A View from the Front Lines of Academic Engineering Research
I will address the forces shaping academic engineering research today from the perspective of one who has been a working engineer for half a century. During this period the nature of engineering, business, and our country has undergone major changes. Even greater changes are indicated for both the near term and the next century.
Many commissions, committees, and boards have been established to address the major problems in industry, government, and academia due to the changing world, and many reports have been written and symposia and workshops held to indicate possible solutions. As I have read the reports and attended the forums, I have found that the views being expressed differed significantly from mine. For the most part, the participants and contributors to these activities were from the executive and administrative offices of industry, government, and universities, prestigious people, decision and policy makers. But where were the people who chose other career paths and continued to do technical work? Would their insights, like mine, be different from those expressed in the reports that were receiving most attention? Are perspectives from executive offices really so different from those from laboratories? It would appear that ''working stiffs'' are perhaps a neglected national resource, so I, with some trepidation, will try to represent them and present a different perspective on the subject. I have modified the subtitle of my paper because I did find at least two reports (National Research Council, 1987; National Science Board, 1992) that express views that are very similar to mine. I am puzzled as to why these have not received more attention.
Until World War II each branch of engineering considered itself a distinct and separate field that was growing in an orderly evolution. It was expected that when students completed a college program in some branch of engineering, they were well prepared for their entire professional career. Their ancillary training in physics, chemistry, and mathematics provided little more than the basic principles of a given field. In effect, current practice was conveyed to such engineers, usually in the form of handbooks or correlations, which were used to solve problems. If the problem being considered was not identical to the known solution, the same formulas were applied but the safety (really ignorance) factors were increased.
The war imposed the need for sonar, radar, the atomic bomb, and many other applications that exceeded both the supply and the capability of engineers. Those products were developed, not empirically, but from basic principles, and physicists were largely responsible for them. This demonstrated rather dramatically that people with a good understanding of general principles could apply them to accomplish, rather quickly, ends that previously took long periods of time to accomplish by crude semiempirical methods. This wartime experience led a number of technological institutes and engineering colleges to believe that if fundamental and comprehensive knowledge of the physical sciences could so remarkably shorten the time from discovery to application, then the sciences and mathematics should receive greater emphasis in their education programs. Thus, in some schools new engineering education programs stressed fundamentals rather than current practice, and this culminated in what is now known as the engineering science curriculum. To further the scientific "coloration" of engineering (a phrase that can be attributed to Donald Frey, Northwestern University), engineering research was introduced in the universities and it has become an important part of the educational enterprise.
After World War II, as the United States assumed the role of a superpower, it was apparent that the nation's defense and economic and social well-being depended directly on engineering. The dominant feature of the environment in which engineering functioned in that period is change. The National Research Council's Committee on the Education and Utilization of the Engineer (1985c) identified four factors as particularly important in that regard for the engineering profession: (1) a large expansion in the roles of government, (2) a rapid increase in the amount of information in daily life and work, (3) the accelerating rate of technological development, and (4) the internationalization of business and the marketplace.
To date the engineering education system has been relatively successful in producing engineers able to cope with the changes, despite such severe constraints as departmental structures established in the nineteenth century,
faculty shortages, uneven and declining student enrollments, obsolete equipment, and funding shortages. The adaptability of engineers to such changes has been attributed to the broad content of physical and engineering science in the current undergraduate curricula (National Research Council, 1985b).
However, despite much success, U.S. industry has experienced strong competition from abroad and has lost many markets in key products. Serious questions have been raised about how to counter those trends and, as a result, a plethora of boards, commissions, reports, symposia, and workshops have been organized around major examinations of the entire engineering profession.
The engineering education system has received its share of the blame, primarily because design and manufacturing (current practice) were not given enough attention in the curricula. Criticism has also been leveled at academic research: it is said to be only self-serving for the faculty to have publication records and that it is irrelevant for industry.
