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.
Assessing Infrastructure Needs: The State of the Art D. Kelly O'Day and Lance A. Neumann INTRODUCTION The Infrastructure Problem There is a growing impression that America's basic public facil- ities its highways, bridges, and water and sewer systems-are badly deteriorated. The national press, trade associations, and now the general public are concerned that these basic systems are in danger of collapse. Each new public works problem is taken as additional proof of the pending crisis facing the urban infrastruc ture. The widespread concern about the condition of the nation's public facilities stems primarily from two sources: the obvious deteriora- tion of highly visible facilities like interstate highways and the fact that both capital and maintenance spending on public infrastruc- ture has been reduced in constant dollars in the past decade. Public works capital investment as a portion of the gross national product has dropped from 4.1 percent in 1965 to 2.3 percent in 1977. Many state and local governments, faced with pressing budget problems, have been forced to reduce capital rehabilitation as well as oper- ating and maintenance budgets for public facilities. A 1981 survey by the American Public Works Association indicates that noncap- ital public works budgets were reduced during the 1970s, if the 67
68 PERSPECTIVES ON URBAN INFRASTRUCTURE effects of inflation are considered. The average relative decrease in the 1970-1976 period was 0.18 and 0.66 percent per year for U.S. cities and counties, respectively. The comparable decreases for the 1977-1979 period were 0.44 and 1.55 percent per year. This means that public works operation and maintenance budgets, in constant dollars, were decreasing at an accelerating rate during the 1970s. This trend of declining capital and maintenance budgets raises serious concern about the future condition of public facilities; main- tenance cutbacks are bound to shorten the useful life of facilities at a time when capital rehabilitation funds are limited. At every level of government, elected officials and public works administra- tors are raising questions about the existing and likely future con- dition of public facilities. The heightened awareness of a potential infrastructure crisis has spawned a host of so-called needs studies that have attempted to define the magnitude of the problem ahead in dollar terms. To date these efforts have not created a comprehensive data base or con- ditions assessment, particularly for county and municipal facility systems. Rather, these studies, based on a variety of methods and assumptions, have focused on particular elements of the infrastruc- ture system. While a precise definition of the problem has eluded us, in some areas (particularly highways, bridges, and transit), additional funds are being made available and the rate of system rehabilitation will be increased. Additional needs studies will be conducted (the re- cently passed federal transportation act calls for a $3 million na- tional infrastructure needs study) to further define the problem. Our purpose in this paper is to examine how these additional needs studies can best be used to guide decisions on both the level of investment appropriate for an infrastructure system and the allo- cation of available resources to specific facility improvements and maintenance strategies. National Needs What are the nation's infrastructure needs? Accurate and reliable cost estimates are not available, but various investigators have developed projected needs based on limited data. Table 2-1 sum- marizes a variety of national estimates of capital-expenditure needs for the next 15 years. These estimates represent $2.5-3.0 trillion, a staggering amount of money, equivalent to the 1979 gross na t
ASSESSING INFRASTRUCTURE NEEDS TABLE 2-1 National Infrastructure Needs, 1982-1997 Sector Nonfederal-aid highway Federal-aid highway Bridges Water supply urban rural Wastewater treatment Mass transit Jails Othera TOTAL Cost ($ billions) 1,000 200-225 50 100-150 45 75-100 80 1,000 2,478-2,578 aIncludes dams, water storage, ports, community buildings, city halls, and recreational facilities. SOURCE: Rochelle Stanfield, National Journal, November 27, 1982. 69 tional product of the United States. Viewed in another way, these costs represent 10 times the estimated total construction activity of the country in 1983, including all public, private, and residential construction. The total nonresidential construction in 1982 was $94.2 billion; the infrastructure-related construction was $25.1 billion, 27 percent of all nonresidential construction or slightly less than 10 percent of all construction. These needs estimates are clearly beyond the country's capabil- ities, raising the question as to whether they are really necessary or represent a "wish list" of projects. Even assuming there is some legitimacy to such overwhelming needs estimates, a critical issue that must be addressed is the degree of priority of different needs within a sector and the appropriate balance in addressing needs in different sectors. The Challenge Ahead There can be little question that the country will face some serious public infrastructure problems in the next decade and beyond. How- ever, there remain some nagging questions about how serious the problem really is and what level of investment is requires! to address it. While elected officials seem increasingly willing to devote ad- ditional resources to rehabilitate selected public facilities, they often have not been armed with the type and quality of information that ought to be a basis for such critical choices. To fell this information
70 PERSPECTIVES ON URBAN INFRASTRUCTURE gap, professionals will have to address questions such as: · Has the quality of facility inventory and conditions assessments been adequate to identify deficiencies in a consistent and precise manner? · Have recent national needs estimates been worth the effort, given the basic methods and assumptions used? · Is it useful to divorce needs studies from a broader investment planning process that ultimately is required to allocate resources to specific improvements? Addressing these questions may lead to the conclusion that im- proved approaches to defining infrastructure needs and estimating required levels of investment are but two steps in restructuring and improving the management of public facilities. While there are many deserving needs that should be met, the challenge confronting this symposium and others like it is to ensure that whatever re- sources can be made available are managed and used in as cost- effective a manner as possible. This will require moving beyond arbitrary definitions of need and design standards to a much more creative approach to public works management. DEFINING NEEDS Needs Versus Desirable Improvements A critical issue involved in defining infrastructure needs is com- municating a realistic sense of the urgency of responding to various levels of need and the consequence of ignoring, or postponing a response to, any unmet needs. Decision makers need to know the real impacts of varying levels of infrastructure investment before they can make meaningful judgments about the appropriate level of resources to devote to the very real and serious problems con- fronting the nation's public investment priorities. In short, for a definition of capital needs to be useful it must describe more than the total dollars required in a particular sector over some long time period. Unfortunately, the approach taken to defining need in many stud- ies has been, and continues to be, very narrow and of limited use- fulness in guiding resource allocation decisions at any level of gov- ernment. In a recent speech focusing on the desirability of a national capital budget, Senator Tsongas (D-Massachusetts) noted that Con
ASSESSING INFRASTRUCTURE NEEDS 71 gross cannot deal with needs estimates ranging from hundreds of billions to tens of trillions of dollars for the highway system. When confronted with such ranges, the needs estimates become almost irrelevant. Needs defined in this traditional manner have referred to the level of investment required to either complete construction of a new system or to bring existing facilities up to some prespecif~ed standard. While the standard or standards chosen often include physical condition and design criteria as well as level of service and demand-related criteria, most needs studies have been based on unstated assumptions about economic, social, and environmental objectives, performance standards, and growth trends. These stand- ards themselves, as well as the ability to clearly relate infrastruc- ture investment and performance levels to broader objectives, par- ticularly economic objectives, are being questioned. The result of the traditional needs approach generally has been estimates of capital investment requirements far in excess of avail- able resources or even the most optimistic projections of new rev- enue sources. As a result, many needs studies have been viewed as self-serving and lacking any real credibility. Long-range needs es- timates, in the abstract and independent of short-range budget de- cisions, are difficult to understand and often are not very useful. Ultimately, tough priority decisions have to be made about how to spend available resources, and too often needs studies have provided no real guidance on how to separate desirable improvements from critically important investments required to maintain essential service levels. It is this lack of differentiation between various levels or priority of needs that has severely limited many past needs studies. Undue emphasis has been placed on coming up with a total dollar amount without careful analysis of the underlying assumptions or the ef- festiveness of providing varying levels of investment. One factor that has led to an overly narrow definition of needs is the view that needs studies are primarily a vehicle to lobby for additional funds. Of course, demonstrating a level of need and jus- tifying particular funding levels is a necessary and important part of the capital investment planning process. However, almost any program can demonstrate a large backlog of unmet and deserving needs. While one approach is to develop needs lists or wish lists for each sector requiring capital investments and hope that decision makers guess well, a sound capital investment strategy is unlikely
72 PERSPECTIVES ON URBAN INFRASTRUCTURE to result from such a process. This is particularly true at the state and local levels, where decisions concerning specific individual fa- cilities must ultimately be made. Alternatively, needs studies can be redefined to relate potential investment levels more explicitly to system performance and pro- vide a clearer sense of the importance of satisfying different levels of investment need. Needs definer! in this way will not be absolute or particularly amenable to naive characterization by one total dollar level. Nonetheless, a broader definition of need and more effort devoted to differentiating the importance of meeting different needs will provide a more effective basis for decisions on both the appropriate level of capital investment and the most cost-effective allocation of any given amount of available resources. The Needs Assessment Process The two key activities in the needs assessment process are: · inventory and conditions assessment of existing facilities both currently and in light of estimates of future usage and · identification of the desired level or levels of maintenance and improvement. Development of a new and broader needs assessment process and a more useful definition of needs requires a careful analysis of the appropriate approaches to both these activities. The inventory and conditions assessment of existing facilities is a relatively straightforward and value-free task in theory. In prac- tice, however, given the enormity of the job for some public infra- structure systems, a wide range of approaches has been taken for getting some estimate of current conditions. To the extent that the inventory and conditions assessment is a detailed facility-by-facility appraisal by trained professionals based on sound engineering data and measurements, debate over the current condition and projected future condition of facilities can be minimized. However, when the conditions assessment is based on a sample of facilities, performed by relatively inexperienced staff or performed using very crude assumptions and measures of conditions, questions about the real condition of a particular system will frustrate any attempt to define the capital investment needed to maintain or improve the system. Again, it is recognized that needs studies conducted at different levels of government are often satisfying different objectives and
