Guidelines for Integrating Safety and Cost-Effectiveness into Resurfacing, Restoration, and Rehabilitation (3R) Projects (2021)

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.

11   Process for 3R Project Development This chapter presents the process for development of 3R projects, including the identification of candidate 3R projects and the role of safety and traffic operational considerations in design decisions. 3.1 How Does the Design Process for 3R Projects Differ from That for New Construction and Reconstruction Projects? The current design process for new construction projects is based primarily on the dimen- sional design criteria presented in the AASHTO Green Book (4) and in the design policies of individual highway agencies. It is appropriate to use established dimensional design criteria for new construction projects because, in such projects, there is no existing roadway with a safety and traffic operational performance history that can be used to guide the design process. Established dimensional design criteria provide an aspirational goal for the design of recon- struction projects. Where a roadway is being fully reconstructed, design improvements may be feasible with limited additional cost, except where such improvements would substantially affect adjacent development, established communities, or sensitive environments; in these situations, highway agencies typically seek a design exception to minimize such impacts. Resurfacing, restoration, and rehabilitation projects are usually initiated on the basis of the need for pavement resurfacing and are most appropriately considered as maintenance activities. A performance-based design process provides the basis for design of 3R projects focusing on the decision about which projects should be resurfaced without accompanying design improve- ments and which projects should have design improvements incorporated. The design process for 3R projects begins with the recognition that the project will be imple- mented on an existing road whose past safety and traffic operational performance is known and should serve as a key factor in design decisions. Unlike new construction and reconstruc- tion projects, which are designed with reference to dimensional design criteria presented in the AASHTO Green Book (4), these guidelines do not establish dimensional design criteria for 3R projects. Research has found that any set of dimensional design criteria for 3R projects is likely to produce suboptimal results in terms of crash frequency and severity reduction (7). Rather, 3R design decisions should be based on an assessment of the safety and traffic operational performance of the existing road and the cost-effectiveness of potential design improvements. This approach is most likely to maximize the reduction in crash frequency and severity that can be obtained within any budget level for 3R projects. Geometric improvements should be considered as part of a 3R project in the following situations: • An analysis of the crash history of the existing road identifies one or more crash patterns that are potentially correctable by a specific geometric improvement; C H A P T E R   3

12 Guidelines for Integrating Safety and Cost-Effectiveness into Resurfacing, Restoration, and Rehabilitation (3R) Projects • An analysis of the traffic operational level of service (LOS) indicates that the LOS is currently lower than the highway agency’s target LOS for the facility or will become lower than the target LOS within the service life of the planned pavement resurfacing (typically 7 to 12 years); or • A design improvement would be expected to reduce sufficient crashes over its service life to be cost-effective; that is, the anticipated crash reduction benefits over the service life of the project should exceed the cost of implementing the improvements (including both construc- tion and right-of-way costs). Procedures for evaluating the cost-effectiveness of geometric improvements are presented in Chapter 5. In the absence of any of the three situations defined above, there is no indication that a design improvement is needed as part of a 3R project, and the existing roadway and roadside geometric features should remain in place. It makes little sense to invest scarce resources in design improvements as part of a 3R project when the existing roadway is performing well and potential design improvements would not be cost-effective. The funds needed for such a project can be better invested in projects that do address documented performance concerns or those in which potential design improvements would be cost-effective. In particular, improvement of systemwide safety across the road network is so important that funds invested with the objective of improving safety should be directed toward projects for which it can be demonstrated that substantial safety benefits will likely be obtained. One potential exception to the guidance given above occurs if the normal pavement cross slope for an existing roadway to be resurfaced is less than the dimensional design criteria for normal cross slope in the AASHTO Green Book (4) or the design policy of the responsible highway agency. For, example, the AASHTO Green Book proposes a normal cross slope of 1.5% to 2% for paved roadways. There are no existing cost-effectiveness analysis procedures to address the selection of normal cross slope values, but appropriate pavement cross slope is needed for proper drainage during precipitation and should be provided. Therefore, it is suggested that, whenever practical, normal pavement cross slope be increased to the criteria applicable to new construction and reconstruction as part of pavement resurfacing. This guidance applies only to normal cross slope but does not apply to superelevated horizontal curves. The pavement cross slope on superelevated curves for rural two-lane and multilane highways is addressed by a crash modification factor (CMF) in the AASHTO HSM (2). There- fore, the need for superelevation improvement/restoration for horizontal curves on these types of facilities can be considered through cost-effectiveness analyses. The reliance on cost-effectiveness to guide design decisions for 3R projects has several advantages: • Highway agencies can have confidence that funds invested in design improvements intended to reduce crashes as part of 3R projects are, in fact, likely to result in reduced crashes. • Since crash frequency for a road generally increases with increasing traffic volume, the use of cost-effectiveness analysis as a basis for design decisions means that the likelihood of design improvements being included in a 3R project increases with increasing traffic volume. This dependence of design decisions on traffic volume levels is logical and desirable and is not fully reflected in most current dimensional design criteria for new construction and reconstruction. • A cost-effectiveness approach focuses improvement needs on low-cost improvements with documented safety effectiveness, which are most consistent with the limited scope of 3R projects. However, the procedures are flexible enough that higher-cost improvements can be considered where benefits are sufficient to justify their implementation. If extensive geometric improvements are found to be cost-effective, consideration may be given to reclassifying the project as a reconstruction project.

