This paper presents the results of a study to evaluate the fatigue performance of pavements that contain RAP materials. The study is based on two different accelerated loading tests conducted in the United States. Each of these two projects has corresponding pavements with and without RAP mixtures that were constructed over the same aggregate base and subgrade and subjected to the same traffic loading and environmental conditions. The primary reason for selecting these two projects is that they exhibited opposite trends in a comparison of the fatigue cracking performance of their pavements with and without RAP. The asphalt mixtures in the test pavements were characterized using the simplified viscoelastic continuum damage model. The fatigue cracking performance of the pavements was then modeled using the characterized material properties, in situ pavement structures and traffic loads, and climate conditions in FlexPAVE™, a three-dimensional layered viscoelastic finite element program with moving load analysis. The resultant fatigue cracking performance predictions matched the field observations well. Both the simulation results and the field observations also showed that the fatigue cracking performance of the asphalt mixtures that

Thermal cracking is the predominant failure mode for asphalt pavements constructed in regions with severe air temperature drops or significant daily temperature variations. The thermal stress restrained specimen test (TSRST) is one of a few tests that can simulate thermal loading in the field. Conducting TSRSTs is a time-consuming task; therefore, a predictive methodology is needed that can predict the fracture temperature of asphalt mixtures measured by TSRSTs. To this end, this paper utilizes dissipated pseudo strain energy (DPSE), which can be predicted using simplified viscoelastic continuum damage theory, as a failure criterion to predict the fracture temperature of asphalt mixtures. In this study, the fracture temperatures of eight asphalt mixtures used in the surface layer of the Minnesota Department of Transportation’s research facility, MnROAD, pavement sections were predicted using the DPSE failure criterion. The results show that the DPSE failure criterion is able to differentiate the mixtures’ thermal cracking performance based on differences in the mixtures’ volumetric properties. The results also indicate that the predicted mixture performance matches the field rankings reasonably well.

Thermal cracking is one of the most prevalent types of distress found in asphalt concrete pavement sections. The coefficient of thermal contraction (CTC) is the intrinsic parameter that determines the thermal contraction of asphalt mixtures subjected to temperature drop. Thus, thermal stress and associated thermal damage are greatly affected by the CTC of the mixture. The direct measurement of a mixture’s CTC is the most reliable and at the same time cumbersome method for predicting the thermal contraction of asphaltic mixtures. In this study, for Level I analysis, we measured the CTCs of mixtures that have a wide range of properties and reported the data. In order to remedy the difficulties associated with directly measuring CTC, we herein propose a simple yet accurate composite model. The suggested approach enables pavement engineers to predict CTC values as the temperature drops. The formulation requires the mixture’s volumetric properties, elastic modulus, and CTC of the aggregate a priori. With regard to binder, the formulation requires the CTC and relaxation modulus as input parameters. This procedure constitutes the Level II analysis for which the binder CTC is required beforehand. A database that constitutes a basis for estimating binder CTC was developed from measurements reported in the literature. The database can be used to fill the value for binder CTC in the developed composite model for the sake of predicting mixture CTC. This procedure constitutes Level III analysis. Levels II and III only differ with respect to the input value for binder CTC. The CTC values from the measurements and predicted from the composite model (Levels II and III) are compared. In order to fully understand the effects of Level II and III errors in estimating mixture CTCs, we also compared the thermal stress obtained from the predicted and measured CTCs.