Near the end of August, 1971, my advisor signed the paper certifying that my dissertation on Analysis of Systems of Constructed Facilities was accepted, fulfilling the last remaining requirement for completing my Ph.D. studies in M.I.T.’s Department of Civil and Environmental Engineering. My thesis had been that decision makers—that is to say, the designers and managers responsible for building, operating, and maintaining highways, dams, houses, and other types of constructed facilities—should have as their goal to provide the facilities’ users with system that exhibit qualities of satisfactory performance throughout a defined service life and in a relatively efficient manner. The novelty lay in bringing together in an explicit and operational way four ideas that were at the time coming into focus in our society and the literatures of engineering, architecture, economics, and political science: First, the concept of a facility’s “performance†has many dimensions. Second, what performance is “satisfactory†depends on users’ values and choices; in a pluralistic society, there will always be debate. Third, the long service lives of constructed facilities, measured in decades or centuries, mandate explicit consideration of uncertainties and risks that performance may become unsatisfactory; something may have to be done in the future to correct the situation. Finally, the resources used to deliver performance cannot adequately be measured on any one scale of value; efficiency can be judged only in relative terms, by comparing available options.
My approach to enabling decision makers to accommodate these ideas drew on principles from economics, psychology, and mathematics to represent performance in terms of three primary measures: serviceability, the degree to which the facility satisfactorily provides the services that users want; reliability, the probability that service will remain satisfactory throughout the facility’s service life; and maintainability, an indication of the effort that may be required for maintenance and repair to ensure satisfactory service. Serviceability, reliability, and maintainability are not independent, and each may be increased—in principle—by using more resources. The decision maker’s problem consists, I asserted, in devising and choosing among available options a design or management strategy that offered the best mix, the optimum performance.
Enactment of the National Environmental Policy Act in 1969 (NEPA) was a tangible demonstration of the emergence of a new way of thinking about constructed facilities, or to use more recently popular terminology, our civil infrastructure or built environment. The law’s timing was fortuitous. As a young professional with a newly minted degree in hand, I became engaged in a thriving consultancy practice, helping government agencies learn how to make their decisions about our infrastructure in a more open, public forum and taking more directly into account the values that a broadly-based user community may place on such resources as parklands, historic associations, wildlife, and clean air.
The one resource that everyone recognized, of course, was money, and infrastructure decision makers soon realized that they needed more of it than in the past to deliver this enhanced concept of satisfactory performance. Limited budgets and competing demands for public-sector spending—notably in the early 1970s on growing military and health-care programs—meant that tradeoffs had to be made. Maintenance might be neglected or planned repairs deferred.  Of course, one can argue that this was just the crest of wave that had been swelling for decades, but by the end of the ‘70s, some people were growing alarmed at what they saw as an impending infrastructure crisis. When America in Ruins: Beyond the Public Works Pork Barrel (Pat Choate and Susan Walter, Council of State Planning Agencies, Washington, 1981) was published, it made headlines in the nation’s leading newspapers, a rare feat for any discussion of constructed facilities (later reprints changed the secondary title to The Decaying Infrastructure). The book argued that the United States as a nation had been investing too little in its infrastructure and in the wrong places for a long time, and the nation’s economy was now at risk.
There followed a decade of federal government studies and intense debate among economists about just how important infrastructure is as a foundation supporting the economy and just how fragile that foundation might have become. The debate formed a backdrop for renewed consideration of performance as a useful facilities-management concept, and by the early 1990s I found myself at the National Academy of Sciences, working with a committee of diverse professionals tasked with recommending how best to measure and improve infrastructure performance. We visited several cities, meeting with municipal and state officials and private-sector professional responsible for building and operating a wide range of infrastructure facilities. The committee’s report, Measuring and Improving Infrastructure Performance, was published in 1996 (Washington, National Academies Press). We observed that practices then current for measuring infrastructure performance were “generally inadequate.†Performance measurement was typically undertaken because the effort was mandated by law or regulatory requirements, or when there was a specific problem to be solved, not because of any broad acceptance that performance measurement is an effective management tool.
More important was the committee’s recommendation that no single measure of performance can adequately represent the varied and complex societal needs that infrastructure is meant to serve. As the report’s summary expressed it, “Performance should be assessed on the basis of multiple measures chosen to reflect community objectives, which may conflict…. The specific measures that communities use to characterize infrastructure performance may often be grouped into three broad categories: effectiveness, reliability, and cost. Each of these categories is itself multidimensional, and the specific measures used will depend on the location and nature of the problem to be solved.â€
The committee’s concept of performance had similarities to what I had proposed 20 years earlier. “Effectiveness†was described as the ability of the system to provide the services the community expects…not so different from what I had defined as “serviceability.â€Â The term “reliability†was used in essentially the same way in my dissertation and the committee’s report. What I had earlier considered as “maintainability†is now more understandably referred to as “resilience†and incorporated as an aspect of reliability. Describing “costâ€â€”deriving from multiple resources and distributed throughout a facility’s service life, but definitely dollar-denominated—as a measure of performance was the major difference from my thesis and an important insight.
While historians may claim causal connections between events separated in time and space, such connections are fundamentally uncertain unless supported by explicit testimony from the people involved in later action linking their motivations to the earlier occurrences. Having myself met twice with Congressional staff to discuss these matters and delivered to them copies of Measuring and Improving Infrastructure Performance and other documents presenting similar perspectives, I would like to imagine that what I and others have learned about infrastructure performance influenced the most recent transportation reauthorization bill Moving Ahead for Progress in the 21st Century (MAP-21, Public Law 112-141, enacted in July 2012) , which features a new federal emphasis on performance measurement. Section 1203 of the act asserts that “Performance management will transform the Federal-aid highway program and provide a means to the most efficient investment of Federal transportation funds by refocusing on national transportation goals, increasing the accountability and transparency of the Federal-aid highway program, and improving project decision making through performance based planning and programming.†(While the U.S. Department of Transportation has for some years issued its biennial Conditions and Performance report to Congress on physical and operating characteristics of the highways, bridges, and transit, MAP-21 is transformative in making an explicit link between performance and national goals.).
The law then states 7 goals that are to be the basis for defining performance, focused primarily on the nation’s highways: (1) safety, reducing traffic fatalities and serious injuries; (2) infrastructure condition, keeping the infrastructure asset system in a state of good repair; (3) congestion reduction; (4) system reliability, improving the system’s operating efficiency; (5) freight movement and economic vitality, improving the national freight network to support trade and economic development; (6) environmental sustainability, enhancing transportation while protecting the natural environment; and (7) reducing project delivery delays, to control costs and promote jobs. Elsewhere the act makes keeping transit system assets in a “state of good repair†a goal as well. The law tasks the Federal Highway Administration (FHWA) and Federal Transit Administration (FTA) with identifying specific performance measures to be used to administer the funding programs covered by the legislation, and with setting targets to be used to judge acceptable performance.
The stated goals and performance measures likely to be selected under MAP-21, while not necessarily comprehensive in their coverage, at least address ideas of effectiveness, reliability, and cost. That it has taken more than 40 years to bring performance-based management into the mainstream one of the principal functional subsystems of the nation’s infrastructure is consistent with the very slow evolution that is a characteristic of civil infrastructure generally.