Use of Pavement Management System Data to Enhance Pavement Performance Specifications in Canada James Dale Smith, Stephen Lee, and Thomas J. Kazmierowski • Limited ability for the agency to account for the changes in performance that arise from differences in quality between the as-designed product and the as-built product. Most agency materials and construction specifications for pavement rehabilitation and construction activities provide little or no linkage between quality assurance methods and the in-service performance of the treatment. The Ministry of Transportation of Ontario (MTO), Canada, is developing and implementing performance specifications to bridge this gap. MTO defines performance specifications as specifications that describe how the finished product should perform over time. This paper describes how MTO used historical pavement management system data to assess performance sensitivity and to create historical performance distributions and how the agency used these distributions to set acceptance criteria, including payment adjustment for performance. In recent years, agencies have explored opportunities to incorporate performance specifications that specify quality, as it relates to desired long-term performance, into their construction contracts (1). These specifications also provide a means to account for the value lost or gained by variations in performance from the specified criteria. Previous Studies Increasing portions of federal, provincial, and municipal resources are being allocated to preserve and maintain Canada’s roadways. Considering the major investments made by these transportation agencies to construct and maintain the roadway infrastructure, it is critical to ensure not only the initial quality but also the long-term performance of these expenditures. Quality assurance specifications that specify end-product quality often have been used by Canadian transportation agencies to ensure the initial construction quality of roadway pavements. Most materials and construction specifications for roadway construction and maintenance offer minimal connection between quality assurance methods and the in-service short- and long-term performance of the delivered product. This quality assurance method has limitations for both the transportation agency and the contracted road builder, including The development of performance specifications and performancerelated specifications (PRS) for flexible pavements is well documented. Initial efforts to develop PRS for new hot-mix asphalt (HMA) pavements began under NCHRP Project 10-26 (2). NCHRP Project 09-20 developed PRS for HMA pavement on the basis of field data (3). A revised PRS method for new HMA pavement that incorporated AASHTO mechanistic–empirical models for predicting HMA pavement performance was developed in NCHRP Project 09-22 (4). The final report of SHRP 2 Project R07, which developed and implemented performance specifications for rapid highway renewal, generally supports the use of performance specifications as a viable option for agencies interested in empowering the private sector to provide creative rapid renewal solutions to save time, minimize disruption, and enhance durability (5). A comprehensive overview and guidelines for pavement warranty programs are presented in NCHRP Report 699 (6). • Increased contract administration burden on the agency; • Minimized flexibility for the contractor to initiate, innovate, and introduce new products; • Inappropriate risks and responsibilities allocated to the agency when the contractor is best able to manage these issues; • Focus of the contractor’s efforts on initial acceptance rather than on long-term performance; and Development of Performance Warranty Requirements Initial Trial Contracts Since 2007, the Ministry of Transportation of Ontario (MTO), Canada, has tendered more than 15 contracts with a 7-year pavement warranty specification. These 7-year pavement warranty projects are unique in North America. Contractors can bid on the pavement portion of the project without any mandatory conventional specifications. The contractor essentially is responsible for all aspects of pavement design, materials, and construction. MTO does not approve the design of the pavement structure; instead, the agency relies on performance requirements during the warranty period. This practice J. D. Smith, Ministry of Transportation of Ontario, 301 Saint Paul Street, Saint Catharines, Ontario L2R 7R4, Canada. S. Lee and T. J. Kazmierowski, Ministry of Transportation