ABSTRACT
Knowledge of the predicted strains and deformations of the individual pavement layers under loading is required for the purpose of accurate pavement design and evaluation. In the current study, a sensitivity analysis of predicted flexible pavement responses is performed through Finite Element Method (FEM) in order to quantify potential variations in the mechanical responses. Variations in responses are investigated depending on different underlying assumptions including the unbound materials variability (nonlinear anisotropic behavior with initial stress-state), variable asphalt concrete layer thickness and modulus of elasticity. The importance of the input assumptions on the derived responses estimated in the nonlinear analysis up to 3.5 times larger than the responses from a corresponding linear analysis is highlighted. This study provides evidence that performance modeling should potentially act as a risk assessment tool for pavement scientists and engineers engaged in both pavement design and QA/QC practices in order to assist reliable performance prediction in a way that bridges the knowledge gap between modeling and pavement construction processes.
