimes have inflated expectations of the value of intellectual property, which must be considered in the context of how its value can be realized. If the university and/or faculty member retains ownership but cannot sell it or develop a product from it, the value is much lower than if ownership is assigned to the company, and the proceeds are shared in some negotiated way. Ownership of intellectual property is not the only issue that must be ad- dressed. Mechanisms are also needed to protect intellectual property, especially in universities with a culture of disbursing knowledge. In Georgia, state laws were changed to allow Georgia Tech to protect intellectual property for three years, which is usually long enough to meet commercial needs. This arrangement has the added advantage of teaching graduate students to respect intellectual property as they will be expected to do in industry. Arrangements to Improve Research Collaborations In the course of committee and workshop discussions, a number of sugges- tions were made for improving research collaborations between universities and industry. Representatives of large and small companies and academics had quite different ideas. Generally, large companies called for changes in the structure of academic research organizations to make them more compatible with industry needs; universities called for more flexibility on the part of companies and a recognition of the limitations under which they operate. Research Partnerships Abroad Some of the problems associated with industry-university collaboration might be solved by the formation of exclusive, one-on-one research partnerships; indus- try partners would conduct research, in return for support from the single, indus- trial partner, and universities would give up the right to accept support from other companies. An example of this approach is Rolls-Royce's university technology centers (UTCs), a program established in 1990 to focus and coordinate Rolls- Royce university research in the UK. The firm targeted university departments with proven track records and expertise in technical areas of strategic importance to Rolls-Royce. UTCs at 12 universities in the UK conduct about half of the research funded by Rolls-Royce across a wide spectrum of short-term and
134 THE IMPACT OF ACADEMIC RESEARCH ON INDUSTRIAL PERFORMANCE long-term needs. The UTCs have quarterly management meetings and annual reviews with Rolls-Royce managers, and they compete for funding to ensure that the company gets value for its money. UTCs are also part of the firm's recruit- ment and training strategy, ensuring access to the best people the UK with the necessary technological training. University Technology Partnerships (UTPs), intended to complement the UTCs, involve suppliers, customers, and other Euro- pean universities. For example, the Rolls-Royce UTP in Engineering Design Processes involves British Aerospace and the Universities of Sheffield, Cam- bridge, and Southampton. It was launched in 1998 to conduct a joint program of research into engineering design processes for the twenty-first century. U.S. Research Partnership One of the presentations at the workshop described the efforts of GE Aircraft Engines (GEAE) to develop a new strategy for funding academic research, simi- lar to the strategy used by Rolls-Royce but tailored to the issues specific to U.S. universities. The strategy, called the University Strategic Alliance (USA), is intended to shift the company's current pattern of funding of 140 small contracts at many universities to funding of much larger contracts at a few universities that could become long-term partners. The change was partly motivated by a 66 percent drop in the company's internal engineering staff, which has made the company anxious to make more effective use of university capabilities. GEAE' s idea behind USA is to integrate university research into the firm' s business strategy and technology road maps. The company would enter into a guaranteed, performance-based, five-year contract with sufficient funding to en- sure a critical mass of capability at each university. Contracts would be focused on specific problems, such as film cooling issues on high-pressure turbine air- foils, in real-world contexts. The company's intent is to ensure reasonable aca- demic freedom, reasonable protection of intellectual property rights, and oppor- tunities for publication (with some restrictions). The universities would have access to company technology. The company would share ownership of intellec- tual property with the university but would generally not pay royalties. Because protecting intellectual property can be a difficult issue, terms would be negotiated with each university. In some cases, the company might own the intellectual property rights but allow publication; in other cases, it might share patents. The same university could have different contracts for the treatment of intellectual property for different projects. A person with technical knowledge on each side would participate in the negotiations so that decisions would not be made solely by legal departments. The company's goal is to enlist academic research to solve problems, many of which require fairly basic research. The company would allow university researchers to identify ways the problems could be solved, as long as those approaches could actually be implemented and would result in solutions in one to
AEROSPACE INDUSTRY 135 three years. The company envisions that faculty and students will come to its facilities to learn about problems in a real-world context and will pursue solutions in both company and university facilities. From the company's point of view, this kind of company-university interac- tion would have significant benefits, particularly compared to the relatively little value the company now receives from the academic research it funds. From the university perspective, however, it may be difficult to participate despite assur- ances of long-term funding. A particular concern is that the problems posed by the company might not be attractive research subjects to faculty members be- cause they would probably not involve problems at the intellectual frontier of the field and, therefore, would not be supported by the academic reward system. The arrangement might also be perceived as too restrictive in terms of publication and the university's freedom to work with other companies. Finally, the company may not gain access to the best researchers, who can raise research funds from other sources without the restrictions imposed by the company and so may not perceive any benefit from participation. This example highlights some of the difficulties in improving university- industry research collaborations. The proposed strategy addresses the company's desires to have universities perform productive research and attempts to accom- modate the university's need for open publication, protection of intellectual prop- erty, and stable funding. Nevertheless, U.S. universities may not consider this plan in a positive light. CONCLUSIONS Conclusion 4-1. Academic research in aerospace is changing rapidly. The gov- ernment policy change reducing government support for aerospace in general means support for research in aerospace will also be reduced. Henceforth, market forces will determine how things change, and industry and universities will have to adjust accordingly. Conclusion 4-2. Although industry values academia for turning out educated graduates and, to a lesser extent, for its research results, industry is not willing to support generalized research programs. Industry will provide support for pro- grams with clearly identifiable impacts on its performance. Conclusion 4-3. In the more mature parts of the aerospace industry (e.g., air- frame and engine manufacturing), better methodologies and tools are the most valued research products. These include tools in computational fluid dynamics and the results of the Lean Aerospace Initiative. In less mature industry sectors (e.g., UAVs, MAVs, and satellite-information systems), new concepts and physi- cal understanding will be most beneficial.
