Using mechanistic-empirical simulations and nonparametric testing to explore variability in the expected performance of reference asphalt pavement designs

Abstract

This study explored the variability in expected performance of eight reference asphalt pavement designs using AASHTOWare pavement ME software. Nonparametric tests showed that structures with a high-modulus asphalt base on an unbound base significantly delayed maintenance. Varying pavement layer materials significantly affected the simulations, except the asphalt surface and asphalt base (when used on an unbound base). Moreover, recycling rates in AC asphalt bases over 60% significantly improved performance over no recycling; however, air void content and effective binder percentage seemed to be key in these findings. Finally, this study showed that meeting material specifications does not guarantee equal performance.