of Ontario, 145 Sir William Hearst Avenue, Room 233, Downsview, Ontario M3M 1J8, Canada. Current affiliation for T. J. Kazmierowski: Golder Associates, 6925 Century Avenue, Suite 100, Mississauga, Ontario L5N 7K2, Canada. Corresponding author: J. D. Smith, J.Dale.Smith@ontario.ca. Transportation Research Record: Journal of the Transportation Research Board, No. 2573, Transportation Research Board, Washington, D.C., 2016, pp. 60–68. DOI: 10.3141/2573-08 60 Smith, Lee, and Kazmierowski 61 motivates contractors to provide an adequate pavement design, to have effective quality management systems, and to construct a durable pavement. The earliest of these contracts were completed in 2007. Fulldepth reclamation is the pavement remedial treatment used for most contracts, with cold in-place recycling used for others. Contractors responded to the challenge, although some would rather not have dealt with warranties. In most cases, the contractor pavement designs were similar to, or more substantial than, what MTO would have designed. Contractors used established materials and construction methods and put significant effort into quality control. With the exception of two cases, the pavements, still within their warranty term, generally are meeting the performance requirements. A few pavement sections have exceeded the roughness and cracking requirements and required corrective work, including surface course replacement and crack sealing. The analysis of historical performance data for the pavement management system described in this paper has helped refine and further develop the specification used for the initial trial contracts. It is expected that the analysis, when shared with Ontario’s road building industry, will increase bidder understanding of performance risks and will improve the outcome of future projects that use pavement performance specifications. of the use of these records in developing pavement performance specifications: Performance Records Performance Sensitivity Evaluation MTO has collected pavement performance records systematically for provincial highway pavement sections since the mid-1960s (7). Since the early 1980s, subjective performance evaluations have been recorded by trained evaluators using standard procedures (8). Since the mid-1990s, objective records of pavement roughness, wheel track rutting, and pavement crossfall have been collected with the MTO Automatic Road Analyzer (ARAN) with a 10-m reporting interval. An analysis of the performance records was conducted to establish reasonable and defensible expectations of pavement performance. Subjective evaluation records for the period from 2005 to 2011 (10,322 records) and objective ARAN records for the period from 2007 to 2010 (5,052 records) were used for the analysis. The time periods selected are not arbitrary but contain recent records that are in a consistent format and are suitable for statistical analysis. The subjective records were compiled by trained evaluators and the objective records were collected with the 36 ultrasonic sensors (wheel track rutting and crossfall) and inertia profilograph (roughness) mounted on the ARAN vehicle. Statistically, the subjective records contain categorical paired data and the objective records contain continuous data. An example of continuous data is wheel track rutting depths measured and collected along the entire length of a pavement section in both wheelpaths, with averages reported at 10-m intervals. An example of the categorical paired data is moderate, frequent flushing. To facilitate the analysis, it was necessary to transform the paired subjective records to a single parameter. This transformation was achieved by applying equal weight to the distress severity and density, with the resulting single parameter called the distress index (DI). The DI follows the MTO convention of reporting pavement condition indexes: a value of 10 means perfect condition, with progressively lower condition as the value decreases. The performance records were collected for the purpose of networklevel pavement management. The following are some limitations As a first step in the analysis, the researchers evaluated the performance sensitivity to such key highway attributes as geographic region, traffic volumes, road classification, and work type. Performance sensitivities were addressed by one of three methods: • Categorical, discrete pavement distress records that describe the predominant distress only; • Limited detail on construction and any subsequent