Safi F, Hossain K, Wu S, Al-Qadi I, Ozer H. (2018). Modeling Creep Compliance of Asphalt Mixtures Containing Recycled Materials. Construction & Building Materials. Elsevier. https://doi.org/10.1016/j.conbuildmat.2018.06.232

Abstract:
Thermal cracking is one of the major distresses in flexible pavements, especially in the northern part of the U.S. and Canada. With the increasing use of recycled materials like reclaimed asphalt pavement (RAP) and recycled asphalt shingle (RAS), asphalt pavements have become more susceptible to thermal distresses. Therefore, such asphalt concrete (AC) mixtures should be carefully investigated to ensure that performance is not compromised in the pursuit of an economical solution. The Pavement ME Design uses the empirically developed model for the prediction of creep compliance as an input for Level 3 analysis, which uses the mix and binder properties. However, this prediction model cannot be applied for every climatic and mix design conditions; therefore, different researchers came up with modified models for different states within the U.S. However, the impact of recycled materials was not captured in the MEPDG as well as in the modified models. This research proposes a new model that captures the effect of recycled materials. First, creep compliance master curves were developed using time-temperature superposition, which were used in extensive exploratory data analysis to better understand the effect of various experimental variables on creep compliance. It was observed that softer binder and higher asphalt content results in more compliant, while recycled materials tend to decrease the compliance. Hence, softer binder and recycled materials tend to counterbalance each other’s impact. From model analysis, it was found that the existing MEPDG model significantly overestimates creep compliance compared with actual measured values. The proposed model is capable of capturing the effect of asphalt binder replacement (ABR). The proposed model has been validated with randomly selected data and found to exhibit good correlation with lab testing data.