ASSESSING INFRASTRUCTURE NEEDS 73 must rely on different analysis approaches. Our primary concern is viewing needs studies from the perspective of state and local governments that ultimately have operating responsibility for maintaining urban infrastructure. There have been many attempts to define highway conditions (and ultimately investment needs) based on a small sample of road segments. A recent study of all rural public facilities was based on a survey of a very few local officials in a very large number of rural communities rather than a very detailed study of actual facility conditions in a much smaller number of communities. Although such approaches may serve some limited objectives at the federal level for getting an idea of how bad the problem might be, they may not provide a very sound basis for determining appropriate capital investment levels and strategies. Ultimately, good invest- ment decisions generally will require a detailed appraisal of each facility compared with overall system conditions, and it would be much better for any serious needs assessment process to start with such an appraisal. While not perfect, the national bridge inventory and inspection program offers an example of an attempt to provide a comprehensive and sound facility conditions data base as a foun- dation for judging the nation's bridge needs. The critical components of an inventory and conditions assess- ment process should be: · an overall description of the system (location, physical descrip- tion, capacity, etc.~; · structural integrity; · the quality of service and level of usage; · safety; and · the role of each facility in the overall system (i.e., some func- tional classification). While the specific data and criteria that are appropriate will vary widely depending on the type of system (highway, water distribu- tion, sewer, etc.) being examined, detailed and specific information on the items identified above will be critical both to define mean- ingfu} investment needs and to maintain, improve, and manage the system over time, irrespective of the level of resources allocated. Once an inventory and a conditions assessment of existing facil- ities have been completed, the appropriate level or levels of main- tenance and improvement can be determined once assumptions about future usage and facility conditions have been made. As described
74 PERSPECTIVES ON URBAN INFRASTRUCTURE earlier, this element of the needs process has often been accom- plished in an extremely mechanical way by comparing each facility to a prespecified set of standards and simply defining needs as the cost of bringing each facility up to standards. While such an ap- proach has an attractive simplicity to it, the results in general will not provide a good basis for determining the appropriate level of investment and how to allocate funds among various facilities. The appropriate level of investment in any component of public infrastructure will depend on the effectiveness of a particular in- vestment level in meeting a variety of economic and social objectives compared with investments in other infrastructure systems or other programs. While measuring the effectiveness of public investments has always been difficult, some measures of effectiveness and output can be defined for each infrastructure system. Explicitly relating investment to output, however defined, will almost always be pref- erable to assuming that a set of design standards can serve as an adequate proxy. Some studies are already moving in this direction. For example, federal highway needs studies that have been con- ducted periodically since the late 1960s have increasingly stressed performance criteria and the monitoring of system conditions. In. fact, a soon-to-be-released federal highway performance study will include an analysis of the economic impacts of highway improve- ments and systems performance. However, further steps can be taken to evaluate the real effectiveness of varying levels of highway investment, and many needs studies still rely exclusively on design standards to define needed levels of investment. The approach to defining investment needs recommended here recognizes that, from a practical point of view, the appropriate level of investment in any particular facility will depend on many factors in addition to the specific physical conditions of, and quality of service provided by, the facility. The general condition of the rest of the system of which the facility is a part; the role and importance of the facility in the overall system; and the total resources avail- able all should influence the type of improvements that are ap- propriate for any particular facility. In fact, given the interdepen- dencies between all the components of an infrastructure system and the relationship between the appropriate level of improvement for a particular facility and the total budget available, it may often be necessary to define several investment levels or scenarios. The ef- fectiveness of each potential investment level, in terms of perform- ance and impact criteria, would have to be evaluated before the
ASSESSING INFRASTRUCTURE NEEDS 75 appropriate or "needed" investment level could be defined with any degree of rigor. Except for certain minimum criteria, as discussed in the next section, design standards or policy may have to be determined as part of the investment needs and resource allocations process not as an input to or a constraint imposed on that process. While in general we are arguing for a broader definition of needs and a more complex needs assessment process, we do recognize that there may be instances in which decision makers simply want to know what it would take to bring elements of the public infrastruc- ture up to certain standards. It is simply our feeling that when such analyses lead to widely varying ranges of needs or needs estimates far in excess of reasonably available resources, the results are of limited value and a poor guide to where to allocate less than the "needed" amount of investment. The Role of Standards While in many needs studies there has been an overreliance on using design standards as a yardstick for measuring needs, there is an important role for standards in defining capital infrastructure needs. Standards can help ensure that consistent approaches are used in improving similar facilities, provide for compatibility of all elements in a system to ensure continuity of service, offer potential cost savings by limiting the scope of potential improvements, and, most important, represent one key mechanism to provide for public safety through good engineering design. Given these important functions, the issue is not whether design standards are required but what level of service (including safety) they should reflect and how much flexibility should be allowed to tailor improvement or maintenance strategies to particular facilities. There is a critical need to reexamine current standards applicable to each public system. Questions that need to be addressed include: · Have standards risen too fast to be realistic guides for wholesale rehabilitation of extensive existing infrastructure that has been put into place over many decades? · Do older facilities really have to meet new facility standards? · Have the reliability versus risk assumptions embedded in cur- rent standards created too great a margin of safety for a given facility in light of systemwide rehabilitation needs?
76 PERSPECTIVES ON URBAN INFRASTRUCTURE When standarcis are being set by different levels of government, an additional question must be asked about different perceptions of needs and good design practice at the federal, state, and local levels. Again, there will be cases in which consistency and uniform stand- ards may be appropriate, but in many others the costs of such uniformity simply does not make sense given a backlog of critical structural or safety needs that can be addressed with "substandard" approaches. The overriding issue in the debate over the appropriate level of standards is whether it is better to improve a few facilities to strin- gent standards or many more to lower standards. Obviously the answer depends on what the lower standards are and what they imply for safety, service, and ultimately the cost-effectiveness of the pattern of investments proposed for an entire infrastructure system. Summary It is becoming clear that the concept of needs embodied in the traditional infrastructure needs study is inappropriate for dealing with the problem of allocating funds to a range of infrastructure areas or particular facilities within~one area. Typically these tra- ditional studies: els; · have been unrelated to what might actually be accomplished with less than the "needed" level of resources; · have definer! projects that may or may not be cost-effective investments, irrespective of the actual budget available; and · have been no real guide to tough priority decisions at the fa- cility-by-facility level. What is necessary to replace this approach is a process that relates the investment level that is thought to be needed to the productivity or effectiveness of those investments compared with different in- vestment levels and ultimately investments in other areas as well. Thus a broader needs approach requires a more refined evaluation process to ensure that scarce funds are employed most procluctively. Because traditional needs studies have been divorced from the process of fund allocation to infrastructure areas, geographic re- gions, or specific projects, the results have tendec! to be estimates · have not reflected any policy choices or alternative service lev
ASSESSING INFRASTRUCTURE NEEDS 77 of dollar needs far in excess of the funds available. Such studies have been of limited usefulness in guiding investment decisions on how to effectively use a much smaller amount of money. If the funds for a major capital investment identified as needed by such a study were simply not available, the study would not be of much use in determining where smaller investments would be most effective. In short, needs cannot be defined in a vacuum. Consideration of realistic budget levels, multiple objectives, and the most efficient way they might be met requires that needs studies evolve as part of a much broader allocation and investment planning process. THE CONDITION ASSESSMENT PROCESS Defining Conditions The previous section pointed out the neec} for accurate, reliable information on facility conditions as the first step in determining need. This section discusses condition assessment in some detail both to review current practice and to suggest approaches that have been successful. Condition assessment inclucles the process of measuring the phys- ical condition of facilities, using specific, clearly defined indicators. It should be based, to the extent possible, on observable and meas- urable indicators to limit judgment and ensure consistency. Various studies have used readily available fiscal measures like mainte- nance budget trends as surrogates for condition measures. It should be clear that maintenance investment per year, even if it is (lropping over time, does not indicate the existing condition of facilities. Like- wise, capital investment trends do not indicate current condition. Condition Assessment Measures Condition assessment is a critical element of the overall needs assessment process. It should be reviewed in some detail. Facility condition must be assessed on several dimensions, including safety and structural integrity, adequacy of capacity, quality of service, and system role. The overall condition of a facility is actually a composite of its rating on each dimension. In addition, it is necessary to recognize that these dimensions embody inherent value judg- ments that affect the evaluation. For example, most would agree that the structural safety of a bridge is more important than its ride quality.
