Fatigue Analysis and Predicting of Hot Mix Asphalt Based on a Fracture Mechanics and Damage Density

In this study, fracture mechanics approach based on modified Paris' law was used for developing constitutive models for predicting fatigue life for HMA tested in dynamic shear rheometer using small cylindrical samples under strain and stress test modes. The methodology of developing constitutive models are based on damage density concept through the relationships between damage state of materials with mechanical properties and volumetric characteristics extracted from tested samples. Where the fundamental fatigue parameters represented by the exponential relaxation coefficient (m), relationships of fracture index (FIc) with normalised stiffness modulus (CN) and cumulative dissipated pseudo-strain energy (DPSE) against number of cycles for strain and stress test modes. Two types of mixes: hot rolled asphalt and dense bitumen macadam (DBM) were fabricated using two types of aggregate (limestone and granite) with two binder grades (40/60 and 160/200) pen. Results analysis showed the capability of these models in predicting the fatigue performance of HMA in terms of number of cycles at 50% and 20% of initial stiffness modulus for strain and stress test modes. Also, bias analysis for the predictively of models revealed that the strain model's prediction accuracy was better than that of the stress model. Analysis of the results also showed that a higher DPSE/N ratio indicates lower fatigue life and greater damage.