maintenance activity; and • Performance evaluation limits that do not necessarily coincide with pavement rehabilitation limits. Recognizing the data limitations, the researchers completed detailed field evaluations of selected pavement sections to validate the analysis results that are described later in this paper. MTO has established a warranty evaluation segment length (or lot size) of 500 lane meters for pavement warranties, with a shorter length used where necessary, such as at the limits of a project. Where applicable, warranty requirements are specified relative to this length (e.g., average rut depth per 500 lane meters). The 500–lane meter length was selected to balance two factors: the contractor’s increasing risk of nonconformance with shorter lengths and the owner’s increasing risk of unrepaired, localized areas of poor performance with longer lengths. For analysis, the performance records were transformed to values for a 500–lane meter length where possible. • Varying performance requirements by attribute. Higher road classifications exhibit lower roughness, as measured with the mean roughness index (MRI), as is expected with the use of higher reliability and lower terminal serviceability as specified in the AASHTO pavement design procedures. • Excluding attribute by specification warrant. Highways classified as local typically exhibit markedly lower performance compared with collector, arterial, and freeway classifications. • Applying the worst-case performance. The performance requirement was established from wheel track rutting depths on Superpave® Traffic Categories D and E pavements (9), on the basis of the approximately 2-mm average increase in observed wheel track rutting depths from Category A to Category E pavements—142 contracts and 2,807 segments of 500 lane meters—shown in Figure 1. Note that MTO uses five traffic categories: Category A: <0.3 million equivalent single axle loads (ESALs), Category B: 0.3 to <3 million ESALs, Category C: 3 to <10 million ESALs, Category D: 10 to <30 million ESALs, and Category E: >30 million ESALs. Figure 2 shows the sensitivity of pavement edge cracking to highway classification, and the evidence of significantly higher amounts of edge cracking for local highways. The reason for this sensitivity is simple: local highways in Ontario typically are very low volume with less HMA thickness and do not warrant partially paved shoulders. It is necessary to evaluate the sensitivity of performance to key highway attributes before further analysis, 62 Transportation Research Record 2573 100 90 Line Distribution (continuous data) (%) 80 70 60 50 40 30 20 10 0 0 2 Traffic Category A 4 6 8 Average Wheel Track Rut Depth (mm) Traffic Category B Traffic Category C 10 Traffic Category D 12 Traffic Category E FIGURE 1 Sensitivity of wheel track rutting to Superpave traffic category after 7 years of service on the basis of 147 contracts and 2,807 segments (500 lane meters per segment). 10 9.5 9 Distress Index 8.5 8 7.5 7 6.5 6 0 2 All highways 4 Freeways 6 8 10 12 Pavement Age (years) Arterial highways FIGURE 2 Sensitivity of edge cracking to highway classification. Collector highways 14 16 Local highways 18 20 Trend (all records) Smith, Lee, and Kazmierowski 63 because application of one of the three methods listed above may require selective extraction of the historical data along with the removal of outliers. described earlier, the performance is sensitive to highway classification, and distributions are shown for freeways and arterial and collector highways. Performance Distribution Performance Acceptance Limits In the second step, the performance records for each distress type for the performance year of interest were extracted and best-fit statistical curves were plotted through the performance distribution. The performance year of interest was 7 years, which was selected by MTO as the performance warranty period for new, reconstructed, and rehabilitated flexible pavement structures and which, on the basis of experience in Ontario, constitutes approximately 40% of the typical 18-year design life of these structures. Figure 3 shows the distribution of flushing performance for 445 pavement sections that were 7 years old at the time of evaluation. This distribution indicates that approximately 65% of the pavements were evaluated as having no flushing distress. A power function best-fit curve was plotted through the lower portion of the distribution with probability density function y = 6 E−08e1.5485(DI). Figure 4 shows the