10
RESEARCH LEADING TO INFRASTRUCTURE IMPROVEMENT
Chapters 3 through 9 have presented the committee's selection of topics for infrastructure research that warrant high priority in the coming decade. Many more topics could be suggested and may warrant research effort. However, the topics included in this agenda comprise a broad scope for researchers and their sponsors, and research resources are inevitably limited.
The committee therefore limited its recommendations to the broad, cross-cutting niches discussed here. Because the boundaries among these infrastructure research niche areas are difficult to define, overlap or duplication among specific research topics may occur within the broad scope of the committee's agenda. Because similar topics can be stated from several perspectives, each possibly leading to fruitful results, some overlap and duplication may be a useful characteristic of a robust research agenda. But it should not become excessive and wasteful. Monitoring of progress is needed.
The committee did not estimate explicitly the reasonable or appropriate cost of research projects that might be developed within this broad scope. Explicit estimates of cost must be based on the specific technical approaches and resources to be applied to each study. Sometimes, initial research results will indicate that resources should be applied differently or that a new approach is needed. Again, monitoring of progress is needed, along with mechanisms for feeding information from evaluation back into the direction of future research.
If infrastructure research is to be effective within this context, there must be a continuing commitment over a period of years. Committee members asserted that the strength of this commitment is even more important than total level of funding in determining research outcomes. The medium-term outlook (i.e., ten to twenty years into the future) that the committee adopted in developing its infrastructure research agenda should apply to program development as well.
As the committee noted early in its study, NSF spending has traditionally been allocated and administered primarily along disciplinary lines, rather than in the cross-cutting patterns that infrastructure warrants. The traditional pattern matches the structure of academic institutions that are the NSF's primary grant recipients. The NSF Directorate for Engineering deals largely with academic staffs in schools of engineering; the younger Division of Social, Behavioral, and Economic Research works with schools of economics and social sciences; and other directorates deal similarly with particular disciplinary fields.
The committee also noted, however, that special programs have from time to time been established to encourage collaboration across academic disciplinary boundaries and to address needs not easily placed within traditional organizational structures. In recent years, the NSF has sponsored creation of research centers focused on such topics as earthquake engineering (at the State University of New York at Buffalo), Portland cement technology (based at Northwestern University), and large-scale structures (at Lehigh University).
These centers typically bring together faculty and students from several parts of the university—or even from several universities and outside institutions1—to work on projects managed under the umbrella of the center's administrative structure. Typically there is also a physical center of activity with offices, equipment, and meeting areas that provide a base of operations and sense of identity for the researchers and the research area.
While their creation and management often pose substantial challenges within the traditional disciplinary organization of a university, such centers offer advantages of bringing together a critical mass of research activity that begins to produce results greater than might be expected of the parts taken separately. Such famous centers as Thomas Edison's Menlo Park, New Jersey laboratory, the Bell Telephone Laboratory, and Princeton University's Institute for Advanced Studies2 are frequently cited as examples of highly productive concentrations of research talent and activity. Other newer research centers have demonstrated success at universities throughout the country.
However, the development and management of infrastructure are largely a local matter. While there are common needs and concerns at the national level the local conditions and demands that infrastructure must
serve vary substantially from place to place, influenced by socio-economic conditions, local political relationships, climate, topography, and a variety of other factors. Centers, which concentrate activity and focus on particular topics, will have difficulty responding effectively to the broad diversity of needs in a comprehensive research agenda. The traditionally widespread and curiosity-driven work of individual researchers is better matched to this broad diversity than are the focused activities of centers.
The committee suggests that both traditional single-researcher efforts and focused centers have a place in infrastructure research programs that grow from this agenda. A program that includes a balanced mix of both types of activity, drawing on research resources and responding to research needs around the nation, is most likely to yield solid payoffs of infrastructure improvement. The research may be based at academic institutions, but could as well draw on the resources of government laboratories now being directed toward commercial activity and civilian concerns.3 This NSF-funded cross-cutting infrastructure research could augment activities at centers established under other programs, such as the 13 University Transportation Centers administered by the Department of Transportation or the NSF's own Engineering Research Centers (ERC).