78 PERSPECTIVES ON URBANINFRASTRUCTURE Safety and StructuralIntegrity Protection of public safety is a primary requirement for public facilities; therefore all facilities must be struc- turally sound and safe for public use. The potential risk from structural failure varies widely by type of facility, with bridges representing very high risks and water and sewer lines considerably less risk. It is nec- essary to evaluate the structural condition and its importance to public safety. Capacity All infrastructure facilities are designed to provide a specific level of service, generally measured in units per time. Examples in- clude highways with capacities in thousands of vehicles per hour and water facilities with capacities of millions of gallons per day. Facilities must be evaluated to determine whether they have adequate capacity to meet demand. It should be recognized that most facilities have a dual capacity, that is, they are rated at so many gallons per day at a given pressure. The rated capacity can be increased by reducing the level of performance. This point is important to remember, because it is possible in many cases to slightly reduce the level of performance, increase the rated capacity, and not significantly affect the public. Quality of Service Measures of the quality of service vary widely by facility type. Smoothness of ride is used as a major criteria in evalu- ating street pavement, while measures of reliability of service can be used to measure the quality of water and sewer lines. Role The relative importance of facilities varies from minor to ab- solutely essential, depending on how the temporary loss of the facility would affect the overall system, the number of people or households involved, and the extent of inconvenience. As an example, the loss of pumping facilities at a city's water treatment plant would have a much greater impact on the water system than a break in a small main on a local city street. The pump loss could disrupt water service to the entire city, causing widespread disruption, while the main break may only inconvenience a few households. In evaluating the condition of facilities, it is necessary to consider both the physical condition and the facilities' role so that priorities can be established. The national bridge inventory and inspection program is an excellent example of a systematic, well-designed condition as- sessment process. The next section discusses it as a good case study of how condition assessment should be conducted. A second case study, on New York City's water distribution system, is presented to show
ASSESSING INFRASTRUCTURE NEEDS 79 how agencies can use readily available records to define conditions better and prioritize replacement needs when it is not possible to do more detailed inventories of each facility. Age In contrast to these case studies, an approach that often has limited value is comparing the annual level of replacement and re- habilitation activity with the system inventory to determine the re- placement cycle for the facilities that can be compared to their esti- mated useful lives. For example, water mains are often reported to have a "useful life" of 100 years. If a city replaces 1 percent of its mains each year, then the replacement cycle would be 100 years, the same as the useful life. If the city replaced 0.5 percent per year, the replace- ment cycle is 200 years, double the useful life. This approach has several major flaws that limit its value as an infrastructure needs assessment measure: (1) there are no uniformly acceptable estimates of useful life; (2) facility deterioration depends on many: factors, including initial design, quality of construction, use of the facility, quality of maintenance, and the conditions to which the facility has been exposed; and (3) it does not recognize actual facility performance. Useful life estimates for infrastructure are subject to wide variations because of major differences in design practice, the loads and stresses experienced, weather effects, and maintenance practices. Thus average values have limited use because they ignore the sub- stantial differences in the forces that cause deterioration, stresses, wear and tear, and corrosion. Differences in weather, design practices, and maintenance are often more significant in assessing replacement need than the facility's age. Finally, the replacement cycle approach ignores the actual operating performance of the facility as an indicator of replacement needs. De- cisions on replacement should be based on actual conditions, not simply a facility's age. Many facilities need to be replaced or rehabilitated before their theoretical useful life, while others may not need replace- ment for many years after. The actual operating experience provides the best indication of facility condition. The National Bridge Inspection Program In response to the collapse of the Silver Bridge in West Virginia in 1967, the U.S. Congress mandated a national bridge inventory and inspection program for all highway bridges over 20 feet in length. Originally only bridges in the federal aid system, eligible
80 PERSPECTIVES ON URBAN INFRASTRUCTURE for federal bridge rehabilitation or replacement funds, were in- cluded in the inventory. Now all highway bridges are inspected every 2 years and are eligible for federal bridge funds (though Congress mandates that 65 percent of the funds must be spent on the federal aid system). State transportation and highway departments are responsible for conducting the inventory and inspection program, which is often done in cooperation with local officials. The Federal Highway Administration mandates that 90 inventory and appraisal items be recorded for every bridge, using a standard rating sheet like that shown in Figure 2-1. Many states use their own rating forms and often collect more information than is needed to meet the federal requirements. The condition and appraisal ratings for items 68-72 on the rating sheet are based on a 9-point scale. To help guide decisions on individual bridge priorities, the Fed- eral Highway Administration develops a sufficiency rating using the same inventory and appraisal data base used to generate the needs estimate and to allocate funds, but it uses a different analysis procedure. A bridge sufficiency rating is developed to identify the specific bridges that are eligible for funds for replacement (suffi- ciency rating of 50 or less based on a 100-point scale) or rehabili- tation (sufficiency rating of 50-801. While the sufficiency rating combines a variety of structural and service characteristics, it is recognized that any such rating system is somewhat arbitrary. Thus, while the sufficiency rating is used to screen bridges for eligibility for replacement and rehabilitation, states and local communities are not expected to and generally do not set bridge priorities strictly on the basis of the federal sufficiency rating. Furthermore, simply because a bridge has a sufficiency rating of 50 or less does not mean that the appropriate action is replacement. Many factors, including a state's total allocation of bridge funds, ultimately determine what improvement is funded for any particular structure. The National Bridge Inspection Program represents an important step in assessing the nation's bridge replacement and rehabilitation needs. It created the first real inventory of the nation's bridges and has attempted to develop a consistent data base ant! rating system for each structure. It has provided a basis for estimating national needs and lobbying for funds, allocating funds, and identifying the specific bridges with the most serious deficiencies. Notwithstanding these positive results, the limitations of the national needs estimate and individual bridge ratings must be recognizes} if the funds avail
ASSESSING INFRASTRUCTURE NEEDS STRUCTURE INVENTORY & APPRAISAL SHEET 81 Revised 1 2-78 . IDENTIFICATION CLASSIFICATION 1 State 24 H ighway System 2 Hwy District 3 County 4 City/Town 25 Administrative 5 Inventory Route OOn I iUnder 6 Features I ntersected 26 Functional _ _ 7 Facility Carried by Structure 8 Structure No. 1 of STRUCTUR E DATA 42 Type ServiceCo de 9 Location 27 Year Built 43 Structure Type-Main 10Min. Vertical Clearance, Inv. Rte._ 28 Lanes on Str. under 44 -Approach 11 Milepoint 29 ADT 30 Year 45 No. of Spans-Main 12 Road Section No 31 Design Load 46 -Approach 13 Defense Bridge Description 32 Appr. Rdwy. Width and Shield 47 Total Horiz. Clearance ft. 14 Defense Milepoint 33 Br. MedianClNone~open OCIosed 48 Max. Span Length ft. 15 Defense Section Length 34 Skew 49 Structure Length ft. 16Latitude 35 Structure Flared OYes=No_so Sidewalk_Length_ft. Width _ft. 17 Longitude 36 Traffic Safety Features 51 Br. Roadway Width (curb-curb) ft. 18Physical Vulnerability 37 52 Deck Width (out-out) ft. 19Bypass, Detour Length 38 Navigation Control ~Yes~No_53 Vert. Clearance over Deck 20Toll 39 Vertical ft._s4 Underclearance-Vertical 21 Custodian 40 Horizontal ft. 55 -Lateral-Right ft. 220wner 41 Open, Posted, or Closed 56 -Left ft. 23 F.A.P. No. 5 7 Wearing Surface CONDITION Materiai Condition Analysis 58 Deck 59 Superstructure 60 Substructure 61 Channel & Channel Protection 62 Culvert & Retaining Walls 63 Estimated Remaining Life 65 Approach Roadway Alignment 64 Operating Rating 66 I nventory Rating APPRAISAL Deficiencies 67 Structural Condition 68 Deck Geometry 69 Underclearance-Vertical (Lateral) 70 Safe Load Capacity 71 Waterway Adequacy 72 Approach Roadway Alignment PROPOSED IMPROVEMENTS 73 Year Needed Completed _ Describe (item 75) 74 Type of Service 75 Type of Work 76 Improvement Length ft. 77 Design Loading 78 Roadway Width ft. 79 Number of Lanes 80 Prop. Rdway Improvement-Year 81 ADT 82 Year 83 -Type I Remarks: 84 Cost of Improvements $_, , 000_ _ 85Prel. Engrg. $_, , 000_ _ 86 Demolition $ , , ooo_ 8 7 Substructure $ , , - 000 _ _ 88 Superstructure $ , , -000 891nsp. Date _ By Date Transfer of Date Maintenance Insp. Condition Analysis Appra isal Cost Estimate General Review Rating Rati ng FIGURE 2-1 Bridge inventory and rating sheet.
