Click for next page ( 10

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement

Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 9
9 Pavement-type selection committee identifies a pool of alternatives (see Figure 2) Select alternatives from the agency pool Project-specific Conduct engineering review data and analysis Develop project-specific criteria for eliminating unfeasible alternatives Is NO alternative feasible for Eliminate alternative project ? YES Project-specific feasible alternatives Next step: see Figure 4 Figure 3. Identification of pavement alternatives for a project. the pavement. Such needs are better addressed by some of feasible alternatives. Decisions regarding the identifica- alternatives than others, thus helping narrow the list of fea- tion of alternatives at the project level should be documented sible alternatives. in the pavement-selection document. This step is presented in Detailed evaluation of existing pavement properties. Cor- Figure 3. ing, nondestructive testing, and other on-site evaluations (drainage, friction) provide information on the causes, 3.5 Development of Pavement extent, and variability of pavement distress and the struc- Life-Cycle Strategies tural and functional capacities of the existing pavement. Roadway peripheral features. Peripheral features such as For each alternative identified and designed, a life-cycle guardrails, curbs and gutters, traffic control devices, over- strategy is developed to sustain the desired functional and head clearances, and on-grade structures may play impor- structural performance level of the pavement over the chosen tant roles in the selection of alternatives. Such features may analysis period. The life-cycle strategy includes the construc- have special bearing on rehabilitation work where grade tion of the original pavement structure, as well as rehabilita- changes are limited. For example, in some cases, recycling tion, preventive maintenance, and corrective maintenance or reconstruction may be more desirable than an overlay. activities. Figure 4 presents a flow chart of this step. Each alternative pavement should be designed for the same In practice, the broadest range of alternatives (including the conditions of traffic level, reliability, life, and terminal per- various forms of recycling) should be considered on each proj- formance thresholds that trigger rehabilitation. However, the ect. However, certain alternatives may not be appropriate for required M&R activities, their timing, and associated costs will certain classes of roads or under certain traffic conditions. In be different for design alternatives over the life of the pave- addition, there may be project features that limit the number ment. In the long term, these differences result in different cost

OCR for page 9
10 Pavement life-cycle models for alternatives Figure 4. Development of a life-cycle model for pavement alternatives. streams for alternatives developed with similar design and terms of pavement-structure design, costs, performance, traf- performance criteria. fic characteristics, climate, materials, or subgrade). Historical Establishing the life-cycle model requires knowledge of the project data may consist of direct source data or could be man- expected service lives of the original or structurally rehabili- ifest as practical-experience information held by the agency, its tated structure and the sequence of expected timing and extent engineering consultants, paving contractors, or various other of future M&R treatments, as illustrated in Figure 5. Viable project stakeholders. Data should be updated and expanded as strategies are selected for structural rehabilitation based on necessary to accommodate the continuous evolution of struc- the predicted terminal pavement condition in the future. A tural design, materials, specifications, and construction prac- viable strategy should be feasible from an engineering and eco- tices. As noted earlier, the suggested approach is to use data nomic standpoint, address existing causes of deterioration, and housed in an agency's pavement management system database. mitigate the recurrence of the same distress over the analysis If sufficient data are unavailable to incorporate new technolo- period. gies, the past data can be considered along with research find- The selection of life-cycle strategies requires the collection ings, specific evaluation and testing results, experience in other and meticulous evaluation of a large amount of data. Such data states, and modeling tools such as the MechanisticEmpirical pertain to the current project and similar past projects (e.g., in Pavement Design Guide (MEPDG) (AASHTO 2008). Initial Construction Rehabilitation 1 Rehabilitation 2 Maintenance Maintenance Maintenance Maintenance Time, years 0 t1 t2 t3 t4 Figure 5. Example pavement life-cycle model.