distribution for roughness (MRI) on the basis of 68 contracts and 1,473 segments (500 lane meters per segment). The distribution reflects the freeze–thaw climate and the ongoing challenges of the low-temperature performance of asphalt cement for road construction in Ontario, with segments of relatively high roughness at the lower end of the distribution. For roughness, as The third step in the process is to use the performance distribution to statistically set the acceptance limits. NCHRP Report 699: Guidelines for the Use of Pavement Warranties on Highway Construction Projects states, “The DOT may set the threshold at 2 σ. . . . With additional experience or improved consistency, the DOT may decide to tighten the threshold (to between 1 σ and 2 σ) or extend the warranty” (6). It is critical to use reasonable performance thresholds that balance road builder risk (and project cost) and agency expectations. Historically, MTO has balanced the risk and agency expectations by establishing acceptance limits based on 1.5 σ or 2.0 σ from the mean of a normalized distribution. For pavement performance specifications, MTO has selected a threshold of 1.5 σ for most distress types, and a threshold of 2 σ for distress types that are objectively measured and combined with payment adjustment based on performance. Payment adjustment for performance is described in the next section of this paper. Some performance distributions were too skewed to permit normalization, for example, the performance distribution for flushing shown in Figure 3. Where normalization was not possible (on 100 Line Distribution (categorical, paired data) (%) 90 80 70 60 y = 6E–08e1.5485(DI) R 2 = .9325 50 40 Slight–few (0% to 10% of surface area affected) 30 20 Moderate–frequent (20% to 50% of surface area affected) 10 0 10 9 8 7 Data points 6 5 Distress Index 4 3 Best fit, probability density function FIGURE 3 Flushing performance distribution after 7 years of service. 2 1 0 64 Transportation Research Record 2573 100 90 Line Distribution (continuous data) (%) 80 70 Resurfacing work excluded Freeway = 446 segments Arterial = 694 segments Collector = 333 segments 60 50 40 30 20 10 0 0 0.5 1 1.5 2 2.5 3 3.5 MRI (m/km) Freeways Arterial highways Collector highways FIGURE 4 Pavement roughness (MRI) performance after 7 years of service. Resurfacing work is excluded. the basis of generally accepted limits for skew and kurtosis), the threshold was based on the best-fit probability distribution or density function, with a probability of exceedance of 6.7%, equivalent to 1.5 σ. The DI value for flushing at a 6.7% probability of exceedance was 8.9. Backcalculation yielded 15.7% moderate flushing. To illustrate, Figure 5 shows the average wheel track rut depths on 1,125 pavement segments (500 lane meters per segment) from 64 Superpave Traffic Categories D and E contracts as measured by ARAN. The distribution of the measured data was normalized by a Box–Cox power transformation (10). Transformation results are shown in Figure 6. Backtransformation yielded values for mean + 1.5 σ and mean + 2 σ of 6.7 mm and 7.8 mm, respectively, as shown in Figure 5. Performance Categories and Requirements For flexible pavement structures, performance requirements have been established for the following eight categories: flushing, coarse aggregate loss, wheel track rutting, roughness (MRI), alligator cracking (or crack density), single and multiple cracking, joint separation, and differential frost heaving. For each category, some requirements apply throughout the warranty period, including the final 12 months, and some requirements only apply in the final 12 months of the warranty period. The requirements that apply throughout the warranty period are called limits and are intended to address conditions that would affect pavement safety, serviceability, or both. By nature, these requirements have a low probability of exceedance. The limit requirements are based on the performance distributions, on expert judgement, and on an assessment of the existing pavement network; for example, Table 1 shows the occurrence of different depths and interval lengths of localized wheel track rutting on 7-year-old pavements in the provincial highway network. The 30-m interval and 14-mm average rut depth were selected as the limit; this limit was exceeded on 3.5% of 7-year-old pavements or contracts. The requirements that apply in the final 12 months of the warranty period are called levels and are the expected reasonable level of pavement performance