82 PERSPECTIVES ON URBANINFRASTRUCTURE able for bridge improvements are to be used as effectively as pos- sible. The General Accounting Office (GAO) reviewed the National Bridge Inspection Program in their August 11, 1981, report to Congress, "Better Targeting of Federal Funds Needed to Eliminate Unsafe Bridges." Among the report's many findings were two of importance to this discussion: incomplete, inaccurate, and unreliable data were detected; and inconsistency in bridge inspection ratings was a major concern. The GAO found several instances of incomplete, inaccurate, and unreliable data, including a case in which 10 percent of a state's bridges were inadvertently deleted from the inventory. In addition, there were many cases of missing data, such as estimated costs for rehabilitation or replacement and bridge ownership. Finally, the GAO found that bridge inventory data were not being updated in a timely manner. Many of these data problems have been corrected since the issuing of the GAO report, which pointed out that these problems were understandable considering the substantial effort required to develop a national survey in a 3-year period. The GAO's other major findings of inconsistency in bridge in- spection ratings are more disturbing: · Inspection officials rated bridge components differently because of the amount of judgment involved. · Some states did not use the Federal Highway Administration's bridge description ratings but had developed their own. · The number of officials used to inspect and rate about the same number of bridges and the cost of inspections varied considerably by state. · A wide variance existed in the percentage of deficient bridges between states that appear to have similar conditions. The GAO's findings point out the difficult problems of trying to determine existing facility conditions. The engineering, procedural, and logistical problems are tremendous These problems, coupled with short schedules and competition for federal aid, could seriously distort the survey results. it is critical to recognize the major dif- ficulties in assessing the true condition of facilities and develop programs that obtain accurate condition information, because sub- stantial investments are based on these data. The decisions will only be as sound as the condition information. Since the bridge survey is one of the best examples of a comprehensive and tech
ASSESSING INFRASTRUCTURE NEEDS 83 nically sound national condition assessment program, its problems should be carefully studied to ensure that they are avoided in other surveys of infrastructure condition. The New York City Water Distribution System The National Bridge Inspection Program is an example of a fed- erally funded, well-designed, comprehensive condition assessment program. There are steps that local public works agencies can take to significantly improve their knowledge of facility conditions. The New York City water distribution system provides a case study of how existing records were used to evaluate the condition of a water distribution system without extensive field surveys. While the re- sults are not as valuable as a complete survey of conditions, it does show how agencies can begin to improve their condition assessment by using available resources. New York City has one of the largest and oldest water distribution systems in the United States. On May 7, 1979, the Office of the City Comptroller issued a report, "Rebuilding During the 1980's: New York City's Capital Requirements for the Next Decade." As part of this report, the water distribution replacement needs were evaluated to determine their capital costs. The report concluded in part, ". . . approximately $2.45 billion should be spent over the next decade to replace 39 percent of the City's water main (2,404 miles) and ~ percent of the valves that are at least 60 years old and have exceeded their design life." This recommendation was based on a replacement cycle concept that took the useful life of water mains and valves to be 60 years. Digging up 39 percent of the city streets in a 10-year period would create major traffic problems, but would it solve the city's water main problems? The state of New York, in cooperation with the city, requested the Army Corps of Engineers to investigate the replacement needs for water distribution mains. The Corps, through a consultant, eval- uated citywide water main condition trends and conducted a de- tailed investigation of Manhattan's water mains. Comprehensive investigations of water main conditions should consider structural conditions of the pipes, the hydraulic adequacy of the system, and water quality. A detailed survey of this nature would cost millions of dollars, far beyond the available funds. Since structural deteri- oration of mains was a major area of concern, the study focused on this facet. Water main breaks were used as an indicator of structural
84 PERSPECTIVES ON URBANINFRASTRUCTURE TABLE 2-2 Trends in New York City's Water Main Breaks, 1940-1978 Borough Manhattan Bronx Brooklyn Queens Staten Island New York City Percentage Increase 0.6 2.2 2.5 2.9 3.8 1.7 deterioration, not necessarily because it is the only or best measure, but because it was an acceptable indicator of structural failure that was readily available. Citywide analysis indicated that the rate of breaks was increasing 1.7 percent per year in the 1940-1978 period. The break rate trends, however, varied markedly by borough, as shown in Table 2-2. Com- parison of break rates among boroughs provided additional infor- mation about the variation in water main breaks across the city. Review of the trends indicated that Manhattan had a relatively slight rate of increase in break rates, but it had a much greater rate of failure. Table 2-3 presents a comparison of break rates by borough for all mains, and 6-inch- and 12-inch-diameter mains. This relatively simple analysis indicated that Manhattan was experiencing a break rate 3 to 7 times greater than the other four boroughs. Thus, while Manhattan's rate of breaks was not increas- ing as rapidly as the other four boroughs, its failure rate was much worse. In fact, there were 12 tracts in Manhattan experiencing 14 times the four-borough break rate for 6-inch mains. In order to investigate water main break problems in more detail, a computerized inventory of Manhattan water mains was developed, and break histories for each main were developed. An extensive set of analyses were conducted to define break patterns and to see if a predictive mode! of main failure could be developed. The results indicated several important findings: · The water distribution system mains were not in danger of imminent collapse. · Main break rates were not related to age. · Location may be the single most important factor in predicting main breaks.
ASSESSING INFRASTRUCTURE NEEDS TABLE 2-3 New York City Annual Breaks 1,000 Miles (1975) 85 Area All Mains 68 167 55 Twelve-Inch Mains Six-Inch Mains 56 116 35 158 610 87 Citywide Manhattan Four boroughs Ratio: Manhattan/four boroughs 3.0 3.3 7.0 · Replacement should be targeted to those mains with the worst break records, regardless of age. This study demonstrated that careful analysis and effective use of existing records could significantly improve both (1) the under- standing of facility conditions and (2) the development of capital replacement programs. Targeted replacement based on demon- strated need could reduce costs dramatically when compared with estimates based on useful life and replacement cycle planning. THE INVESTMENT DECISION-MAKING PROCESS Estimating Investment Requirements A number of different approaches have been used to estimate the level of investment required for various public facility systems. Some of these approaches have been very simplistic, reflecting the absence of sound inventory and conditions data. For example, ex- trapolating past trends of investment levels in a particular sector and using a replacement cycle rule of thumb have both been used to forecast future needle. Although such approaches may be neces- sary in the absence of good information on conditions, they are likely to produce only a very crude estimate of real investment needs. When facility conditions data have been available (on the basis of either a comprehensive inventory and inspection or surveys and other indicators d, most needs studies have compared observed con- ditions with a set of accepted design standards to (refine and esti- mate the cost of"needed" improvements. Even when such an ap- proach is applied to the results of a rigorous and comprehensive conditions assessment (as recommended above), the estimates of need may be of limited use. As discussed earlier, uniform design
86 PERSPECTIVES ON URBAN INFRASTRUCTURE standards are often not a good guide to the appropriate level of investment for a specific facility when budgets are tight and there are many systemwide needs. There simply may be no easy shortcut from a comprehensive facility conditions assessment to recommendations about the ap- propriate level of investment. Such recommendations would depend on the specific pattern of facility improvements that can be obtained for any given level of resources. Recommendations on needed levels of investment should therefore be a product of a much broader decision-making process about investments rather than a basis for it. In many cases, it will be more productive to improve the decision- making process at different levels of government than attempt, in the absence of changes to the investment planning process, to de- velop more sophisticated improvement standards or definitions of need. A Framework for Investment Decisions In maintaining existing infrastructure, public works officials face two interrelated questions: · What level of investment is required? · What specific projects should be selected? Officials reach these decisions each year either systematically or in an ad hoc manner. Of course, in many cases the answer to the first question is highly constrained, and sometimes the question is not explicitly addressed. Increasingly, local officials will turn to the staff of the public works agency for guidance. Most agencies make these decisions in a loosely structured fashion. Lists of"necessary" projects are reviewed and those within the available budget re- sources are selected. Decisions are often based on the strength of individual arguments and whether projects are ready for imple- mentation, rather than on a step-by-step review of project needs and the assignment of priorities. This approach, proven and workable, rests on the premise that the important projects will climb to the top of the list. Experience shows that this does happen in most cases. There is no assurance, however, that the most effective mix of projects is selected. In sit- uations in which projected need exceeds available funding, there remains a question of whether the agency is getting the most from its investment.
ASSESSING INFRASTRUCTURE NEEDS 87 To provide a better approach to both estimating the level of in- vestment required and selecting the appropriate mix of specific facility improvements, we suggest a broader investment planning framework. The major purpose of such a framework is to emphasize that decisions about the appropriate level of investment and the specific pattern of facility improvements to fund cannot be treated separately. The major activities in the proposed needs assessment framework are shown in Figure 2-2. While there is a danger in oversimplifying any decision-making process, several critical steps must be taken to develop investment needs estimates: 1. develop a facility inventory; 2. establish performance criteria and conduct a conditions as sessment; 3. identify deficiencies; 4. develop funding scenarios and program priorities; 5. develop and evaluate alternative projects; 6. evaluate program/project alternatives; and 7. select a program option. l I Facilityinventory I and Condition Assessment ]~ j _ _ _ _ _ . . _ . . ... .. ~ 1 Funding Scenarios it__ _. ~ Develop Alternative ~Facility Program Priorities Deficiencies Program I Performance . Measures ~ Or red ~ | Available Budget l I and Staff I 1 r Performance Criteria Research National Standards Economics Studies Future Analyses Develop and Evaluate Alternative Projects LO ~ . ~ Alternative Programs/ Projects , Needs Report Selected Program and Projects FIGURE 2-2 A framework for infrastructure needs assessment.
