Assessment of different flexure fatigue failure analysis methods to estimate the number of cycles to failure of asphalt mixtures

Load associated fatigue cracking is one of the major distress types occurring in flexible pavements. Beam fatigue testing has been used for several decades and is considered an integral part of advanced characterization procedures. However, there have been several fatigue failure analysis methods based on the failure stiffness of the mixture. These methods may not produce the same results, and vary depending on the method of detecting the failure point. In this paper, beam fatigue test results conducted at Arizona State University were used for the analysis. The mixtures included conventional, Asphalt Rubber (AR), and Fiber-Reinforced (Aramid and Polyolefin) asphalt mixtures. Deflection controlled fatigue tests were conducted according to American Association of State Highway and Transportation Officials (AASHTO T321) procedure at three temperatures. Paired-t statistical analysis approach was utilized to compare five different fatigue failure methods. These methods are commonly known as the Pronk, Hopman, Carpenter, ASU, and Rowe. An additional method using the Franken fitting model was also introduced. A recommendation is presented to select a potential method that unifies the current fatigue analyses using a rational energy-based approach. This new method provides a well-defined fatigue failure point compared to other methods.