as determined by the statistical analysis of historical performance described in the preceding sections. The intent of this two-part warranty approach is (a) to reduce the owner’s oversight burden by calling for a detailed evaluation of performance to be carried out only at the end of the warranty period and (b) to reduce the risk to the road builder of multiple repair-related returns to the project site. Before the final year of the warranty, when trained evaluators conduct a detailed evaluation, including a project-level ARAN survey, routine assessments of pavement condition, maintenance patrols, and a network-level ARAN survey are relied on to identify limit exceedances. Warranty performance requirements for this approach are shown in Table 2, along with the repair that must be completed if requirements are not met. Repairs are completed according to standard material and construction specifications with normal owner oversight, including material sampling and testing. The performance requirements in the final 12 months of the warranty period are based on a +1.5 σ threshold, except for those categories with a payment adjustment for performance, which are based on a +2 σ threshold. Payment adjustment for performance is applied only to categories measured objectively by ARAN: cracking, wheel track rutting, and roughness. These three categories are the key Smith, Lee, and Kazmierowski 65 300 250 Count 200 150 Skew = 1.501 Kurtosis = 3.014 100 +1.5 σ = 6.74 mm +2 σ = 7.84 mm 50 10 9 .8 11 8 .3 11 6 .8 12 4 .3 12 3 .8 13 1 .2 13 9 .7 8 14 .2 6 91 10 .3 43 9. 94 9. 46 8. 98 8. 49 7. 01 7. 53 7. 04 6. 56 6. 08 5. 59 5. 11 4. 63 4. 14 3. 66 3. 18 2. 69 2. 21 1. 73 1. 0. 0. 24 0 Average Segment Wheel Track Rut Depth (mm) FIGURE 5 Average depth of wheel track ruts after 7 years of service. 140 120 100 λ = –0.71601 Skew = –3.4 E–08 Kurtosis = –0.378 Count 80 60 40 20 0. 7 0. 7 0. 8 0. 8 0. 9 0. 9 0. 9 1. 0 1. 0 1. 1 1. 1 1. 1 1. 2 1. 2 1. 3 1. 3 1. 3 1. 4 1. 4 1. 5 1. 5 1. 6 1. 6 1. 6 1. 7 1. 7 1. 8 1. 8 1. 8 1. 9 0 Average Segment Wheel Track Rut Depth (normalized) FIGURE 6 Normalized distribution of wheel track rut depth (l 5 Box–Cox transformation parameter). 66 Transportation Research Record 2573 TABLE 1 Occurrence of Localized Wheel Track Rutting on 7-Year-Old Pavement sufficient to shift the road builder’s performance target from one that fulfills the minimum warranty requirements to one that will result in a payment increase. Because of the analysis’s limitations described earlier, detailed field evaluations were carried out to verify the results of the statistical analysis. Figure 7 shows the results of detailed field evaluations of coarse aggregate loss on 729 segments (500 lane meters per segment) of 7-year-old pavement from 36 contracts. These evaluations showed that 65 of the 729 segments (8.9%) exceeded the level requirement of no more than 10% loss of moderate coarse aggregate, and 25 of the 729 segments (3.4%) exceeded the limit requirement that there be no severe or very severe coarse aggregate loss. Of the 65 segments that exceeded the level requirement, 36 were part of a single problematic contract, and most of the locations of severe or very severe coarse aggregate loss were localized potholes requiring minor repairs. The detailed field evaluations generally verified the expected statistical nonconformance rates of 6.7% (1.5 σ) for levels and 2.3% (2 σ) for limits, for all distresses. The performance analysis was based on subjective evaluations by experienced, trained evaluators and on ARAN measurements of network-level roughness and wheel track rutting. In 2012, MTO purchased a new ARAN to replace the older ARAN that provided the analysis data. The new ARAN is equipped with an advanced laser measurement system whose capabilities include crack measurement. Data collected by the new ARAN are analyzed as they become available each year for comparison with the analysis results; reasonable agreement has been found to date. Future refinements to the Table 2 performance requirements are anticipated after sufficient data are obtained from the new ARAN. Occurrence (%), by Average Rut Depth over Interval Interval (m) 16 mm 14 mm 7.7 4.2 2.4 0.7 11.3 4.9 4.2 3.5 10 20 25 30 performance categories of flexible pavement. Measurements taken by ARAN have been found to be repeatable and accurate, and maintenance and calibration of the equipment is annual and ongoing. In addition, a protocol is being developed for field calibration immediately before the measurement