88 PERSPECTIVES ON URBAN INFRASTRUCTURE As described in the previous section, the first three steps are needed to provide the type of facility-specific information on which to base systemwide investment decisions. These steps are de- scribec! in more detail below. 1. Develop a facility inventory An accurate inventory of all in- frastructure facilities with sufficient technical and mainte- nance information should be established and updated on a regular basis. This inventory should be permanent and should be updated with information on all changes, such as the build- ing of new facilities and major changes to existing facilities. 2. Establish performance criteria and conduct a conditions as- sessment A set of facility performance criteria should be es- tablishec! to evaluate the condition of facilities. These criteria should be reviewed perioclically and should not be overly re- strictive. Current national standards may not be appropriate in many cases because they may not reflect local conditions and problems. Systematic surveys of all facilities should be conducted on a routine basis, e.g., annually or biennially. The nature ant} extent of the surveys will vary by infrastructure type; however, the idea in each case is to assemble consistent information on all facilities so that the current and likely future condition of these facilities can be determined. 3. Identify deficiencies The facility condition survey results shouicT be compared with the facility performance criteria to identify those elements of the system that do not meet the criteria. It is important that the survey data provide sufficient information to allow comparisons of conditions with perform- ance data. Those facilities identified as deficient are candi- clates for follow-up analysis to define alternative projects to correct the deficiencies. The next four steps in the proposed investment planning frame- work focus on how to proceed from a sound conditions assessment to judgments about needed levels of investment and particular proj- ect priorities. They reflect a strong belief that needs estimates (le- velopec! independent of any sense of overall fiscal constraints and explicit priority setting are simply not likely to be very meaningful for setting overall budget levels and will be irrelevant in making facility-level resource allocation decisions: 4. Develop funding scenarios and program priorities While the
ASSESSING INFRASTRUCTURE NEEDS 89 ultimate purpose of needs studies is to provide insight into the investment requirements for a particular facility system, it will generally not be useful to develop facility needs lists in the abstract with no reference to potential overall budget ]@V- els. Definitions of needs and priorities can change dramatically in the face of specific resource constraints. The most practical way to get some measure of the needs being addressed and the urgency of different needs is to apply the discipline imposed by specific funding constraints. These constraints, or funcling scenarios, should be used to show what can be done for various levels of budget above existing levels. In addition to a level of funding being used to define each scenario, some guidelines on the priorities that should be addressed at any level should also be established. These guidelines will necessarily repre- sent explicit value judgments about the most critical deficien- cies uncovered in Step 3. 5. Develop ant! evaluate alternative projects-For each of the fa- cilities identified as deficient (or possibly only for facilities with severe deficiencies), alternative improvement levels or projects should be developed. The specific level of improvement appropriate for any facility will depend on the nature and severity of the deficiencies identified, the priority established for addressing specific deficiencies, the relative condition of other facilities in the system, and the overall resources avail- able. 6. Evaluate program/project alternatives For each funding sce- nario established in Step 4, a program option (consisting of specific facility-level improvement) can be developed to reflect the priority established for different types of (deficiencies (e.g., safety versus capacity) and facilities (e.g., water mains versus distribution systems). By evaluating various program options defined both in terms of a resource level and the deficiencies to be addressed, a much more realistic estimate of investment requirements can be made. While such an approach implies a level of effort and detail generally not a part of the traclitional needs study, ultimately such a process is necessary to allocate available resources; it should simply be extended to generate needs reports as well as a specific program recommendation. 7. Select a program option- While selection of an actual invest- ment program is generally beyond the scope of most needs studies, it is included here simply to suggest the logical con- clusion to the steps outlined above. The selection of a final
go PERSPECTIVES ON URBAN INFRASTRUCTURE program reflects many factors, including the resources ac- tually made available. The steps in this framework for investment decisions describe an approach to investment decision making that integrates the typical needs study in a much broader planning process. The two case studies that follow provide a comparison of two approaches to es- timating investment needs. The Environmental Protection Agency (EPA) survey adopted the traditional ad hoc approach to estimating needs, yet it clearly demonstrates how definitions of need and re- source allocation decisions are dependent on the standards or cri- teria selected and the total resources realistically available. The Wisconsin Department of Transportation, in attempting to convince the state legislature that more resources were required for its high- way program, adopted an approach more like that outlined above. The EPA Needs Survey The Environmental Protection Agency has conducted needs as- sessments since 1973 to obtain information about investment re- quirements for publicly owned sewerage facilities. These estimates primarily reflect needs related to national goals for pollution abate- ment and do not provide a complete definition of sewerage needs. Since these estimates are primarily used to allocate federal grant funds, they may seriously understate or overstate needs. The enactment of the Federal Clean Water Act of 1972 caused dramatic change in water pollution control, including substantial increases in investments in treatment facilities. The law specified minimum treatment levels for municipal and industrial wastes, established specific time schedules for compliance, and enlarged the federal role in funding pollution abatement efforts by local governments. The goals of the act included secondary treatment of all municipal wastewater by 1977, fishable and swimmable streams by 1983, and the elimination of all pollution discharge by 1985. In the subsequent 11 years, there have been many changes to the 1972 act, resulting from experience with this overly ambitious program and the realization that the costs of such an effort were much greater than originally anticipated. In this sense, the bian- nual needs surveys have provided useful information to help re- clirect the program. Table 2-4 presents the results of the EPA needs survey, which
ASSESSING INFRASTRUCTURE NEEDS TABLE 2-4 Environmental Protection Agency Survey of Needs for Water Pollution Control ($ Billions Current Dollars) 91 Needs Category 1973 1974 1976 1978 1980 1982 Secondary treat- ment IIa Advanced second ary IIb Advanced treat 17 13 13 15 6 16 21 29 31 21 4 5 ment 1 1 IIIa Infiltration/inflow 1 5 3 2 3 3 IIIb Sewer infiltration 7 6 5 6 5 IVa New collectors 14 18 17 19 18 21 IVb New interceptors 11 18 18 18 21 18 V Combined sewer overflows 13 31 18 26 37 36 (SUBTOTAL) (60) (108) (96) (106) (120) (118) VI Stormwater needs 235 54 62 114 93 TOTAL 60 343 150 168 234 212 NOTE: Survey results are not comparable due to changes in procedures, definition of allowable projects, and cost estimating procedures. has been carried out every other year from 1974 to 1982. In re- viewing these estimates, it is interesting to note the fluctuations in needs, from a low of $60 billion in 1973 to a high of $343 billion in 1974. These variations are due to the different methods used to compute needs and changing criteria for projects considered eligible for construction grant aid. The EPA conducted the first survey in 1973, using existing project descriptions and leaving gaps where cost estimates were unavailable. In 1974, EPA requested the states to prepare needs estimates that were submitted to EPA for com- pilation. This approach resulted in a fivefold increase in needs over the 1973 survey. The 1976 survey was conducted by an outside consultant with assistance from EPA's regional offices. Certain stormwater control projects were not allowed, resulting in a sig- nificant reduction in needs, compared with the 1974 survey. The 1977 amendments to the Clean Water Act eliminated storm sewer projects as grant eligible, significantly reducing federal ob- ligations for pollution control costs. The 1978 needs survey showed a 30 percent reduction in eligible needs compared with the 1976 survey because of this change in grant eligibility. Other changes include modifications in minimum treatment requirements for coastal communities, again reducing the needs estimates. In the 1972-1981 period, EPA obligated $32.2 billion to 22,000
92 PERSPECTIVES ON URBAN INFRASTRUCTURE projects for planning, design, and construction. By 1981, however, only 4,000 had been completed. Even with this large investment in federal grants, the needs estimates have increased from $60 billion in the 1973 survey to $118 billion in the 1982 survey, if only grant- eligible projects are included, or $212 billion, if stormwater needs are included. These increases cannot be explained by increasing pollution or inflation alone. Rather, they reflect those factors as well as increased information about the costs of implementing the Clean Water Act. Since the early needs surveys included only avail- able estimates of project costs, they did not reflect projects that were later added to the survey as more data became available. In 1981 Congress reviewed the nearly 10 years of progress in pollution control under the Clean Water Act and rewrote the act to reduce the federal responsibility for constructing sewerage fa- cilities. The 1981 amendments cut federal spending from $3.9 bil- lion in fiscal 1981 to $2.4 billion in fiscal 1982. It also reduced the federal share of costs from 75 percent of costs prior to October 1, 1984, to 55 percent thereafter. The needs surveys, with cost estimates that fluctuate because of changing methods and policies, provided EPA and Congress with valuable information on the costs resulting from legislation and regulatory procedures. Partly because of the continually rising costs that far exceeded available funding, grant eligibility and pollution abatement requirements were modified to bring needs more closely in balance with fiscal resources. The needs surveys, while useful in defining overall fiscal implications of the Clean Water Act, were of limited value in setting priorities for specific projects. The needs surveys focused on the treatment facilities required to meet the act's requirements rather than facilities needed to meet water qual- ity goals. As costs have increased and progress has been slow, Con- gress has shifted the overall program toward specific water quality goals, focusing needs more narrowly. The Wisconsin Department of Transportation's Approach to Investment Programming Traditional highway programming efforts in many states have suffered from a variety of problems, including the use of needs studies or system plans that reflect unrealistic revenue assump- tions, the inability to weight factors to allow trade-offs within and between program areas (e.g., bridge replacement versus highway