136 THE IMPACT OF ACADEMIC RESEARCH ON INDUSTRIAL PERFORMANCE Conclusion 4-4. Industry considers academic research contributions most valu- able (i.e., most relevant and easiest to transfer) when they result from cooperative programs with industry. Conclusion 4-5. Issues concerning proprietary rights pose substantial barriers to improving collaborations between academia and industry. Resolving these issues will require compromises between industry's desire to protect intellectual prop- erty and academ~a's desire to maintain a free and open intellectual community. Conclusion 4-6. Formal research partnerships between industry and academia are evolving. For new approaches to succeed, however, the academic reward system will have to find a way to recognize the value of close collaboration with industry. NOTES 1Many engineering colleges have merged aerospace engineering departments with other depart- ments, typically departments of mechanical engineering. 2Because the five subsectors studied by the committee do not align with the industry subsector definitions used to report sales and shipments data, these numbers are rough estimates that draw upon data from the following sources: AIA, 2001; McGraw-Hill/DOC, 2000; and CRS, 2003. 3In 1986, commercial launches accounted for only 13 percent of commercial and government launches (excluding space shuttle launches). By 1996, commercial launches accounted for half of the total (McGraw-Hill/U.S. Department of Commerce, 1998). 4Researchers at the University of Pennsylvania developed the underpinnings of laminate theory in the 1960s. Since then, Michigan State University, Case Western University, Stanford University, University of Connecticut, and University of Wyoming, among others, have contributed to the un- derstanding of fiber-matrix interactions and to the modeling of mechanical analysis of compos- ite materials. 5Satellite LLC purchased the assets of Iridium in late 2000 and relaunched operations of the 73 LEO satellites previously launched by Iridium. An additional five satellites were launched in February 2002. See Washington Post, February 12, 2002, p. E4. REFERENCES AIA (Aerospace Industries Association). 2001. Aerospace Facts and Figures. Washington, D.C.: Aerospace Industries Association. American Society for Engineering Education. 1994. Engineering Education for a Changing World. Washington, D.C.: American Society for Engineering Education. CRS (Congressional Research Service). 2003. Unmanned Aerial Vehicles: Background and Issues for Congress. Washington, D.C.: Library of Congress. Available online at: http:// www.fas.org/irp/crs/RL31872.pdf. [August 21, 2003] Hicks, D., and F. Narin. 2001. Strategic Research Alliances and 360 Degree Bibliometric Indicators. Pp. 133-145 in Strategic Research Partnerships: Proceedings from an NSF Workshop. NSF 01-336. Arlington, Va.: National Science Foundation. IUGREEE. 2001. Vision 2010. Available online at: http://wwwl.eng.iastate.edu/iugreee/. [June 24, 2003] Kline, S.J. 1995. Conceptual Foundations of Multi-Disciplinary Thinking. Palo Alto, Calif.: Stanford University Press.