Regardless of the institutional setting, the research should be undertaken as a partnership of researchers and research users. In particular, municipal and state governments should be participants in the development and management of infrastructure research programs, to assure both that the research reflects problems, and operating conditions actually encountered in U.S. infrastructure systems and that results can be put quickly into practice. Inclusion of private-sector partners (i.e., utility companies, construction companies, and manufacturers) should be encouraged as well. However, financial contribution of all such participants, while highly desirable and often required in other centers programs, should not be a condition for NSF support of a new cross-cutting infrastructure research center. Such a requirement can be a strong incentive to focus research on topics that are closer to commercial development than is the intent of this agenda. Nevertheless, the committee recommends that all NSF-sponsored infrastructure research should be guided by the contributions to be made to practical improvement of the nation's infrastructure, with substantive input from public agencies and industry.
Because the range of research disciplines encompassed in this crosscutting agenda is broad, the committee recommends that a single, unified program announcement should be issued to invite proposals. This announcement should span all areas of infrastructure research funded by the NSF, and all seven niche areas recommended here.
Such a broad announcement will not match the scope of any current single program or directorate within the NSF, but rather bridges several
operating units. Some adjustment in NSF organization and procedures may be needed to relax constraining influences of traditional disciplinary relationships and points of view. However, the NSF's ongoing internal studies to develop CIS programs suggest to the committee that such change will not be particularly disruptive and may already be under way.4
Can broad programs of infrastructure research be justified in economic terms? As described in Chapter 2, infrastructure R&D spending has declined substantially in recent years. Committee members pointed out that the nation's private sector routinely expects that financial returns on assets should be 3 to 7 percent annually or more. If the nation's infrastructure, with a value probably exceeding $3 trillion, yields a similar return, the system's aggregate contribution to the nation's economy would be between $90 and $210 billion annually, equivalent to approximately 2 to 4 percent of GNP (gross national product).5 In addition, infrastructure needs are global, and new infrastructure technology can have high export potential.
If NSF-sponsored research can improve the productivity of the nation's infrastructure by only 1 percent, the benefits would amount to perhaps $1 billion to $2 billion annually for years into the future. But productivity improvements are not the only justification for infrastructure research. Research results will help to conserve and extend the productive life of existing infrastructure assets, and will preempt the failures of infrastructure that, while rare, have such high nonmonetary costs for a community.
The conduct of the research will contribute to the education and training of a new generation of infrastructure professionals. These new professionals will enter their disciplines with a functional appreciation of the multidisciplinary nature of infrastructure and its impacts on the economy, the environment, and society. New teamwork among traditional disciplines will be fostered as well.
Most importantly, new knowledge and technology resulting from this cross-cutting infrastructure research program will ultimately improve the quality of life throughout the nation. By global commercialization and transfers of technology, these benefits can spread to other countries as well.
These many benefits represent a high return to be gained from new investments in infrastructure research, and increased investments may be justified. However, realizing these benefits through infrastructure innovation is achievable only if research results are ultimately put into practice. While support of undirected research that can lead to serendipitous results should certainly continue, priority should be given to research that can plausibly be linked to practical infrastructure improvement. Imagining these links is analogous to preparing a marketing plan for new products nearing readiness for commercial release. The committee has taken a first step in imagining the research topics and improvements to be gained, but
it is up to researchers and users of the results to achieve real improvements in infrastructure performance.
President John F. Kennedy and his brothers adopted the words of playwright George Bernard Shaw: "Some men see things as they are and say 'Why?' I dream of things that never were, and say, 'Why not?'" Infrastructure research should be an effort to dream of improvements and find ways to make them happen. That is the aim of the committee, and should shape the NSF's role in supporting infrastructure research.