Fatigue behaviour of conventional and rubber-modified gap-graded asphalt mixtures using bending and axial fatigue tests

The study presented in this paper assesses and compares the laboratory fatigue performance of two gap-graded asphalt concrete (AC) mixture: one using crumb rubber (CR) (rubber-modified) and without using CR (conventional). The rubber-modified and conventional gap-graded mixtures were sampled from the E18 highway located in the Stockholm area of Sweden. The testing program consisted of dynamic modulus, |E*|, bending beam, and axial fatigue tests. The |E*| master curves of the respective AC mixtures and their generalised fatigue model coefficients (k ₁, k ₂, k ₃) from the axial fatigue tests were used to predict and compare the long-term pavement performance of each mixture using the AASHTOWare Pavement ME Design software. This analysis was performed systematically using parametric variations to assess the sensitivity of the predicted performance to individual material factors. Fatigue test results from both bending beam and axial fatigue tests showed clear evidence of improved fatigue behaviour for the rubber-modified gap-graded mixture compared to the conventional gap-graded mixture. The Pavement ME design analysis showed that the rubber-modified gap-graded mixture outperformed the conventional mixture. The parametric sensitivity analysis showed that this improved behaviour was the result of an increase in k ₁ and decrease in k ₂ of the fatigue life relationships at individual test temperatures of the rubber-modified gap-graded mixture. It was also found that the |E*| and the fatigue coefficient k ₃ of the rubber-modified gap-graded mixture has an adverse impact on its fatigue performance compared to that of the conventional gap-graded mixture.