The end of the Cold War, with the associated reduction in defense budgets, the significant decreases or elimination of research in industry, and the policy that research must serve national goals—a policy promulgated by the new government administration with the support of influential members of Congress—all presage even more major changes for engineering research, in particular for such research performed in universities. Therefore, an examination of the nature of engineering research and its role in the profession and for the welfare of the nation is in order.
THE ROLE OF RESEARCH IN EDUCATION AND INDUSTRY
Research has become an important part of engineering education. Research is also performed in industry, more in some industries than in others. It is necessary to understand the purposes, goals, and types of research in these two different kinds of institutions before discussing possible synergisms.
The principal responsibility of universities is to the students, their primary customers. The support that industry gives to universities is certainly helpful, but it is small in relation to the investment made by students (and their parents). Industry's support of research constitutes the smallest source of R&D expenditures at U.S. academic institutions (Dickens, this volume). In Dickens's view, the situation is somewhat better for "organized engineering research units at academic institutions," which receive almost 23 percent of their funds from U.S. business and industry. However, without data
on the specifics of the disbursements, I suspect most of that support goes to just a few of the largest and most prestigious engineering schools. Thus, it is evident that industry is a user of university-developed products, in much the same way as the National Football League is. This clarification, however, in no way implies that meaningful interactions between industry and academe do not exist or are not desirable. On the contrary, it is merely intended to identify clearly the differences in objectives and functions of the two types of institutions.
The university, then, must transmit knowledge and understanding to young people, give them an opportunity to develop their capabilities, help them gain an understanding and appreciation of the world around them, teach them to think independently, and in some curricula, like engineering, enable them to obtain skills that will serve them well throughout their lives.
We have seen that before World War II the skills transmitted to engineering students were those of "current practice," which were based on lore, empiricisms, and intuition. It should be kept in mind that it was during that period that many industries were developed by those means. The wartime lesson that the time from discovery to application could be considerably shortened by fundamental knowledge and research led to major curricular changes that deemphasized current practice, including design and manufacturing. The new curricula did produce engineers who were flexible, versatile, and adaptable and who functioned well during the many and rapid changes that have occurred in engineering since the war.
Engineering research was intended to confront a student, for the first time, with a complex problem that was not well specified, would need defining, and would require synthesis of all the student's knowledge for its solution. In this way the student would experience the loneliness of individual inquiry and the anxiety of the unknown and would develop the discipline, tenacity, and perseverance required for exploring the unknown and for independent thinking. This type of research requires persistent work for a period of several years before the crucial insights and results are obtained. Significant advances have been made in this way, but the main purpose is to provide new and unique educational experiences for the students.
The situation portrayed above would seem to present a clear picture of what is needed for the future: more engineering science education and research. However, there are serious shortcomings to that approach if one desires the products of such an education to be gainfully employed, contribute to meaningful technological developments, enhance an employer's productivity, or help the nation's economic growth.
The engineering science curriculum in attempting to emulate the pure sciences and thereby gain academic respectability, developed courses that were analytic, formalizable, and teachable. Thus, well-posed problems were presented and emphasis was given to solution methods and their results,
which sometimes illustrated interesting physical phenomena. The problems considered were chosen primarily for their mathematical tractability and, thereby, represent highly idealized situations with limited, if any, relation to real engineering systems. Furthermore, the conditions under which the simplified results could be applied to real problems were rarely, if ever, delineated. Although such criticism is mostly made about theoretical studies, it applies equally to experimental work that is designed for ease of observation and measurement rather than its relation to real situations. Much of the academic research has this character as well. It is thus, perhaps, not surprising that academic research is said to be "pure" or "basic" and is relegated to the "ivory tower" and considered useless by industry.
Also, this approach has deprived many students of one of the most essential of all engineering skills, the ability to determine a priori, the essence of a complex situation, that is, to define the meaningful problem. More emphasis is required for problem definition, and consideration of open-ended real-world problems, the type of interest to industry, is urgently needed.
To determine the appropriate role of engineering research in industry, it is first necessary to recall some changes in engineering practice that have occurred in the past half century. Almost all industrial and manufacturing processes were developed empirically, as was the related equipment. Throughout the years, those processes remained essentially unchanged as long as the companies were profitable and the industries were unchallenged. This is true not only of traditional, heavy industries, because many high-tech industries also have empirical origins and processes.
