Sensitivity analysis of a new dynamic modulus (| E*|) model for asphalt mixtures

Abstract

Asphalt pavement designers have been using the dynamic modulus (| E*|) as a fundamental parameter to reflect the hot-mix asphalt (HMA) performance, and since | E*| is considered as an essential input in the new AASHTO Mechanistic-Empirical Pavement Design Guide (MEPDG), many models were developed for evaluating and/or predicting this property for HMA. This article focuses on the verification of a newly developed model to predict | E*| by utilising robust sensitivity analyses through tornado plots and extreme tail analyses, which are used to identify the relative importance of the input variables on the predicted | E*|. Results showed that the binder dynamic shear modulus (| G*|) has the most significant effect, followed by the air void content, on the predicted | E*|, and both properties contributed largely to the extreme predicted values.

Keywords:

• asphalt mixes
• dynamic modulus
• sensitivity analysis
• tornado plot
• extreme tail analysis