Probabilistic and Sensitivity Analyses Applied to the Study of an Asphalt Pavement with a Prony Series Modeled Overlay

Abstract

Probabilistic and sensitivity analysis of the mechanical behavior of an asphalt pavement subjected to varying loads over time, originating from vehicle traffic, is discussed in this study. The pavement is structured in three layers: an asphalt overlay and a granular base, supported by a subgrade. The probabilistic model of the system is implemented using the Finite Element Method in the ANSYS software. The 2D model's input variables include the geometry and mechanical properties of the pavement and subgrade, and they are implemented using the Monte Carlo method. A two-dimensional quadratic element with eight nodes and two translational degrees of freedom per node is used in modeling the system. The asphalt overlay is assumed to be viscoelastic-linear and is modeled using the Prony series, while the other materials are linear-elastic. The maximum average vertical displacement experienced by the asphalt overlay is the only random output variable of the system, and its value is compared and validated against deterministic results provided in the scientific literature. The probabilistic results demonstrate, with 95% probability and confidence, that the average vertical displacements are greater than their corresponding deterministic values. The uncertainty associated with the average vertical displacement tends to be higher at the point where the vertical load reaches its maximum value, decreasing and stabilizing over the analysis period. Sensitivity analysis shows that the asphalt overlay thickness and the base's elastic modulus are the parameters that most influence the maximum vertical displacement values and the permanent displacement of the pavement, respectively.