Process for 3R Project Development 13   Section 4.4 demonstrates that reliance on dimensional design criteria provides suboptimal results, with some investments made at locations where they are not cost-effective and other investments not made at locations where they would be cost-effective. 3.2 How Should Candidate 3R Projects Be Identified? Candidate 3R projects should be identified primarily on the basis of pavement condition and the need for pavement resurfacing. Most highway agencies assess pavement condition and establish resurfacing priorities with the assistance of pavement management systems. Resurfacing at the proper time in the life of the pavement prolongs the life of the pavement and avoids more costly repairs. There are essentially three approaches that may be used in pavement management for project and treatment selection under fiscally constrained situations: ranking, prioritization, and opti- mization. Each of these three analysis approaches provides a method for identifying an optimal strategy for preserving the condition of the network, given any constraints that may exist. One of the easiest methods for selecting projects is to rank needs on the basis of some type of agency priority, such as pavement condition or traffic levels, or both. A common method of ranking needs is to list road sections in sequential order by pavement condition rating and to fund projects with the worst pavement condition until the available funding limits have been reached. This approach is commonly referred to as a “worst-first” strategy, in which the pavements in the worst condition are assigned the highest priority for funding. The next level of sophistication in project and treatment selection is a prioritization process, in which the most cost-effective use of available funding over the analysis period is identified. This approach is preferred over a ranking approach because the consequences of delaying or accelerating a treatment are evaluated and the cost-effectiveness of a treatment is taken into account in developing the program recommendations. An optimization analysis is a more complex means of determining how to efficiently allocate resources so that network conditions are maximized and costs are minimized. In some cases, agencies add additional constraints to help meet agency objectives. For example, an additional constraint that prevents any interstate pavement from dropping below a particular condition level may be included as an analysis parameter. An advantage of the optimization analysis is the ability to incorporate risk into the analysis. One of the challenges associated with the use of pure optimization is the difficulty in linking network-level results with project-level results. At the network level, the results of an optimization analysis typically provide the percentage of the road network that should be moved from one condition state to another through mainte- nance or rehabilitation actions. While most candidate 3R projects are identified on the basis of the need for pavement resur- facing, any project that does not involve new construction or reconstruction can be addressed with the design guidelines in this report. For example, projects on existing roads identified with crash analysis tools, congestion management systems, or any other tool can be designed as a 3R project so long as the scope of the project does not fall within the definition of reconstruc- tion presented in Section 2.1.2. Thus, highway agencies have substantial flexibility to identify candidate 3R projects in any manner, so long as the resulting project falls within the definition of a 3R project. Most 3R projects, even if they are not initiated originally by pavement resurfacing needs, are scheduled for implementation in conjunction with pavement resurfacing to obtain economies of scale. However, a design improvement that falls within the definition of 3R work can be designed and implemented in accordance with the guidelines presented in this report, regardless of whether pavement resurfacing is performed as part of the project.