ASSESSING INFRASTRUCTURE NEEDS 93 rehabilitation or improvement), the lack of a systematic method for maximizing statewide versus local or project benefits, and a failure to explicitly consider the relationship between maintenance and improvement programs. Generally only one design alternative or potential investment level is considered for each project in devel- oping a program, and projects are ranked either subjectively or by using a more technical method such as a sufficiency rating, a prior- ity index, or benefit-cost analysis. For the most part, little formal program evaluation occurs. In short, the process of program devel- opment has been viewed as a somewhat mechanical one of checking off projects on a priority list until available funds are exhausted. To address these shortcomings, the Wisconsin Department of Transportation developed a new approach to highway investment analysis and programming that was designed to: · provide a range of policy choices to top management, not simply a single recommended program alternative; · maximize system benefits over individual project benefits; · consider alternative design concepts (i.e., investment levels) for each project; · explicitly develop alternative programs for evaluation; and · make use of a range of consistent evaluation procedures to eval- uate project and program options. Since the most important objective of the process was to improve the department's investment decision-making capability by provid- ing management with fully evaluated policy choices, it was nec- essary to develop explicit alternative improvement programs. De- veloping meaningful alternative programs required, in turn, project alternatives that is, alternative leYels of improvement for a given highway segment. Under certain program assumptions (e.g., con- strained revenue), the appropriate level of improvement for a given segment might be a resurfacing or minor reconditioning; under other assumptions (e.g., a revenue increase), a more substantial improvement might be warranted. Unless this dynamic relation- ship between the scale of projects and program alternatives is ex- plicitly recognized, a key element of program choice is ignored and program alternatives are simply different combinations of projects, each having only one proposed design. The first step in developing a multiyear program was to thor- oughly assess the highway system's physical and service conditions. The assessment of deficiencies for purposes of the 6-year program
94 PERSPECTIVES ON URBAN INFRASTRUCTURE specifically avoided a needs study approach ant! the reliance on traditional highway standards. It instead recognized that, as a prac- tical matter, definitions of need and deficiency vary from time to time on the basis of factors including the public acceptability of existing conditions, the cost of improvements, and revenue avail- ability, among others. To provide some objective measures of roadway condition, a set of data was collected for the 12,000 miles of roadway in the state system. These data included surface age and pavement condition, accident rates and occurrences, volume-to-capacity ratios, percent no-passing zones, and other geometric and structural criteria. The data for each segment were placed in a computer file for efficient editing, sorting, analysis, and display. Using the computer information system, a series of reports was produced summarizing the extent and severity of various deficien- cies statewide and by district, functional class, etc. These reports were used to help guide the development of specific program alter- natives, to evaluate them, and to summarize program performance. In parallel with the analysis of deficiencies in the state highway system, conditions and deficiencies in the other program areas were identified as well. For bridges, the results of the Federal Highway Administration's Sufficiency Rating Formula and the state trans- portation department's own priority listing based on load-carrying capacity, overall structural condition, and geometries were used to assess replacement needs. The cost estimate prepared for the Fed- eral Highway Administration served as a basis for assessing po- tential improvements on Wisconsin's portion of the interstate sys- tem. Once the screening of deficiencies in the state trunk highway system was completed, alternative improvement plans were devel- oped for those segments judged most deficient. In identifying po- tential improvement projects, emphasis was placed on segments requiring surface renewal within the 6-year program period and on safety, geometric, and capacity deficiencies. The minimum improve- ment alternative proposed for each segment was a resurfacing proj- ect or a resurfacing project plus the minimum structural renewal necessary to support the new surface. Depending on the severity of the deficiencies present, higher levels of improvement proposed for a given segment varied from minor reconditioning projects to major reconditioning, reconstruction, and major new alignments. For each alternative improvement proposed for each segment,
ASSESSING INFRASTRUCTURE NEEDS 95 data on the key design elements, potential impacts, cost estimates, and schedule were collected and placed in a computer file that could be cross-referenced with the deficiency data file to produce sum- maries of the deficiencies addressed by different sets of projects and programs. From the analysis of deficiencies it was clear that a range of key policy issues had to be explored in developing alterative programs, including: · the benefits available from a revenue increase under varying assumptions about how additional revenues might be spent; · the benefits of greater emphasis on safety or capacity improve- ments versus pavement preservation; · the most cost-effective mix of resurfacing and reconditioning work for maintaining some minimum pavement quality; and · the trade-off of funding a relatively few major improvements versus many more small improvements. Given the expected trend in gas tax revenues, the need to explore the potential for a revenue increase and to demonstrate how ad- ditional revenues could be used was identified as the most critical issue facing the Department of Transportation. While the number of alternative programs that could be devel- oped and evaluated was limited, a range of reasonable funding levels was defined for each of four program areas: resurface, recon- struction, and recondition (RRR); bridge replacement; interstate improvement; and major projects. Table 2-5 shows the funding levels selected for each area. They were based on the results of deficiency analyses described earlier, expected federal funding availability, previous program commit TABLE 2-5 Structure of Program Alternatives: Wisconsin Department of Transportation Alternative Program Levels (Millions of 1978 Dollars) Program Area Low Mid High Resurface, recondition, reconstruc tion 200 300 400 Interstate improvements 90 135-195 245 Bridge replacement 70 100 140 Major projects 120 listing of additional projects
96 PERSPECTIVES ON URBAN INFRASTRUCTURE meets, and the policy preferences of top management. The objective was to identify the likely range of expenditures by program area, assuming different total revenue levels and policy directions. Thus, for the general RRR program area, expenditures of at least $200 million (1978 dollars) were estimated to be necessary during the period 1980-1985 simply to meet objectives for surface renewal. Even under the most optimistic scenario (i.e., assuming a major revenue increase), combined with minimum expenditure levels in the other program areas, it was unlikely that RRR program area expenditures would exceed $400 million. Alternative programs for the RRR area were developed using guidelines that identified targets for miles of surface renewal and other criteria as well as overall funding levels. While district offices were given initial funding targets, it was made clear that their final funding levels would depend on a statewide evaluation of initial district submittals and the desire to develop a consistent program from district to district. Figure 2-3 illustrates the criteria that were expected to be used in determining the appropriate level of im- provement; however, overall funding levels and deficiency criteria had to be combined in making project selections. At the lowest funding level for the RRR program area ($200 million), district choices were very constrained by the goal for miles of surface re- newal, and the majority of projects were resurfacing and minor conditioning. At higher funding levels there was increasing flexi LEVEES OF IMPROVEMENT: STATE HIGHWAY SYSTEM RESURFACING RECONDITION 1 RECONDITION 2 RECONSTRUCTION Pavement Age Maintainabil ity Pavement Serviceabil ity Index (PSI) (Minor Reconditioning) _ Plus: Pavement Width Shoulder Paving Minor Shoulder Widening (Major Reconditioning) Plus: Pavement Failure Safety Isolated Curve Federal Aid Eligibility Crest Hazard Safety Geometries Capacity Combinations Plus: FIGURE 2-3 Criteria used to determine the appropriate level of improvement.
ASSESSING INFRASTRUCTURE NEEDS 97 bility to fund major reconditioning projects while still meeting sur- face renewal goals. Some consideration was given to specifying relatively rigid rules or priority thresholds (e.g., accident rate above a specified level, etc.) for projects proposed for higher-level improvements. However, subject to meeting goals for surface renewal, districts were given wide latitude to set priorities. Initially this was a more prudent approach, given variations in conditions from district to district and a lack of agreement on the acceptable range for any threshold cri- terion. More defensible threshold criteria could be set in future cycles, depending on the degree of variation occurring in initial district submittals. The development of alternative bridge, interstate, and major highway programs was also guided by an explicit set of priority and policy guidelines, yet the use of a strict formula was also avoided. For bridges, primary consideration was given to load-carrying ca- pacity and posted limits, overall structural conditions, geometries, age, and traffic levels. For interstate improvements, priority was given to completion of the system and selected operational and safety improvements on existing facilities. Unlike the National Bridge Inspection Program and the EPA survey, the Wisconsin Department of Transportation's approach to capital investment programming was not designed to produce an overall needs estimate for highways based on a set of physical and service standards. Rather, it was developed to provide a realistic appraisal of what could and could not be accomplished with a num- ber of different revenue levels to guide individual project selection. It did serve the purpose of many needs studies by providing the basis for state legislation to increase highway revenues twice since 1980. SUMMARY AND CONCLUSIONS Needs Assessment Issues In evaluating the state of the art in needs assessment, we have identified several issues that should be considered in discussing infrastructure needs. While there are many technical, financial, managerial, and administrative issues, we fee! these three deserve particular attention.
