Evaluation of Test Data Variability and Suitability of Predictive Equation Related to the Determination of Dynamic Modulus (E*) for Hot-Mix Asphalt

Abstract

The dynamic modulus (|E*|) of hot-mix asphalt (HMA) is one of the fundamental inputs in the mechanistic-empirical (M-E) Design Guide developed in NCHRP Project 1-37A. The M-E Design Guide provides three levels for |E*| inputs, which depend on the importance of the pavement in service. Level 1 |E*| inputs require laboratory measured |E*| values while level 2 and 3 |E*| inputs are estimated using a predictive equation. To provide the laboratory measured |E*| inputs for implementation of the M-E Design Guide, a significant |E*| testing program was completed in Arkansas. The testing program included three replicate specimens from each of four aggregate types, three nominal maximum aggregate sizes, two PG binder grades, and two air-void levels. The |E*| tests were conducted using five test temperatures and six loading frequencies. The |E*| values obtained in this testing program exhibited much lower variability than those in other studies and complied with the required variability level specified in AASHTO TP 62-03. Therefore, the laboratory measured |E*| values can be used with confidence for level 1 material characterization inputs for HMA in the M-E Design Guide. Level 2 and 3 |E*| inputs, estimated using the predictive model provided sufficiently accurate predictions of pavement performance compared to performance predictions generated by the Level 1 inputs . In addition, the differences between level 2 and 3 predicted distresses were not significant. Thus, level 3 |E*| inputs could be used with confidence for initial implementation of the M-E Design Guide.