AEROSPACE INDUSTRY 137 McGraw-Hill/U.S. Department of Commerce. 2000. U. S. Industry and Trade Outlook '00. New York: McGraw-Hill. McMasters, J., B.J. White, and T. Okiishi. 1999. Industry-University-Government Roundtable for Enhancing Engineering Education. AIAA 99-0281. Reston, Va.: American Institute of Aero- nautics and Astronautics. Narin, F., K. Hamilton, and D. Olivastro. 1997. The increasing linkage between U.S. technology and public science. Research Policy (26)3: 317-330. NASA (National Aeronautics and Space Administration). 1997. Report of the University Strategy Task Force. Washington, D.C.: Office of Aeronautics, NASA. NRC (National Research Council). 1995. Engineering Education: Designing an Adaptive System. Washington, D.C.: National Academy Press. NSF (National Science Foundation). 2001a. Research and Development in Industry: 2000. NSF 03- 318. Table A-3. Online at: http://www.nsfgov/sbe/srs/nsfO3318/start.htm. [June 24, 2003] NSF. 2001b. Science and Engineering Doctorate Awards: 2000, S.T. Hill, author. NSF 02-305. Arlington, Va.: Division of Science Resources Statistics, National Science Foundation. Spencer, J. 2001. How relevant is university-based scientific research to private high-technology firms? A United States-Japan comparison. Academy of Management Journal 44(2): 432-440. Toon, J. 2001. Flying on Mars. Research Horizons 19(1): 19-23. Available online at: http:// gtresearchnews.gatech.edu/reshor/rh-fOl/mars.html. [June 24, 2003] U.S. Bureau of the Census. 2002. Statistical Abstract of the United States 2002. Washington, D.C.: U.S. Government Printing Office. Available online at: http://www.census.gov/prod/www/ statistical-abstract-02.html. [June 24, 2003] Womack, J.P., D.T. Jones, and D. Roos. 1991. The Machine That Changed the World. New York: Harper-Collins.
138 ADDENDUM Questionnaire The following questionnaire was sent to selected individuals in various seg- ments of the aerospace industry, some of whom attended the December 1998 workshop. Included among the questionnaire respondents were senior executives at Aerospace Corporation, Boeing, Draper Laboratories, GE Aircraft Engines, Lincoln Laboratories, NASA Lewis Research Center, Northrop Grumman, Or- bital Sciences Launch Systems, Orbital Space Systems, Rolls-Royce Allison, SAIC, TRW, Inc., and professors with expertise in aerodynamcs, heat transfer, high-speed instrumentation, and turbomachinery from Iowa State University, Ohio State University, and Carnegie Mellon University. STUDY OF THE IMPACT OF ACADEMIC RESEARCH ON INDUSTRIAL PERFORMANCE 1. Please identify the targeted sector or sectors of the aerospace industry with which you are most familiar. Space-based information systems Launch vehicles Transport aircraft Unmanned aerial vehicles ~ Gas turbine propulsion systems 2. Please describe any cases you are aware of in which the contributions and impact of academic research to seaports) have been clearly evident. What were the circumstances that led to the favorable outcome? (If possible, please supply references to published information.) Please use additional pages, as necessary. 3. Overall, would you describe the impact of academic research on industrial performance in your rectory) of the aerospace industry as (please put an X in one box): 1. very large 2. large 3. medium 4. small ~ 5. very small/nonexistent 4. The Panel has identified a number of mechanisms (listed below) by which it believes academic research has an impact on performance in the aerospace
AEROSPACE INDUSTRY 139 industry. Please rate the importance of each mechanism using the same 1-5 scale used above [where 1 is "very large" and 5 is "very small/nonexistent." Where possible, please write down under each item some specific contributions you are aware of that have been made by academic research to industry via the mecha- nism. Please use additional pages, as necessary. A. Research-related Education end training (of graduate students) i. et the M.S. level ii. at the Ph.D. level B. Invention and Innovation For example, please consider any specific patents owned by a uni- versity or specific innovations resulting from university research that have been of benefit to your business or industry. C. Consulting This refers to faculty providing expert advice to the industry. D. Technology Filtering This refers to the role of academic research in helping compa- nies identify research/technological opportunities and research/ technological "dead ends." E. Tools and Productivity This includes the development of analytical, computational, and experimental tools and methods that are adopted by industry. F. Pontification, Professionalism, Foundations This refers to the function offaculty in documenting and presenting information critical to aerospace; in conveying relevant informa- tion to others in the profession, investors, and legislators; and to service to the profession, for instance, in organizing and sustain- ing professional societies, industry associations, and other trade groups. 5. Are there systematic trends that will either increase or decrease the impact of academic research in the future? For example, the Panel perceives that indus- trial in-house research is being de-emphasized as a result of downward pressure on engineering staff sizes. Is this a trend you perceive also? If so, will this result in more or less opportunity for productive research partnerships between industry and academe? Are there other trends of importance? Please use additional pages, as necessary.
