Experimental validation of a fractional model for creep/recovery testing of asphalt mixtures

Prediction of asphalt mixtures’ behavior during their service life is a challenge due to its complexity and sensitivity to environmental and loading conditions. It has been proved that, when subjected to loading conditions comparable with most pavement operating conditions, asphalt mixtures behave as linear visco-elastic (LVE) materials. Traditionally the LVE behavior of bituminous material is modeled via creep/recovery functions. In the past, several rheological models constituted by elastic and viscous elements arranged in series or in parallel (analogical models) have been proposed and specified for both bitumen and asphalt mixtures. The corresponding constitutive laws always involve first order derivatives of time with exponential type solutions but problems in setting parameters arise when both the creep and recovery behavior have to be modeled. In this paper it is shown that experimental creep data follow a power decay law, rather than an exponential one. As a consequence, a simple fractional model is here proposed for predicting creep/recovery behavior of asphalt mixtures with a small number of parameters and low computational efforts with respect to the classical analogical models. The proposed model is then calibrated by a best fitting procedure on experimental data from creep and creep/recovery tests carried out on asphalt mixtures under different load and temperature conditions.