Investigating different approaches for evaluation of fatigue performance of warm mix asphalt mixtures

Fatigue is one of two major performance issues in flexible pavements that caused by repeated traffic loading. There are several methods for determination of fatigue life like 50% stiffness reduction (AASHTO), energy-based methods, and fracture mechanics. The primary objective of this study is exploring the effect of choosing different methods to determine the fatigue life of warm mix asphalt (WMA) in comparison with hot mix asphalt (HMA). WMA mixtures were prepared using Aspha-min and Sasobit additives. Four-point flexural beam test was employed according to AASHTO T321 to evaluate fatigue performance of WMA and HMA mixtures. The tests were conducted at four strain levels 650, 800, 1000, and 1200 (more than typical strain levels) and two frequencies of 7.5 and 15 Hz at 20 degrees C. The fatigue life of asphalt mixes determined using three different approaches including AASHTO, Rowe and Bouldin (R&B), and rate of dissipated energy changes (RDEC). In addition, a fatigue damage analysis using linear (Ecole Nationale des Travaux Publics de l'Etat) ENTPE model was performed to compare damage evolution of WMA and HMA mixtures in given loading conditions. AASHTO method was a more sensitive approach for evaluating effects of loading conditions on fatigue life of WMA mixtures. Based on the AASHTO method results, the fatigue life of WMA mixtures at 15 Hz was lower than HMA. In practical strain levels and at 7.5 Hz, WMA was more fatigue resistant than HMA. The fatigue lives estimated using both of the energy methods (R&B and RDEC) were approximately same. They predict fatigue lives of WM Amixtures about five times of standard AASHTO method, specifically using RDEC method. In addition, damage analysis method justifies that the AASHTO is not a reliable approach for fatigue life prediction and the energy methods are better approaches.