140 THE IMPACT OF ACADEMIC RESEARCH ON INDUSTRIAL PERFORMANCE 6. What changes are required, if any, in academic research if it is to be responsive to trends and emerging challenges in the industry? Please use addi- tional pages, as necessary. 7. To what extent is academic research relevant to aerospace justified for the research results it produces versus the academic infrastructure (faculty, post- docs, other research personnel, facilities) it supports, which enables the other contributions outlined above in question #4? Please use additional pages, as necessary. 8. Do you see any downside to enhancing university-industry research col- laboration? Things to be avoided? Please use additional pages, as necessary. 9. Other comments? Please use additional pages, as necessary.
AEROSPACE INDUSTRY Workshop Agenda WORKSHOP ON ENHANCING ACADEMIC RESEARCH CONTRIBUTIONS TO THE AEROSPACE INDUSTRY December 4, 1998 National Academies Building 2101 Constitution Avenue, N.W. Washington, D.C. 9:00 a.m. Introduction to Study and Summary of Status J. Kerrebrock 9:20 a.m. Summary of Findings from Questionnaire T. Mahoney 9:30 a.m. Session I: General Discussion 141 How have changes in the aerospace industry, as well as changes in academia, especially engineering, affected university-industry re- search cooperation? How might the future of research cooperation be different from the past? What role does academic research play in the total research enter- prise (industry, university, government) in this industry? 10:15a.m. Break 10:30 a.m. Session II: Discussion of Specific University-industry Interactions: Presentations by Industry Participants criteria to determine what types of research universities do best managing the research interaction: definition of deliverables, monitoring progress · long-term vs. short-term collaborations · single firm vs. consortia-based collaborations · stellar examples of success and failure 12:00 p.m. Working Lunch
142 THE IMPACT OF ACADEMIC RESEARCH ON INDUSTRIAL PERFORMANCE 12:30 p.m. Session II Continued: Presentations by Academic Participants: · sources of research funding · preferred research for industry and expectations of industry · managing interaction with industry, both for specific research projects and overall · long-term vs. short-term collaborations · single firm vs. consortia-based collaborations · stellar examples of success and failure 1:30 p.m. Session III: Detailed Discussions (in breakout groups, if desired) 2:30 p.m. Presentation and Discussion of Findings 1. Can best practices be identified? Do they correlate to research with significant impact? Can the impact from academic research be clearly differentiated from the impact of internal and govern- ment research? 2. How might university-industry research collaboration be man- aged better to meet the needs of both? 3. How essential is industry participation in academic research, as a funder, definer, and active participant, both from the industry viewpoint and to meet the academic mission of research and edu- cation? 4. Is it possible to maintain the desired educational role of academia, in the absence of traditional academic research programs? If not, then how do we motivate and justify academic research, espe- cially at the Ph.D. level, in the absence of a strong need for its research output? 3:30 p.m. Formulation of Findings, Conclusions, and Recommendations 4:30 p.m. Adjourn
AEROSPACE INDUSTRY Jack L. Kerrebrock, chairs Professor of Aeronautics and Astronautics Massachusetts Institute of Technology William F. Ballhaus, Jr. Corporate Vice President Science and Engineering Lockheed Martin Corporation John S. Baras Director, Center for Satellite and Hybrid Communication Networks University of Maryland Jewel B. Barlow Director, Glenn L. Martin Wind Tunnel University of Maryland Robert A. Delaney Chief, Design Technology Rolls-Royce Allison Robert M. Ehrenreich Senior Program Officer National Materials Advisory Board National Research Council Antonio L. Elias* Senior Vice President, Advanced Programs Orbital Sciences Corporation *Parley member 143 Workshop Attendees Kenneth C. Hall* Associate Professor Department of Mechanical Engineer- ing and Materials Science Duke University David Heebner (retired) SAIC Robert J. Hermann* Senior Partner Global Technology Partners, LLC Wei H. Kao Director, MMTC Structural Materials Department The Aerospace Corporation Kent Kresa* Chairman, President, and CEO Northrop Grumman Corporation John S. Langford Chairman Aurora Flight Sciences Corp. James McMichael Program Manager DARPA/TTO Art Roch Director, Advanced Technology Commercial Aircraft Division Northrop Grumman Corporation
144 THE IMPACT OF ACADEMIC RESEARCH ON INDUSTRIAL PERFORMANCE Arun K. Sehra Deputy, Aeronautics Directorate NASA Lewis Research Center S.K. Varma Aerospace Consultant Bethesda, Maryland David C. Wisler Manager, University Programs and Aero Technology Labs GE Aircraft Engines Consultant to the Aerospace Industry Panel Thomas C. Mahoney Director, WV-MEP West Virginia University NAE Program Office Staff Tom Weimer, Director Proctor Reid, Associate Director Robert Morgan, NAE Fellow and Senior Analyst Penny Gibbs, Administrative Assistant