Development of a predictive model for the compressive relaxation modulus of asphalt mixtures

Abstract

Relaxation modulus as an important viscoelastic property of asphalt mixes is usually determined using time-consuming and expensive test methods. Therefore, this research was carried out to develop a predictive model for the time-dependent compressive relaxation modulus of asphalt mixtures based on the mix specifications, volumetric properties, ageing condition, and temperature. To this end, crushed stone aggregates together with 60/70 penetration asphalt binder (corresponding to PG 64-22) were used to prepare the asphalt mixture specimens at two aggregate gradations, two binder contents, two air void levels, and three ageing conditions with four replications. Direct compression relaxation modulus tests were conducted at four different temperatures, and the sigmoidal model was used together with the numerical temperature-shifting technique to construct the compressive relaxation modulus master curves for all the experimental combinations. Statistical analysis of variance and regression analysis were then performed on the test data, and a model was developed to predict the compressive relaxation modulus values. Finally, the proposed predictive model was validated using another set of data previously published by other researchers.