As major U.S. industries began to experience strong competition from abroad, it was suggested that matters could be improved by the use of computers and robots. In fact, in numerous studies of changes in engineering practice, new engineering tools based on the computer are said to be part of a revolutionary change in how engineers work (see, for example, National Research Council, 1985a). Thus, the popular buzzwords associated with modern engineering are terms such as robotics, CAD/CAM and CAI/CAP. The improvements made in this way are most welcome, worthwhile, and overdue. However, it must be understood that, for the most part, the same basic elements of the system (machinery) are employed, albeit faster, more accurately, and more uniformly.
The deemphasis or elimination of design and manufacturing in engineering education is sometimes said to be a contributing factor in the loss of industrial competitiveness. Much pressure is being applied to increase emphasis on those subjects in engineering schools, and considerable federal support is being given to programs dealing with those subjects. What does
not seem to be recognized, however, is that all the changes in engineering have also changed the nature of design and manufacturing, that is, engineering practice has changed. It does not seem to be fully recognized that with demands for products with greater purity, ability to withstand more severe operating conditions than ever before, and greater precision and economy of manufacturing, there is little or no experience or knowledge on which to base such designs. In examining various industries, it is readily apparent that there is a large margin between actual existing industrial systems and the limiting physical behavior, as determined by the laws of nature. Thus, there is great potential for improving industrial and manufacturing productivity by enhancing the effectiveness of the related processes. To accomplish this, it is necessary to "research" the industrial processes, that is, to gain an understanding of the phenomena involved in the process and the factors on which they depend. Vigorous and comprehensive engineering research programs that are directly related to real problems are necessary to develop the knowledge base and physical principles on which advances in design and production can be based. In so doing, gaps in existing knowledge are identified for further study and, also, it is readily apparent that such research is essentially cross-disciplinary.
A National Academy of Engineering (NAE) study committee expressed a similar view: "The technical intensity of most manufacturing and service industries will continue to grow at an accelerating pace, and commercial technology will become increasingly science-based and interdisciplinary" (NAE, 1993, p. 92). The implication of this statement is that R&D activity must be pushed "further downstream into design, production, and marketing, as well as factoring production and marketing considerations into the earlier phases of upstream development activities" (NAE, 1993, p. 31).
It might appear impossible to deal with the complex and diverse phenomena that occur in industrial processes. In fact, most industries either feel no need to apply new knowledge or think their processes are too complex for detailed study and so depend on empiricisms and gross correlations. On the other hand, much academic research is too specialized or idealized to be of much value to industry. Thus, there is now a need for new and intimate relationships between industries and engineering schools so that there can be a coupling of technology with all the latest developments of engineering research, such as the increasing power of theory and computation, meaningful model systems, and sophisticated measurement and diagnostic tools. Such university/industry relationships should not be expected to yield, for example, a generalized computer code that will solve all the company's problems, an approach that is, unfortunately, being pursued too frequently.
The research being advocated here is fundamental engineering research. This is distinguished from fundamental science research in many ways. For
example, science research primarily seeks new knowledge about the natural world without regard for its utility. Engineering research focuses on the man-made world in order to expand the knowledge base and to identify and exploit the physical principles on which advances in design and production can be based (National Research Council, 1987). In many cases there are interactions between science and engineering research, and the boundaries between them are often difficult to discern. However, basic engineering research that provides the underlying competence on which applications or applied research is based is often cross-disciplinary, whereas basic science research is mostly constrained by scientific disciplines. Some basic engineering research does not directly involve the laws of nature but addresses the functional characteristics of large systems consisting of intricate components. The knowledge base for manufacturing, for example, will ultimately consist of engineering principles drawn from many engineering disciplines and activities. There seems to be general misunderstanding of these distinctions, as is evidenced by the usual relegation of basic research to science and applied research to engineering.
