A study was performed to determine the influence of material properties on the thermal cracking performance of hot mix asphalt (HMA), and to determine the ability to predict thermal cracking from pavements of known field performance. The testing device used to measure the HMA properties was the thermal-stress, restrained-specimen test (TSRST), and the device used to measure the binder properties was the bending beam rheometer (BBR). The laboratory study was conducted to determine the variability of test results as an influence of 1) asphalt cement stiffness, 2) asphalt cement quantity, 3) mixes with various aggregate qualities, 4) aging, and 5) the presence of hydrated lime. The influence of the asphalt cement stiffness was the single largest factor that controlled the test results.

TRB's National Cooperative Highway Research Program (NCHRP) Report 669: Models for Predicting Reflection Cracking of Hot-Mix Asphalt Overlays explores mechanistic-based models for predicting the extent and severity of reflection cracking in hot-mix asphalt overlays. Appendices A through T to NCHRP Report 669 are available online--

Asphalt mixes are designed to provide adequate resistance to various distresses including cracking, rutting, and moisture damage. Recently, more efforts are directed towards including performance assessment tests during the design and production of asphalt mixes. Performance-Engineered Mix Design (PEMD) or Balanced Mix Design (BMD) is a new and innovative design approach that incorporates performance assessment tests to optimize the design of asphalt mixes to provide adequate performance. Although transportation agencies are motivated to implement the PEMD approach, several research knowledge gaps and concerns need to be addressed before PEMD successful implementation. This research study aims to advance, develop, and implement performance-engineered design approach and specifications to extend the service life of asphalt pavements.The first phase of this research developed and evaluated a new and innovative monotonic cracking performance indicator called Weibull Cracking Resistance Index (WeibullCRI). The proposed indicator describes the entire load-displacement curve, which overcomes the limitations of the existing performance indicators. First, WeibullCRI was examined using an extensive laboratory evaluation of 16 different asphalt mixes. The results indicated that WeibullCRI was sensitive to variation in binder content and binder PG and the results were in good agreement with the expected cracking resistance based on the composition of the studied mixes. In addition, WeibullCRI had low variability in test results and higher number of various statistical groups. Next, the applicability of WeibullCRI as a unified approach to analyze the results of various monotonic assessment tests was investigated using data generated by other researchers and reported in the literature. The results indicated that WeibullCRI is able to interpret the testing results of various monotonic performance assessment tests (i.e., IDT- intermediate temperature, Semi-Circle Bending [SCB]- intermediate temperature, SCB-low temperature, Disk-Shaped Compact Tension [DCT], and Simple Punching Shear Test [SPST]) and various displacement measurement methods (i.e., actuator vertical displacement and Crack Mouth Opening Displacement [CMOD]). WeibullCRI was also sensitive to variation in test conditions (i.e., specimen notch depth, thickness, and air void content) and mix composition proportions (i.e., binder content, binder grade, aggregate type, NMAS, aging, rejuvenator dosages, and Recycled Asphalt Pavement [RAP] materials).The second phase of this study reviewed and evaluated the current monotonic cracking performance assessment tests and indicators including the developed WeibullCRI used to assess asphalt mix resistance to cracking. In this phase, the testing requirements of various test standards, key publications, concepts, calculation methods, physical meaning, and advantages and disadvantages of various performance indicators were reviewed. Then, the study investigated the validity of the most promising testing standards and indicators. Three testing standards and 12 performance indicators were considered. Several aspects were examined including 1) investigate the fundamental meaning of the variation in the load-displacement curve in terms of the change in mix resistance to cracking, 2) sensitivity of performance indicators to mix compositions, 3) variability in test results, 4) number of various statistical groups, 5) correlation between various performance indicators, 6) direct correlation between laboratory results of monotonic performance tests and indicators with the observed field cracking, and 7) ability to develop PEMD specifications. A comprehensive laboratory investigation was conducted using 33 different asphalt mixes included six Laboratory Mixed-Laboratory Compacted (LMLC) and 10 Plant Mixed-Laboratory Compacted (PMLC) asphalt mixes, and 17 field projects with known cracking performance. The results showed that WeibullCRI calculated from the IDT test to have the lowest test variability, maximum number of Tukey's honestly significant difference (HSD) groups, and have excellent correlation with cyclic cracking resistance assessment indicators as compared to the other monotonic performance indicators. In addition, the results demonstrated that there was no direct correlation between all monotonic performance indicators and the observed field cracking performance, therefore an alternative approach was proposed, evaluated, and validated to develop performance thresholds for the selected performance indicators. Three pass/fail cracking performance thresholds were proposed for WeibullCRI to distinguish between asphalt mixes with good, fair, and poor cracking resistance using the proposed approach.The third phase of this study focused on the development and evaluation of a new cyclic cracking assessment test called Multi-Stage Semi-circle bending Dynamic (MSSD). The test offers advantages over the available monotonic and dynamic cracking assessment tests and addresses major concerns to implement the PEMD (i.e., performance test validity, complex specimen preparation, and testing time). The developed MSSD test simulates the repeated loading (cyclic) in a reasonable testing time (less than 9 hours per test regardless of mix type), has a fixed loading sequence that works for mixes with different characteristics (e.g., mix composition, percent air void content, thickness, etc.), and utilizes testing equipment and specimen geometry similar to that used in monotonic tests. The laboratory evaluation results showed that the proposed test and its derived performance indicators were sensitive to mix composition and had lower variability compared to other dynamic tests. In addition, the MSSD performance indicators correlated well with the observed cracking performance in the field and were able to distinguish between projects with good and poor resistance to cracking. Based on the evaluation results, three pass/fail cracking performance thresholds were proposed to distinguish between asphalt mixes with good, fair, and poor resistance to cracking.The fourth phase of this research examined the most promising tests and performance indicators to evaluate the resistance of asphalt mixtures to rutting. Two tests (i.e., Hamburg Wheel Tracking test [HWTT], and Asphalt Pavement Analyzer [APA] rut test) and three rutting performance indicators (i.e., HWTT rut depth after 15,000 passes [HWTT15000], HWTT rut depth at 20,000 passes [HWTT20000], and APA rut depth after 8,000 cycles [APA8000]) were considered. An intensive laboratory investigation was conducted that included six LMLC, 10 PMLC, and field cores extracted from 17 field projects. The research findings showed that both HWTT and APA rut test provided similar rutting assessment for the evaluated mixes. The study recommended using the HWTT over the APA rut test since HWTT can be also used to assess the resistance of asphalt mixtures to moisture damage to moisture damage. Also, the study recommended using HWTT15000 over HWTT20000 as a performance indicator since it requires less testing time.The final phase of this research provided recommendations of the best testing standards, performance indicators, and performance specifications to assess asphalt mix resistance to cracking and rutting. In addition, it provided guidelines to demonstrate the use of the proposed tools during the design and/or production of asphalt mixes. It also proposed standards testing procedures for the newly developed WeibullCRI performance indicator and MSSD test.