of pavement warranty segments. Warranty performance evaluations are completed by the owner or a representative of the owner. The contractor may challenge the evaluation results and request remeasurement. Payment adjustment calculations apply to each warranty evaluation segment, which typically is 500 lane meters. A payment increase applies to performance that exceeds the −1 σ threshold and a payment decrease applies to performance between the +1 σ and +2 σ thresholds. Performance measured beyond +2 σ requires repair. Payment adjustments are intended to be quality incentives that are TABLE 2 Performance Requirements for Warranty Contracts Entire Warranty Period (Limits) Final 12 Months of Warranty Period (Levels) Measurement Performance Requirement Repair Performance Requirement Repair Coarse aggregate loss SP-024 evaluation No severe or very severe requirements Flushing SP-024 evaluation No severe or very severe requirements Moderate coarse aggregate loss shall be <10% of segment length Moderate flushing shall be <10% of segment length Alligator cracking SP-024 evaluation Remove HMA surface course uniform depth and replace Remove HMA surface course uniform depth and replace Reconstruct Cracking ARAN No moderate, severe, or very severe requirements na Remove HMA surface course uniform depth and replace Remove HMA surface course uniform depth and replace Reconstruct Wheel track rutting ARAN Roughness ARAN ASTM E 950-09 profilometer Joint separation Direct measure na Differential frost heaving ASTM E 950-09 profilometer No lane segments ≥ 30 m with MRI > 6.0 m/km Performance Category Note: na = not applicable. No wheelpath locations ≥ 30 m with average rut depth > 14 mm No lane segments ≥ 30 m with MRI > 2.6 m/km Reconstruct Remove HMA variable depth and replace Slight alligator cracking shall be <10% of segment length Length of cracking up to 20 mm in width shall be <160% of segment length No cracks >20 mm Average depth of wheel track rutting per segment <8.0 mm Average MRI per segment <1.6 m/km (freeway) <1.75 m/km (arterial) <1.90 m/km (collector) Separation shall be <5 mm Separation shall be <20 mm Reconstruct na Remove and replace HMA full depth Full-depth crack repair Remove HMA variable depth and replace Remove HMA variable depth and replace Seal joints 500-mm-wide strip repair Smith, Lee, and Kazmierowski 67 100 90 Pavement Area Affected (%) 80 70 60 Nonconformances Level: 65/729 (8.9%)* Limit: 25/729 (3.4%) 50 *36 on one contract 40 30 20 10 0 0 51 101 151 201 251 301 Moderate 351 401 Segment 451 501 551 601 651 701 Severe to very severe FIGURE 7 Detailed field evaluation of coarse aggregate loss from 7-year-old pavement. Warranty Limitations Summary and Future Work The performance requirements were established with data from pavements across Ontario and reflect typical in-service conditions. The requirements are not valid for extreme or unexpected in-service conditions and events that cannot be accounted for by pavement design. To minimize contractual dispute, conditions are specified that will void the pavement warranty, in whole or in part. Limitations have been established for unexpected traffic increases, temperature extremes, and damage to or alteration of the work resulting from vehicular accidents, fuel spills, natural disasters, nonroutine maintenance, construction, and similar events, as shown in Table 3. This paper details how compiled data from MTO’s pavement management system were used to develop distributions for several categories of flexible pavement performance. Through the use of these distributions, practical and measurable performance acceptance limits are being established to meet the performance specification objective of describing how the pavement should perform over time. Detailed field evaluations were used to validate the acceptance limits, and analysis of data from the new, more advanced ARAN acquired by MTO in 2012 is expected to lead to future refinements to the acceptance limits. TABLE 3 Warranty Limitations (11) Condition Limitation Impact Traffic increase Cumulative actual ESAL during the warranty period exceeds projected warranty period ESAL by more than 40% Annual lowest daily temperature is less than the mean + 2 σ value from LTPPBind v3.1 (11) Total degree days above 10°C in a year is greater than the mean + 2 standard deviation value from LTPPBind v3.1 (11) Pavement damaged by vehicular accidents, fires, hazardous materials, chemical and fuel spills, military actions, civil commotions, natural disasters, or other force