The National Science Board (1992, p. 47) stated this crucial issue in the following way:
A pervasive problem in the United States today is the insufficient attention given to fundamental engineering research in industry, government, and the universities. Every firm must have an ever-expanding, relevant engineering knowledge base, and the hardware and software techniques for translating that base quickly into practice, in order to convert ideas into products rapidly and efficiently. The often-cited lack of emphasis on process improvement and manufacturing, along with excessive time delays from concept to available product, attest to a pervasive lack of understanding of, appreciation for, and sufficient attention to the vital role of fundamental engineering research by U.S. industry, government, and universities. Yet, there is no sufficiently broad and deep fundamental engineering research base on which to build; furthermore, there are an inadequate number of engineering researchers in U.S. industry who are equipped to, and called upon to, extend that base as needed. The greater the storehouse of fundamental engineering research, and the greater the ability in industry and government to extend it as needed for proprietary or national reasons, the better able the qualified engineer is to innovate in an integrated system of design, manufacture, and maintenance.
That report also says that too little support is given to process-oriented R&D. U.S. industrial R&D is weighted much more heavily toward product technology than process technology. In relation to their Japanese counterparts, U.S. firms also allocate a disproportionately small share of their R&D budgets to the search for new or improved processes.
The perspectives presented above provide a basis for meaningful interactions between industry and academe. However, some observations need to be made first to gain an appreciation of why such relationships have not developed to the degree required.
Birnbaum (1994) states the university's position as follows: "Unfortunately, it has become all too common to place the onus for the supposed failure of basic research to contribute to economic competitiveness on the basic research sector."
At a recent meeting on "World Leadership in Basic Science, Mathematics, and Engineering" sponsored by the Office of Science and Technology Policy, there was throughout a tacit assumption that industry was waiting impatiently for research results, which were not forthcoming. That is contrary to my experience and does not seem to be representative of industry's position.
Basic research has not failed—it has fulfilled its mandate. The universities have not failed; if anything, they have succeeded too well in educating large numbers of excellent scientists and engineers. What has failed is industry's ability to translate the fruits of basic research into products and profits. The reasons for the failure are manifold—too numerous to enumerate at this point. It is important to point out that the source of failure of industry to capitalize on available basic science is primarily a failure of management and not a failure of the scientists and engineers (Birnbaum, 1994).
Concurrence with this conclusion from the industrial side seems to be presented by Armstrong (1993, p. 5), who states that "responsibility for deficiencies in our industrial performance rests largely with failures in the private sector, failures of strategy, investment, and training—in short, failures of management." However, the polarity between the university and industrial positions, or perhaps between the view from the executive office and the workbench, is well illustrated by Armstrong's further remarks:
These (failures) will not be cured, or even helped by more research. Trying to cure poor industrial performance in the short term by more university research is like asking for helpers when pushing on a rope. . . . Poor technology transfer from the university or national tabs to industry has not been a major cause of our competitiveness problem (Armstrong, 1993, pp. 5–6).
It is, thus, not surprising that industry has largely abandoned basic research. Armstrong gives emphasis to the short term and indicates many other factors involved in industrial competitiveness, such as fiscal strategies and policies, marketing, the economic climate, and trade policies. He then states that "it is fair and accurate to say that universities lack deep understanding of products or markets, have no responsibility for development or manufacturing, and tend to overestimate the importance of science in technology competitiveness" (Armstrong, 1993, pp. 6–7). All this is valid, since competitiveness is a highly complex combination of attributes. However, as indicated above, one of the primary skills of well-educated engineers is that they can extract the essence of a complex problem. From that vantage point, I find that the greatest leverage that can be applied, by industry itself, is to "research" its processes (the heart of the endeavor) to improve, modify, or replace them as necessary to bridge the chasm between existing, empirically derived processes and the possible improvements that are more efficient and effective. Orders-of-magnitude improvements in time required to complete a process or in the quality of the product, or both, are possible. Surely, such developments could tip the scales in global competition. Unfortunately, too many executives and policymakers seem to be unaware of the power of technical solutions to industrial problems.