98 PERSPECTIVES ON URBAN INFRASTRUCTURE The Limited Use of Traditional Needs Surveys Traditional needs surveys can be classified as studies that de- termine projected needs based on whether facilities satisfy a given set of standards. These surveys are necessarily general, based on assumptions that, while reasonable at the abstract level, are of limited use at the specific facility level. The broader the survey in scope, the greater the potential of overgeneralizing the situation, because of the use of broad assumptions and the failure to eliminate marginal projects, which, while desirable, could be eliminated with- out undue burden on the public. There are two possibly conflicting desires in defining needs, es- pecially at the national level. The first is to develop total costs so that the magnitude of the problem can be determined. The other is to develop detailed, useful information at the local level where trade-offs and project selections can be made. The National Bridge Inspection Program satisfied both desires well because it was based on sound, technically reliable data that could be used at the fecleral, state, and local levels. Surveys in other areas should follow the bridge prototype by building an information base that is both technically realistic and can be updated. For some infrastructure systems and at some levels of government, surveys of all facilities will not be possible. More- over, caution must be exercised at the local level in interpreting and using the results of such studies for estimating needs and se- lecting projects. Single-purpose surveys to develop theoretical needs should be avoided, because they will not reflect real needs and they do not further state and local information and decision-making requirements. The Importance of Technically Sound Condition Information to Effective Infrastructure Management No one knows the true condition of the nation's infrastructure, because assessment data are highly fragmented and there are major gaps in information. There is a serious need for technically sound and accurate information to avoid unwise investments based on questionable criteria. Public works agencies should place high priority on the tasks of determining and continually updating information about the con- dition of public facilities. This information is necessary to provide
ASSESSING INFRASTRUCTURE NEEDS 99 a base for all investments in maintenance. Without reliable infor- mation, there are no assurances that facility conditions will be affected by investments. Replacement needs can be more precisely determined if sound conditions data are available. Agencies should adopt a three-phase approach to developing re- liable condition data: · develop a facility inventory with a performance monitoring rec- ords system; · conduct routine condition surveys; and · conduct research on facility failures, trends, and alternative solutions. Conditions vary widely from agency to agency and within agency systems. Pragmatic applied research into agency patterns, coupled with sound information in a usable information management sys- tem, will benefit the agency substantially. The Importance of the Investment Decision-Making Process Public works agencies must reevaluate their decision-making process for capital and maintenance investments to ensure that the best allocation of resources is made to programs and to projects within programs. With substantial needs and limited resources, it is critical to get maximum value from available funds. In many cases, this may require significant changes in program manage- ment and maintenance practices. Careful systematic analysis, like that suggested above, cannot increase funding; it can, however, ensure that the available funds are spent more electively. This will require reassessing both capital and operating budgets in setting priorities to ensure that the most critical projects are selected. Research Agenda There are many potential areas of research on the management of infrastructure maintenance. We believe that research on the technical aspects of condition assessment, the role of standards on investment, and the investment decision-making process are fruit- fu! areas for study. Particular areas of research include: · Conditions Assessment Research on existing methods of as- sessing infrastructure for each type of facility should be conducted
100 PERSPECTIVES ON URBAN INFRASTRUCTURE with the objective of recommending measures for assessing struc- tural integrity, adequacy of capacity, quality of service, and facility role. Additional research should also be done to determine facility deterioration rates under varying conditions and the appropriate criteria for measuring and monitoring facility conditions over time. · The Value of Information Systems in Condition Assessment It would be useful to analyze the experiences of federal, state, ant! local agencies in the use of information systems in infrastructure management. Are the full potentials of these technologies being realized? · The Role of Stanclards Research is needed on the technical, economic, and legal implications of the standards promulgated by professional associations, trade groups, and the federal government. Are standards continually rising, and, if so, what are the impli- cations of"standards creep" on infrastructure decisions? More anal- ysis of the cost of additional standards or regulations should be performed. · The Role of Risk Analysis in Infrastructure Decision Making- Are risk analysis methodologies being used in decision making, and, if not, what are their potential benefits in determining con- ditions and evaluating options? · Life-Cycle Cost Analysis Research is needed to determine the most appropriate repair, rehabilitation, and maintenance strategies for different facility systems, including a further exploration of the different maintenance options available for each facility system. · Methods for Evaluating Benefits ant! Costs-Knowledge is very limited about the benefits and costs of various levels of system conditions and performance and the trade-offs involved in main- taining different levels of service. Improved approaches for this type of analysis are needed within each infrastructure area and to com- pare the effects of improving one infrastructure system (e.g., high- ways) with another (e.g., water distribution). BIBLIOGRAPHY American Public Works Association 1981 Public Works Management Trends and Development. Special Report 47. Chi- cago: American Public Works Association. 1981 Revenue Shortfall. Chicago: American Public Works Association. Choate, Pat, and Walter, Susan 1981 America in Ruins: Beyond the Public Works Pork Barrel. Washington, D.C. Council of State Planning Agencies.
ASSESSING INFRASTRUCTURE NEEDS 101 Comptroller General of the United States 1981 Deteriorating Highways and Lagging Revenues: A Need to Reassess the Fed- eral Highway Program. Report prepared for U.S. Congress, March 5. CONSAD 1980 A Study of Public Works Investment in United States. Washington, D.C.: U.S. Department of Commerce. Available from the National Technical Information Service, Springfield, Va. Cooper, Thomas W. 1981 State Highway Finance Trends. Report prepared for the Federal Highway Administration, Washington, D.C. Federal Highway Administration 1981 Highway Investment Practices and Trends. Federal Highway Administration, Washington, D.C. National Chamber Foundation 1981 Transport Tomorrow: A National Priority. Report prepared by Paul O. Roberts and The Center for Transportation Policy Research at the University of Cal- ifornia, Berkeley. National Cooperative Highway Research Program 1980 Synthesis 72: Transportation Needs Studies and Financial Constraints. Report prepared by Thomas F. Humphrey. Washington, D.C.: Transportation Re- search Board. National Transportation Policy Study Commission 1979 Final Report. Washington, D.C.: U.S. Government Printing Office. Neumann, Lance A., and Dresser, Joseph 1980 A New Approach for Analyzing Highway Program Choices and Tradeoffs. Washington, D.C.: Transportation Research Board. O'Day, D. Kelly 1981 Philadelphia Infrastructure Survey. Center for Philadelphia Studies, School of Public and Urban Policy, 4025 Chestnut St., Suite 600-T7, University of Pennsylvania, Philadelphia, Pa. 19104. Peterson, George E., ed. 1981 America's Urban Capital Stock. Six vols. Washington, D.C.: Urban Institute. Phillips, Bruce A. 1980 The Deterioration of U.S. Roads: Estimates of Dollar Needs. General Motors Research Pub. GMR:3515. Reed, Marshal 1981 Principles of Highway Finance. Highway Users' Federation, Washington, D.C. U.S. Army Corps of Engineers 1980 New York City Water Supply Infrastructure Study. Vol. 1: Manhattan. N.Y. District, U.S. Army Corps of Engineers. U.S. Secretary of Transportation 1981 A Revised Estimate of the Cost of Completing the National System of Inter- state and Defense Highways. Report to the U.S. Congress. 1981 The Status of the Nation's Highways: Conditions and Performance. Report to the U.S. Congress. Wisconsin Department of Transportation 1980 Six Year Highway Improvement Program 1980-85. Unpublished report.
102 Harry Ha try PERSPECTIVES ON URBANINFRASTRUCTURE DISCUSSION I basically agree with the major points made by O'Day and Neu- mann, and ~ will expand on their material to suggest a research agenda. ~ particularly like the way they have defined needs as- sessment in a broad way, not only looking at condition assessments, but also looking at specific facilities, the alternatives available, and the levels of available funding. Let me address briefly the problem of choice, drawing on work we have been conducting at the Urban Institute. It is a classic systems problem. First, there is a vital need for inventories and condition assessments at the local level; unfortunately, there are a number of dependent variables that need to be identified. In each specific case a number of objective criteria must be examined, such as the number of water main breaks, the number of people served, etc. Second, there are independent, exogenous variables that over- lap each system. For example, soil conditions, weather, traffic loads, and other demand conditions must be considered. They will vary among local areas and, in many cases, within different parts of local areas. For each situation there will be alternative actions that can be taken. These may range from replacement of a facility to emergency repairs and variations of preventive maintenance. For each possible action there are effects on service quality and costs for the long and the short term. Another important factor is citizen expectation for the level of service a facility provides. There is also, of course, uncertainty about the future and funding constraints. Nonlinearities are common in these problems. including inter ~ ~ 7 0 relatedness. Scheduling situations arise, for example, when it is possible to deal with more than one problem at a time due to an emergency response to a single problem, such as major road repair. (It may be possible and more efficient to repair the sewers and water mains at the same time.) While local governments must make the ultimate choices about what to do, our focus should be on a national research agenda, which can ultimately help local and state governments help themselves. Six major topics should be included in the agenda. 1. Improver' condition assessment tools. While methods have im- proved in recent years, there is room for still more improve
ASSESSING INFRASTRUCTURE NEEDS 103 meet. The basic issue is one of reliability. As pointed out by the example of the bridge survey, the question is whether different raters will come to the same determination with re- gard to the condition of a facility. We can improve both the techniques and the training involved so that more reliable results can be obtained and they can be used by decision mak- ers with greater confidence. We must also be concerned with the cost of those procedures. For local and state government to regularly monitor the condition of their facilities, they must have methods that are relatively practical and inexpensive. Therefore, we should look particularly for low-cost approaches. 2. Improved information on the rates of deterioration of different types of infrastructure. Some work has been done on this topic, but it is just a beginning. There is a need to develop data on deterioration rates that take into consideration such condi- tions as soils, weather, loading factors, materials used, and construction methods rather than merely to provide averages for particular facilities at national or even citywide levels. 3. Better analytical tools for making trade-off analyses. There is a need to use risk, cost-effectiveness, and cost-benefit analyses and other tools to deal with multiple criteria, nonlinearities, and uncertainties. Local governments need to know which technique should be used and to what extent for specific sit- uations. Too many governments use a simple "worst first" approach to determining how to set their priorities. That a facility is near collapse does not necessarily mean that its repair or replacement should be given top priority. It may, for instance, be far more cost-effective to devote resources to pre- ventive maintenance of facilities that are heavily used than to repair a facility that is less important in the functioning of the system. In other cases, such as water mains, a low rate of failure may offer a policy choice of accepting the temporary disruption caused by such failures and to undertake a program of emergency repairs rather than a far more costly systematic replacement program. 4. More comprehensive information on individual maintenance alternatives. This includes information on the service quality, life effects, and durability of a facility. It would be desirable to develop a cost and durability handbook providing ranges of costs, under different conditions, for different types of main- tenance options.
