Revisiting the correlation between the dynamic modulus and the flexural modulus of hot mixture asphalt

The compressive dynamic modulus (E*) and the flexural modulus (M-F) of an asphalt mixture are two of the main criteria in structural design and evaluation of flexible pavements. The correlation between these properties and the characterisation of their trends indicates a dynamic response of the asphalt concrete mixtures. In this study, a non-dimensional compressive dynamic modulus index (phi E*) and a non-dimensional flexural modulus index (phi M-F) are proposed as ways to characterise the trends of changes of the dynamic modulus at different temperatures and frequencies. Analysis of Variance (ANOVA) and Ridge Regression were used in analysis of different variables of experiments and statistical modelling, respectively. ANOVA results showed that the proposed parameters are function of the temperature, frequency, and the interactions. Also, the analysis of trends showed that the values of phi M-F converge at 3% as temperature increases, while only a decreasing trend was observed for phi E*. Furthermore, the correlation between phi M-F and phi E* depends on the type of binder. The effect of binder type can be seen in the master curves. Furthermore, the analysis showed that the values of M-F are at least 75% of E*, so this can be used as proximity factor to convert M-F into E* for use in mechanistic pavement design. In addition, a statistical predictive model was proposed as function of M-F, temperature and frequency. The outputs showed that M-F and temperature at the frequency domain have significant effects on E* with a high correlation. In conclusion, the proposed parameters and the model can be used as a basis for development of a universal predictive model of dynamic modulus through trends of changes and material properties. (C) 2021 Elsevier Ltd. All rights reserved.

Automated summary

The study proposed non-dimensional indices to characterize the trends of dynamic modulus changes at different temperatures and frequencies. ANOVA results indicated parameters are functions of temperature, frequency, and interactions. Furthermore, trends analysis showed that phi M-F values converge at 3% as temperature increases, while phi E* shows a decreasing trend.