majeure incidents Pavement altered by the failure or deterioration of underlying roadway elements such as drainage structures or by construction or nonroutine maintenance carried out by others Entire warranty voided Low temperature High temperature Third-party damage Alteration Performance requirement for cracking voided Performance requirements for flushing and wheel track rutting voided Entire warranty voided at location of damage Entire warranty voided at location of alteration 68 Initial trials that used a 7-year pavement warranty have shown that pavement performance specifications can be incorporated into innovative alternate delivery contracts. The analysis described in this paper has been incorporated into an updated 7-year pavement warranty specification, along with guidelines for associated project selection, design, estimation, contract oversight, and performance evaluation. MTO is evaluating automated performance data collection for surface distresses to improve the safety of data collection and the objectivity of the performance evaluations. Cost and performance information on all 7-year pavement warranty specification projects will be compiled and evaluated. When repairs are required during the warranty period, the contractor’s response will be monitored to assess the strengths and weaknesses of the specification. This information will be used to guide future updates of the specification. MTO will continue to pursue the use of alternate models of contract delivery to expand the available options for program delivery in the face of increasing resource constraints, to allocate appropriate risk to those who can best manage these responsibilities, and to promote partnerships and collaboration between stakeholders. In recent years, MTO has awarded design–build contracts in addition to traditional design–bid–build contracts. In the next 5 years, MTO plans to increase the use of design–build projects. Each MTO region will develop a 5-year plan both for rehabilitation and for expansion projects. Performance specifications will be integral to these ambitious plans. Transportation Research Record 2573 References 2. Anderson, D. A., D. R. Luhr, and C. E. Antle. NCHRP Report 332: Framework for Development of Performance-Related Specifications for Hot-Mix Asphaltic Concrete. TRB, National Research Council, Washington, D.C., 1990. 3. Epps, J. A., A. Hand, S. Seeds, T. Scholz, S. Alavi, C. Ashmore, C. L. Monismith, J. A. Deacon, J. T. Harvey, and R. Leahy. NCHRP Report 455: Recommended Performance-Related Specifications for Hot-Mix Asphalt Construction: Results of the WesTrack Project. TRB, National Research Council, Washington, D.C., 2002. 4. El-Basyouny, M., and M. G. Jeong. Probabilistic Performance-Related Specifications Methodology Based on Mechanistic–Empirical Pavement Design Guide. In Transportation Research Record: Journal of the Transportation Research Board, No. 2151, Transportation Research Board of the National Academies, Washington, D.C., 2010, pp. 93–102. 5. Scott, S., III, L. Konrath, and T. Ferragut. SHRP 2 Report S2-R07-RR-1: Performance Specifications for Rapid Highway Renewal—Final Report. Transportation Research Board of the National Academies, Washington, D.C., 2013. 6.Scott, S., III, T. Ferragut, M. Syrnick, and S. Anderson. NCHRP Report 699: Guidelines for the Use of Pavement Warranties on Highway Construction Projects. Transportation Research Board of the National Academies, Washington, D.C., 2011. 7.Pavement Design and Rehabilitation Manual, 2nd ed. Ministry of Transportation, Ontario, Canada, 2013, pp. 362–379. 8. Chong, G. J., W. A. Phang, and G. A. Wrong. Manual for Condition Rating of Flexible Pavements: Distress Manifestations. In Publication SP-024, Ministry of Transportation, Ontario, Canada, 1990. 9.Standard Practice for Superpave Volumetric Design for Asphalt Mixtures. AASHTO R 35-15. AASHTO, Washington, D.C., 2015. 10. Box, G. E. P., and D. R. Cox. An Analysis of Transformations. Journal of the Royal Statistical Society: Series B (Methodological), Vol. 26, No. 2, 1964, pp. 211–252. 11. LTPPBind: A New Tool for Selecting Cost-Effective Superpave Asphalt Binder Performance Grades. Report FHWA-RD-99-082. FHWA, U.S. Department of Transportation, 1999. 1.Transportation Research Circular E-C137: Glossary of Highway Quality Assurance Terms, Fourth Update. Transportation Research Board of the National Academies, Washington, D.C., 2009. The Standing Committee on Flexible Pavement Construction and Rehabilitation peer-reviewed this paper.