14 Guidelines for Integrating Safety and Cost-Effectiveness into Resurfacing, Restoration, and Rehabilitation (3R) Projects 3.3 Assessment of Needs for Improvements in Addition to Resurfacing As noted earlier, there are three types of assessments that are generally conducted to investigate whether a design improvement is appropriate in conjunction with pavement resurfacing: • Review and analysis of crash history, • Traffic operational analysis, and • Cost-effectiveness analysis. All three of these assessment approaches should be applied to every candidate 3R project. In some cases, one or more of these assessments can be applied very quickly. For example, if the crash history for a candidate project site includes no crashes, no crash history review is needed; if the candidate project site has experienced very few crashes, the crash history review should require only a limited effort. If a candidate project has a very low annual average daily traffic (AADT) volume, a quick traffic operational review may be all that is needed to establish that the LOS of the project site is acceptable. Conversely, sites with high crash frequencies or high traffic volumes require, and deserve, more involved analyses. Each of these assessment types is discussed below. 3.3.1 Crash History Review A crash history review should be a routine part of the design of every 3R project. Crash history data for a period of at least 3 years—and preferably 5—should be reviewed to identify exist- ing crash patterns and assess the need for design improvements. Depending on the length and complexity of the project and the number of crashes that have occurred during the assessment period, the crash history review may involve • Review of automated crash data or hard copy police crash reports; • Tabulation of crash severities and types overall and by location within the project, including comparison with average crash characteristics for similar corridors; • Preparation of collision diagrams; and • Comparison of the average crash frequencies of intersections and segments within the project limits with those of similar intersections and segments for the same facility type. The scope of the crash history review is flexible and will vary from project to project. If a candidate 3R project has experienced no crashes during the assessment period, no further crash review is needed and the assessment can move on to traffic operational assessment and cost- effectiveness analysis (see below). If only a few crashes have occurred, the crash history review can be accomplished quickly by reviewing automated crash data or hard copy police crash reports. The assessment should focus on determining whether there are repeated crashes of any given type or at any given location that are indicative of a crash pattern and that are potentially correctable by a design improvement. If a substantial number of crashes has occurred in the assessment period, then more effort will need to be devoted to the review, because the potential for crash patterns to be identified is greater. Tabulations of crash severities and types, overall and by location, should be prepared, and collision diagrams may be useful in identifying crash patterns. Table 1 shows an example from a report in the AASHTOWare Safety Analyst software tools (9, 10) illustrating the format of crash tabulations that may be useful in crash history review. Collision diagrams illustrate the crash location and the pre-crash movements and manner of collision of the vehicles involved in each crash. Pre-crash movements are illustrated with arrows showing the intended direction of travel for each vehicle and whether the vehicle was going straight ahead, turning left, turning right, or backing. The manner of collision for multiple- vehicle crashes is illustrated by the orientation at which the pre-crash movement arrows intersect

Process for 3R Project Development 15   and shows the classification of a crash as head-on; sideswipe, opposite direction; sideswipe, same direction; angle; or rear end. Procedures for preparing collision diagrams are presented in HSM Chapter 5, “Diagnosis” (2). The preparation of collision diagrams can be automated with the AASHTOWare Safety Analyst software tools (9, 10) and with other commercially available software. Figure 1 shows a typical collision diagram. Collision diagrams are most commonly used to evaluate intersection crashes, as shown in the figure, but they can be prepared for nonintersection locations as well. In particular, the AASHTOWare Safety Analyst software tools can automate the preparation of collision diagrams for nonintersection locations. The crash history review should focus on fatal and serious injury crashes, which are the highest priority to address in ongoing safety programs. Crash patterns that consist primarily of property- damage-only (PDO) or minor-injury crashes are not unimportant, but sites with patterns of fatal or serious-injury crashes should receive priority in programming design improvements. Chapter 5 of the AASHTO HSM (2) provides a good guide to the process of diagnosing crash patterns at particular sites. HSM Chapter 5 includes guidance on the use of crash tabulations, statistical tests, and collision diagrams in the diagnosis of crash patterns. Use of HSM Chapter 5 enables the investigator to review quantitative results and apply quantitative tests. There is no specific minimum number of crashes that constitutes a crash pattern. Identifica- tion of crash patterns depends on the context of particular sites. Engineering judgment based on highway agency experience with particular types of sites may be the best guide in identifying crash patterns. One approach that is potentially applicable to some types of sites is to apply the testing procedures in HSM Chapter 5 to determine whether crash experience at a given site exceeds the expected crash experience for sites of that particular type and traffic volume level. HSM Chapter 6, “Select Countermeasures,” explains how crash patterns can be used in identifying and selecting particular types of countermeasures with applications likely to reduce particular crash types. The concepts presented in HSM Chapters 5 and 6 have been automated in the AASHTOWare Safety Analyst software, which includes safety management tools for application to specific highway types. Use of Safety Analyst requires initial effort to organize a highway agency’s roadway, intersection, ramp, and crash data, but the software can provide substantial benefits by automating the diagnosis and process of countermeasure selection. Table 1. Example of crash type summary from the AASHTOWare Safety Analyst software (9, 10).

16 Guidelines for Integrating Safety and Cost-Effectiveness into Resurfacing, Restoration, and Rehabilitation (3R) Projects Figure 1. Example of collision diagram from the AASHTOWare Safety Analyst software (9, 10).