Other arguments diminish the role of research in industry. The Committee on Science, Engineering, and Public Policy (1993) reported that for "industries that rely on high technology but are technically self-contained (such as the semiconductor industry) and industries that do not depend heavily on current science (such as the automobile industry), the results of current fundamental research are generally not decisive." I am not sure what is meant by "technically self-contained," but I do believe both those industries have large margins for technical improvements. That report goes on to point out that Japan, which is not a leading research power, does very well in such industries by "strategies largely separate from scientific research, but highly dependent on engineering." Japan's industries developed after the war in an era when engineering was evolving away from empiricisms and did not carry the baggage of capital investments from another era, as does the U.S. industry. Since research results are not constrained by national boundaries, the Japanese used them freely to send back to us improved products. Whether what they used is basic research or "engineering" is arguable. The fact is that they started fresh with mid-twentieth-century knowledge and an openness of mind to try new ideas, which is in sharp contrast to U.S. industry.
Obviously, there are different views and opinions about the role of academic research in support of industry's needs and the nation's welfare. To define meaningful university/industry relations, the boundaries of participa-
tion must be delineated. The major purpose of academic research is, and should remain, as an essential element in the education of engineers who will later do or supervise the high level of engineering required by industry and the government, including the "proprietary" fundamental and applied research needed when the knowledge base is inadequate. Academic research done in cooperation with industry will be of mutual benefit when both know their respective roles and are prepared to learn from each other. Exhorting universities to do more "relevant" research is likely to be counterproductive. It is unlikely to move an industry forward when that industry does no such relevant research and, more important, when it does not employ a sufficient number of highly educated engineers to use existing relevant knowledge and extend it as needed.
As pointed out by the President's Council of Advisors on Science and Technology,
Some of the cultural differences that have long surrounded industrial research and university research have had the unfortunate effect of unnecessarily inhibiting the most effective interaction between industry and universities. The notion that each sector had its own well-delineated and isolated role and that new knowledge would flow as rapidly as necessary in one direction from the university to industry is completely at odds with today's world . . . . Despite recent gains in building links between U.S. universities and industry, there are still too many individuals in each sector who hold negative perspectives, attitudes, and stereotypes with respect to the other sector. The nation cannot afford to have this situation persist, and much more effort is required to overcome it. Even fundamental research that is not expected to yield short-term answers to industry's problems can benefit from being informed by the technical concerns of industry. Conversely, U.S. industry should have the benefit of easy and immediate access to new knowledge and new talent generated by the universities.
A couple of the relevant "cultural differences" require comment. The time scale for academic research is on the order of years, whereas industry looks for answers in periods of months. Such a mismatch must be acknowledged and addressed in any good partnership. Many of industry's activities are multidisciplinary in that they involve many people other than engineers, such as economists, lawyers, managers, marketers, and the like. Therefore, teamwork is an important and desirable mode of operation. As a result, there are increasing pressures on engineering schools to give "team" experience to the students. This is certainly worthwhile and needs to be done. However, it is being suggested that research also be done by teams, because it too must now be more interdisciplinary. However, inter- or cross-disciplinary means across academic or professional disciplines, and it is different from multidisciplinary.
Independent research develops abilities and qualities that are vital to industry, government, and universities. Engineers with such experience will play an increasingly important role in technical problem solving. Good (1993) states that ''In the future, the complexity of engineering design tasks will require engineers with a doctorate degree.'' Many important discoveries have been made by a small number of very gifted people who were given the opportunity and time to pursue their ideas and intellectual interests. Therefore, independent research must be continued, supplemented by an educational program that emphasizes cross-disciplinary subject matter.