Asphalt overlays are commonly used to rehabilitate deteriorated Portland cement concrete (PCC) pavements. However, mechanically or thermally driven movements at joints and cracks in the underlying pavement usually lead to development of reflective cracks in the overlay. The formation and propagation of reflection cracking is controlled by the mechanical properties of the asphalt and the condition of the overlaid pavement. Current state of practice for asphalt overlay design is policy oriented and lacking an engineered design approach. There is need for establishing state of practice in design of overlays as well as for assessment of PCC pavement condition and recommending improvements to existing pavement prior to overlay construction. The objective of this study is to develop a simple decision tree-based tool for selecting suitable asphalt mixtures and overlay designs to prolong overlay lives by lowering reflective cracking and improving in-situ density. This research will leverage the current National Road Research Alliance (NRRA) effort of constructing, instrumenting, and monitoring 12 MnROAD test sections, laboratory performance tests on asphalt mixtures from the test sections, and past field performance data. The proposed tool incorporates field performance data, performance modelling, and life-cycle cost analysisto develop best practices for rehabilitation of PCC with asphalt overlays.

The first part of this dissertation is focused on comparing the mechanisms and fundamental properties that control the performance of asphalt mixtures during the uniaxial flow number testing. It is shown that aggregate packing, as measured using an image analysis method, is the key property affecting the failure and deformation characteristics in uniaxial testing. Additionally, it is shown that the main cause of tertiary flow of mixtures in the flow number test is structural instability and bulging (dilation) of aggregate skeleton. It is observed that dense graded mixtures with better aggregate packing showed a better rutting performance due to two principal mechanisms: first, lower stress level within the binder phase during loading due to aggregate skeleton serving as the main stress path, and second, the aggregates in proximity or contact showed a supporting structure (higher aggregate particle stability), which delays the tertiary flow in material, and reduces the rate of permanent deformation. The second part of this dissertation is focused on the mechanisms and factors controlling aggregate mobility and structure formation during compaction and their application for optimization of the construction process and for the development of a performance based mix design methodology. It is shown that for sufficient packing of aggregates during compaction, the viscosity of mastic should be low enough for lubrication, but high enough to prevent dry contacts of aggregates before the required aggregate packing is obtained. Depending on the stress state at the proximity zone of aggregates, there is a minimum mastic viscosity that is required to provide sufficient film thickness to prevent locking of the mixture at the early stage of compaction. A set of recommendations are proposed to estimate the optimum conditions for effective packing of aggregates. Based on the recommendations it is shown that packing of mixtures (as measured by total proximity zone length between aggregates) is a function of mastic viscosity, aggregate gradation, and compaction effort. The levels of these factors can be selected in order to achieve the required packing characteristic that is suitable for the design traffic loading.