Process for 3R Project Development 17   Economic analyses are not typically needed as part of crash history reviews. The identification of what the highway agency considers to be a potentially correctable crash pattern, especially a pattern involving severe crashes, is sufficient justification to include a design improvement in a 3R project. Cost-effectiveness analyses are typically applied where no crash patterns are found (see Section 3.3.3). However, some agencies may prefer to verify through economic analysis that correction of a specific pattern would be cost-effective. Both HSM Chapter 7, “Economic Appraisal,” and Safety Analyst provide procedures for economic appraisal that can be applied for this purpose. 3.3.2 Traffic Operational Analysis If the current traffic operational LOS for a roadway or the anticipated LOS at any time during the service life of the planned pavement resurfacing (typically 7 to 12 years) is less than the highway agency’s target LOS for the roadway, then a design treatment to improve the LOS is appropriate as part of a 3R project. Examples of design treatments that may improve LOS are lane widening, shoulder widening, and addition of turn lanes. It is also appropriate to consider alternatives to design improvements, such as changes in traffic control devices, that might accomplish the same objective. Traffic operational analyses are generally performed with the procedures given in the Highway Capacity Manual 2010 (HCM) (11). For any specific facility type, the HCM specifies the service measure(s) that define LOS (see Table 2), presents thresholds that define specific LOS categories, and presents procedures for estimating the service measure(s) and, therefore, determining LOS. The HCM procedures provide one method of quantifying service measure(s), but other methods, including field measurements and simulation modeling, are also possible. Such alternatives to the HCM procedures can be applied at any site but would normally be needed only for complex situations that fall outside the scope of the HCM procedures. Improvements to intersection design may also be included to improve the LOS as part of a 3R project. Delay is a service measure used to define LOS at intersections. The addition of a left- or right-turn lane may make operational sense as well as provide safety benefits. The target LOS for any roadway is determined by the responsible highway agency. Design policies such as those in the AASHTO Green Book (4) or the policies of individual highway agencies do not normally define target LOS criteria for application to specific projects. As a practical matter, most highway agencies find that appropriate target LOS levels vary with the project context in terms of topography, area type (urban or rural), metropolitan area size, and project scope. Thus, the target LOS is often a project-by-project decision, and highway agencies have substantial flexibility in deciding whether to incorporate traffic operational improvements in particular 3R projects. 3.3.3 Cost-Effectiveness Analysis Cost-effectiveness analysis of 3R projects applies a systemic approach to safety assessment that is based on crash prediction models, as opposed to the analysis of observed crashes for Road Type HCM Service Measure Rural two-lane roadway Average travel speed, percentage of time spent following Rural multilane highway Traffic density Urban or suburban arterial Travel speed as a percentage of base free-flow speed Table 2. Traffic operational service measures for specific roadway facility types (11).

18 Guidelines for Integrating Safety and Cost-Effectiveness into Resurfacing, Restoration, and Rehabilitation (3R) Projects review of crash history discussed in Section 3.3.1. Every roadway section has some nonzero risk of crash occurrence, even if no crashes have actually occurred. A systemic approach considers the need for design improvements on the basis of potential crash risk, whether that risk has resulted in crashes or not. The cost-effectiveness analysis can be based solely on assessment of predicted crash frequency, or predicted and observed crash frequencies can be applied by using the empirical Bayes (EB) procedure to estimate the long-term expected crash frequency. The type of cost-effectiveness analysis proposed for the assessment of 3R projects is benefit– cost analysis, in which the results can be expressed as benefit–cost ratios (project benefits divided by project costs) or as net present benefits (project benefits minus project costs). The benefit–cost analysis procedures applicable to 3R projects are presented in Chapter 5. These procedures are tedious to apply manually, so two spreadsheet-based benefit–cost tools have been created for application by highway agencies; these benefit–cost tools are described in Section 5.6 of these guidelines and can be downloaded from the TRB website (trb.org) by searching for “NCHRP Research Report 876”. Highway agencies can apply these benefit–cost tools in any of three ways: • Evaluation of an individual candidate 3R project to determine whether a specific proposed design improvement would be cost-effective, • Evaluation of an individual candidate 3R project to determine whether any of a range of specific design improvements would be cost-effective, and • Evaluation of a set of potential improvements for generic 3R project scenarios to establish minimum AADT guidelines for consideration of specific design improvements for specific roadway types. The first two of these applications, which involve analysis of individual candidate 3R projects, are the preferred approaches to benefit–cost analysis. Chapter 5 shows why these approaches are preferred. The third application, establishment of minimum AADT guidelines for specific design improvements on specific roadway types, is less desirable than the first two applications but is an acceptable approach for highway agencies that prefer not to conduct analyses of individual projects. Procedures for application of the three approaches listed above are pre- sented in Chapter 5, specifically, • Examples of how benefit–cost analyses can be used, including application to – Rural two-lane highways, – Rural multilane highways, and – Freeways; and • An example of how benefit–cost analysis can be used to derive AADT-based guidelines for specific improvement types.