The fact that industries have significantly downsized their basic research or abandoned it completely indicates that a natural role for universities is to carry out the basic research required for industry and that industry focus on applied research and product development. In fact, engineering education would be enriched by consideration of real problems. However, the reasons for the abandonment of basic research by industry must be understood. Obviously, research is not deemed essential, so is industry even interested in interacting in a meaningful way with universities? Also, it is clear that fiscal support cannot come from the universities, and it probably will not come from industry, although it would be less costly than in-house work. Therefore, the federal government will have to be involved, which brings in its own set of problems. Engineering research is an essential area of technical activity that is seriously undersupported in the United States. As the National Research Council's Engineering Research Board wrote in 1987,
This research is essential because all creative technological development in an intensely competitive world rests on it; yet it is undersupported because its central role in the development of productive goods and services is not clearly understood or recognized.
Despite the recent awareness of the increasing cross-disciplinary nature of engineering research, there is little overt support for such activities.
From the perspective of a working engineer, I have pointed out a number of aspects of the changing nature of engineering that do not seem to be widely recognized and that directly impact the matter of academic research in a changing world. In particular, "engineering practice" has changed so that time-consuming empirical approaches are no longer competitive. Because technological advances have surpassed general knowledge, research is now required to develop a knowledge base for design. What is required is essentially cross-disciplinary, basic research, which is different from basic scientific research. Technical solutions to the problem of industrial competitiveness require more process-oriented research.
Academic engineering research done independently by individuals is an essential element of the educational process. Emphasis on problem solution rather than problem formulation is a deficiency of modern engineering education. Consideration of open-ended problems of importance to industry would enrich the education process.
Differences in viewpoints exist between academics and industrial people and between executives, administrators, and managers, and working-level people, just as "cultural" differences exist between academe and industry. If the necessary partnerships are to develop between the two sectors, then those disparate views must be addressed to find the bases for accord. Such dialogues should involve people from all groups, industry, universities, and government, and from all positions, executives, administrators, managers, and particularly, working-level people, who seem to be very much underrepresented.
Armstrong, J. A. 1993. Research and competitiveness: Problems of a new rationale. The Bridge 23(1):3–10
Birnbaum, H. K. 1994. Research Interactions: University, Industry, National Laboratory. Briefing paper for the Forum on Science in the National Interest, World Leadership in Basic Science, Mathematics, and Engineering, Executive Office of the President, Office of Science and Technology Policy, January 31 – February 1, 1994, Washington, D.C.
Committee on Science, Engineering, and Public Policy. 1993. Science, Technology, and the Federal Government: National Goals for a New Era. Washington, D.C.: National Academy Press.
Good, M. L. 1993. Industry needs and the curriculum. Issues in Engineering Education, Vol. 2, October. Washington, D.C.: Board on Engineering Education, National Research Council.
National Academy of Engineering. 1993. Mastering a New Role: Shaping Technology Policy for National Economic Performance. Report of the Committee on Technology Policy Options in a Global Economy. Washington, D.C.: National Academy Press.
National Research Council. 1985a. Engineering Education and Practice in the United States: Foundations of Our Techno-Economic Future. Report in the series Engineering Education and Practice in the United States, by the Committee on the Education and Utilization of the Engineer, Commission on Engineering and Technical Systems. Washington, D.C.: National Academy Press.
National Research Council. 1985b. Engineering Employment Characteristics. Report in the series Engineering Education and Practice in the United States, by the Committee on the Education and Utilization of the Engineer, Commission on Engineering and Technical Systems. Washington, D.C.: National Academy Press.
National Research Council. 1985c. Engineering in Society. Report in the series Engineering Education and Practice in the United States, by the Committee on the Education and Utilization of the Engineer, Commission on Engineering and Technical Systems. Washington. D.C.: National Academy Press.
National Research Council. 1987. Directions in Engineering Research: An Assessment of Opportunities and Needs. Report of the Engineering Research Board, Commission on Engineering and Technical Systems. Washington, D.C.: National Academy Press.
National Science Board. 1992. The Competitive Strength of U.S. Industrial Science and Technology: Strategic Issues. Report of the Committee on Industrial Support for R&D. Washington, D.C.: U.S. Government Printing Office.
President's Council of Advisors on Science and Technology. 1992. Renewing the Promise: Research-Intensive Universities and the Nation. Washington, D.C.: U.S. Government Printing Office.