ABSTRACT
Pavement design methods have been developed from the purely empirical 1993 AASHTO Guide towards the more realistic Mechanistic-Empirical Pavement Design Guide (MEPDG). The ME Design is more accurate and realistic as it overcomes the limitations of the empirical pavement design methods through the incorporation of the principles of material mechanics. Still, it lacks a robust quantification of the reliability of the resulting pavement design. ME Design recognizes that pavement performance is governed by a large amount of uncertainty and variability related to design, construction, traffic loading, and climatic conditions over the expected design life. However, ME Design provides an analytical solution that incorporates reliability uniformly for all pavement types allowing the design of a pavement within a desired level of reliability. This study aims at studying the effect of the uncertainty in the dynamic modulus (|E*|) of asphalt concrete (AC) on the performance prediction using MEPDG. Monte Carlo Simulations are used to model the uncertainty in the |E*| mastercurve of various AC mixtures. Simulated |E*| mastercurves are used as input for MEPDG to predict rutting and fatigue cracking for various pavement structures under different loading speeds and climatic conditions resulting in a suit of 15,000 MEPDG runs. A sensitivity analysis is carried out to check how the uncertainty in |E*| will be forward propagated to predict the performance of pavements. Such results are utilized to assess how uncertainties in material properties, presented by |E*|, need to be incorporated in quality assurance practices of pavement construction.