104 PERSPECTIVES ON URBANINFRASTRUCTURE 5. Better information on new technologies. There are a number of new technologies that are available but have not gotten to the public works departments. Information on new technolo- gies is fragmentary and incomplete. Information is especially needed on the conditions under which these technologies are applicable. 6. The issue of regulations and standards, particularly those that are generated nationally. Standards can cause many problems. What is needed is an independent and professional check of standards and analysis of their implications and effects on immediate costs and on operating costs over the long run. These are difficult research issues, but action on them could make the process of choice substantially easier for local governments. Kurt W. Bauer O'Day and Neumann have provided us with a valid critique of the shortcomings of infrastructure needs studies conducted to date. Such studies have tended to produce needs assessments, which, because of the sheer magnitude of the estimated need, have low credibility with elected officials and the public. The chapter clearly identifies the need to improve these studies if they are to be used to guide decisions about the level of investment appropriate and the allocation of available resources to specific improvements within a system. It also identifies the questions that should be addressed to improve needs assessments. In particular, needs studies should address issues of priorities between sectors or within sectors so that available resources can be used in the most cost-effective manner. The key problem is to provide a continuing and accurate inventory of the capacity of facilities to meet both current and future require- ments for use. My principal criticism of the chapter is in what it does not say. It does not deal explicitly with one of the most important issues concerning infrastructure: how needs assessment relates to com- prehensive planning efforts, including the comprehensive land use plan. This issue is raised but not addressed. For instance, they discuss how needs assessments should be related to the broader investment planning process so as to more rationally allocate re- sources between and within sectors. These priorities and allocations should be based on a variety of broader social ant! economic objec
ASSESSING INFRASTRUCTURE NEEDS 105 fives. This is a task that can be accomplished properly only in the context of a comprehensive plan. Indeed, the description of the im- proved needs assessment process set forth in the paper is a descrip- tion of the classic comprehensive planning process. The case studies too are inadequate in their relevance to a com- prehensive planning process. The Wisconsin transportation needs study, for example, is clearly an improvement over earlier studies of a similar type, yet it does not relate alternative improvement and maintenance strategies to the highway system as a whole. It did not consider how the development of the highway system would relate to other mode} systems. These are serious shortcomings impeding the development of the whole system. They are now being corrected by the state transportation department as it integrates its highway plans into the broader state transportation plan. A similar observation can be made with respect to the EPA survey of wastewater treatment systems. It failed to relate needs for water pollution abatement to detailed areawide plans for water quality management, even though the requirement for such plans was fed- erally mandated. The serious shortcomings of any needs assessment process, how- ever technically sophisticated, outside the context of the compre- hensive planning process are illustrated by a few examples. If, for instance, a needs assessment indicates that a section of a combined sanitary and storm sewer system should be reconstructed, how, in the absence of a comprehensive plan, does one determine whether to reconstruct it as a combined or separated facility? The far-reach- ing implications of that question deserve some contemplation in the design of needs assessment systems. Such questions extend to other kinds of facilities and systems, such as wastewater treatment, surface and groundwater quality management, and street improve- ment and maintenance issues. Moreover, such questions extend to issues of land use, development, and redevelopment. The choices made can have important implications for the economic develop- ment of an area as well as for issues of social equity in a community. Similarly, if there is a need for a waste treatment plant, how does one decide, in the absence of a comprehensive plan, the size of the plant, the level of treatment to be provided, and the service area to be used? If a bridge must be reconstructed, how can a decision be made on the design capacity and the level of service to be provided without reference to some broader system plan for transportation and land use? These few examples suggest the need to relate in
106 PERSPECTIVES ON URBAN INFRASTRUCTURE frastructure neecis assessments ant} the maintenance ant! improve- ment process to a comprehensive plan. In summary, the chapter provides a useful critique of the state of the art of needs assessment. It offers some sounc! suggestions for incremental improvement of these processes. But it stops short of aciciressing one of the key policy issues in infrastructure needs as- sessment-the relationship to the comprehensive planning process. Without this relationship such assessments cannot be user! as sounc} guides for determining the appropriate levels of investment in a given infrastructure system or the allocation of resources to specific facilities. Only within the context of the comprehensive planning process can these two important functions be aclequately aciciressecI. More important, only in that context can one determine the extent to which investment decisions meet broacler social ant! economic objectives. Thus, one of the key policy issues is the relationship of infrastructure programs to comprehensive planning objectives. SUMMARY Preparation of Inventories and Needs Analyses The owners of facilities should be responsible for needs analyses and condition assessments, developing an ongoing set of tools ant! processes. Fecleral ant} state guidance in methods ant! stanciarcis can be helpful, but the owners shouIcI clo the actual evaluations themselves. They may need assistance the first time. The bridge survey was successful because it was concluctec! uncler a fecleral- state-Iocal partnership. The fecleral government proviclecI 100 per- cent of the funcling for the survey, but the work was clone at state and local levels. The kind of partnership and the ratio of funcling may cliffer for other facilities, however, such as water systems. The important point is to involve those who must use the information in its clevelopment. Cost is an important consideration in needs assessment. Local governments must be convinced that the benefit to them in im- provec! decision making is worth the cost of fincling out the condition of their facilities. They are more likely to unclerstand the value of needs assessment if they help produce it ant! use it in making · ~ c ,eclslons. Guidance from higher levels is also a key concept. One important consideration is the use that may be macle of aggregates! ciata from
ASSESSING INFRASTRUCTURE NEEDS 107 local sources at the state or national level. Where they may be used to develop formulas for the allocation of funds to the states, as in the case of funding for bridge replacement, it is important that the data be both reliable and comparable across jurisdictions. Needs studies that are driven wholly by federal programs may distort the problem at the local level, however. National Versus Local-Leve! Data The idea that the collection of data should necessarily make for better decisions was challenged, particularly for data at the na- tional level. A major use of the national bridge survey data is to facilitate decisions on resource allocation among the states. At the local level, however, the data are more useful in planning, and most of the necessary data are available at the local level. They are not always good, but there are no magic solutions to the information problem of telling managers what they need to do. Planning, to be effective, should be on a relatively small scale and within the scope of what can be done. National inventory data, it was argued, are of questionable value. Some went so far as to characterize national inventories as a waste of money. The collection of data at the local level should be designed to facilitate local decisions. The process should be one that proceeds from the bottom up and ought not go overboard in the collection of information for its own sake. It was pointed out that the majority of decisions by public works directors are made in a continuing, incremental process. Most of them want to improve the process and the quality of the data they use. The acid test of the utility of condition and needs assessments or inventories is whether they help elected officials in making a case for facility improvements and maintenance. Inventories are often used primarily to justify federal or other funding for projects. We should move from inventories to planning, including a consideration of whether all the infrastructure we have . · . . in every city IS necessary. Interim Assessments The bridge inventory took 5 years to complete. Underground sys- tems are far more difficult to inspect and assess. A critical question involves interim assessments. Many cities already have reasonably
108 PERSPECTIVES ON URBAN INFRASTRUCTURE good information on their experience with breaks and pressure problems in water mains or sewers. Most could improve their ability to establish priorities by relatively simple computerized manipu- lations of these data and could improve their access to the infor- mation already obtained, which could be used as a springboard to a more comprehensive inventory system. There is, of course, a wide spectrum of capability. Some cities, such as Houston and New York, keep track of their experience and use it to set program priorities. Others record nothing. The important thing is to use the infor- mation on experience that is available to avoid having to use rules of thumb, such as replacement cycles, as a basis for capital im- provement planning and programming. A city can get away from generalized numbers and rules of thumb by looking at its own experience. Data Available for Neects Assessments A lot of information is available. The federal surveys of needs for various facilities, for instance, produce a great (leal of data at the state level. The important consideration in developing a local data system is to build as much as possible from the bottom up, while developing and using common instruments and tools. The federal government could contribute most by developing tools for use by local govern- ment, in contrast to mandating that certain data be collected. Needs should be defined in terms of the mission of the agency and the purpose of a facility, rather than in terms of a checklist mentality. This can help identify facilities that are no longer needed, sug- gesting the wisdom of bringing the assessment process into a broader planning process, in contrast to the planning process required by Section 208 of the Clean Water Act (P.~. 92-500), in which building came before planning. Planning involves stating the mission and getting agreement on it. The notion of using comprehensive planning as the context for assessment was challenged, however, on the ground that planning often is not a useful process for those who must make capital in- vestment decisions. Data, it was asserted, are not useful to anyone other than those for whom they are collected. It is important to develop data that get politicians to pay attention to a problem. A comprehensive plan does no more than address the needs of a par- ticular set of decision makers. The task of getting together all the
ASSESSING INFRASTRUCTURE NEEDS 109 knowledge about a problem in one place for a single decision maker is futile. A more realistic goal is to provide specific information to particular decision makers. In response, the Wisconsin local planning process, involving the clevelopment of an annual and a 5-year capital program, was cited as a workable planning system. It was asserted, however, that cap- ital programs were frequently changed and that projects proposed for the last years of any program do not resemble what is eventually built. It was pointed out that a working capital improvement process contemplates changes in the program in light of events and new information and therefore must be a continuous process. It is critical to have the key decision makers involved in the capital budgeting and programming process. One participant suggested that the best argument in favor of planning is the way decisions are currently being made. In planning it is important to create ant! discuss multiyear financial scenarios so that choices can be clarified for those who must make the finan- cial decisions. The iclea that all participants professionals, citizens, and elected officials will agree on the same information and its meaning is utopian. All information is self-serving and should be. Not everyone has the same role in the process. The Qua1tity of Engineering Knowledge Need is not an absolute quality but is often in the eye of the beholder. Infrastructure problems may be satisfied in some in- stances by more efficient operation, upgrading performance instead of building better or rebuilding. In other cases, allowing further deterioration of a facility may be more logical than repairing or replacing it. Professionals shout make the alternatives and the risks involved in these choices more explicit for the political decision makers, understanding that the professional's "best choice" may not be the decisive one, as the political process has the final say. Early on the engineers need to make their own assumptions, de- velop possible alternatives and their risks, and clarify the tracle- offs to give the public an opportunity to make decisions.