OCR for page 9
11 3.5.1 Service Lives of Initial Pavement and The commonly used approaches in pavement Future Rehabilitation Treatments managementbased estimations are performance trends and survival analyses. The service life of the initial pavement structure before the first major rehabilitation and the service life of future rehabil- itation treatments are estimated for each proposed alternative. Performance-Trend Analysis The service life until rehabilitation depends on factors such as In performance-trend analysis, historical pavement con- material durability, climate condition, construction quality, dition data representing a given strategy are compiled and and traffic loading and may be shorter or longer than the regressed over time. A threshold pavement condition level is design life of the pavement (Darter and Hall 1990). then established that indicates the need to restore deteriorat- Pavement service life can be estimated using various tech- ing functional and/or structural capacity. On the regression niques, ranging from expert modeling using the opinions of curve, the time at which the pavement reaches the threshold experienced engineers to detailed performance-prediction condition level is determined. The estimated time reflects the modeling using historical pavement performance data to con- average age for a rehabilitation need or the estimated service struct survival curves. The following guidelines are suggested life of the pavement. when considering a technique for developing service-life Performance-trend analysis is essentially a four-step process: estimates: Historical pavement performance data documented in the 1. Existing pavement sections with similar structural fea- agency's pavement management system should be the first tures, traffic loadings, subgrade type, and functional classes and foremost source. to those of the proposed alternatives are identified. The Expert opinions are prone to biases and should be consid- historical condition data of these sections are extracted ered only when reliable historical performance data are from the pavement management system or other sources. unavailable or are greatly limited, such as if the pavement Careful attention is given to the "representative" aspect of or rehabilitation types being considered are substantially the selected pavement sections, and the data are screened different, due to changes in traffic or use of new materials for any factors that might have an unusual effect on pave- or technologies. ment condition (e.g., design or construction issues and Experience-based estimates, in conjunction with data trends traffic loading). Sections that are not representative may from other sources, can be used in the absence of reliable warrant removal from the dataset. and adequate historical performance data. 2. Time-series performance plots are created using the con- dition data for each family of pavements. Best-fit linear or The following items should be taken into consideration nonlinear models relating pavement condition to age are when developing service-life estimates: developed. To the extent possible, the time-series plots include separate trends for structural and functional indi- The reliability and accuracy of estimates greatly depend on cators. Some data filtering usually is needed in eliminating the data quality and the number of data points available. the time-series data that reflect any significant improve- It is important to assess how closely the pavement manage- ment in pavement condition due to M&R activities. ment sections used in the analysis represent the pavement 3. Acceptable condition thresholds that serve as triggers for alternative under consideration. This requires grouping of any major intervention should be identified. The threshold pavement sections with similar characteristics as a "family" trigger values depend on factors such as the roadway type of pavements. A factorial table can be used to construct such (rural or urban), functional class, size of highway network, pavement families. The grouping factors include, but are not agency's resources, policies, and programmatic constraints. limited to, similarities in structural design, traffic loading, 4. Service-life estimates are made for each family of pavements functional class, climate, geographic region, pavement type, based on their condition trends and threshold values. These design features, design type, subgrade, and materials. For estimates can be developed using both deterministic and simplicity and clarity, the number of factors should be min- probabilistic analyses. imized as much as possible while remaining comprehensive enough to cover the entire range of distinct characteristics. Figure 6 illustrates the estimation of functional service life If the design, materials, construction specifications, funding, for a family of pavements using smoothness data. In this exam- or operating conditions of the pavement alternative are dif- ple, a threshold International Roughness Index (IRI) value of ferent from those available from pavement management 150 in./mi was used. The model trend line intersects the thresh- files, the pavement-type selection committee should discuss old IRI where the estimated service life is about 18 years. There- appropriate adjustments. fore, in this example, the estimated mean for functional life

OCR for page 9
12 300 100% IRI = 51.13e0.0521AGE Est. std dev. R = 0.52 1.5 years 250 N=242 80% Roughness (IRI), in./mi 200 Sections Surviving Threshold IRI = 150 in./mi 60% 50% sections surviving 150 40% 100 Est. median life Est. Service life 17.5 years 20.5 years 50 Est. std. dev. 20% 2.5 years 0 0% 0 5 10 15 20 25 30 0 5 10 15 20 25 Age, years Age, years Figure 6. An example of functional life estimation for Figure 7. An example of pavement-family survival a family of pavements (N number of observations). analysis. (ride quality) is about 18 years, and the estimated standard in too few roadway segments for a survival analysis to be repre- deviation is about 2.5 years. sentative of the agency's operational policies and procedures. Another drawback to this approach is the inability to account for the benefits or improvements in pavement design, materi- Survival Analysis als, and specifications. This drawback can be overcome by using The "survival" of a pavement section is defined as the non- survival analysis in conjunction with mechanistic-empirical occurrence of failure--or, in other words, the nonoccurrence methods to estimate pavement lives. Such a distress-dependent of major rehabilitation. This technique uses historical con- approach can be used to determine the effect of using materials struction and rehabilitation data for a family of pavements to and design features that are not represented adequately in the construct a survival curve that plots the percent survival as a agency's pavement management system database. function of time (or traffic loadings). Using a value of 50 per- cent pavement sections surviving (or, conversely, 50 percent 3.5.2 Timing and Extent sections failed), an estimate of the median life (and standard of M&R Treatments deviation) for a pavement with similar features and loading conditions can be developed. Figure 7 shows an example sur- The timing and extent of future rehabilitation treatments vival curve developed for a family of pavements. The estimated for functional improvement (e.g., thin overlays and diamond median service life and the standard deviation are 17.5 and grinding), as well as future maintenance treatments (e.g., rou- 1.5 years, respectively. tine maintenance, preventive maintenance, and minor repairs), The preferred method of survival analysis is to develop pave- should be estimated based on pavement management system ment survival curves within each district or region and to con- and/or history records. The focus of maintenance costs should firm those survival relationships periodically over time. Trying be on the timing and extent of preventive and major forms of to segregate the data into too many groups, however, can result maintenance.