Abstract
This study analyzes data obtained from laboratory prepared asphalt specimens by means of statistical tools. The complex permittivity of 111 laboratory prepared asphalt specimens used as the bearing and wearing courses of pavements is measured under dry and wet conditions. A multiple linear regression model is developed to relate the complex permittivity of the specimens at 201.6 MHz and the moisture content, bitumen content, aggregate mix type, specimen length and bulk density. The sensitivity and significance of the regression model and individual variables are evaluated using statistical analysis tools. The statistical analysis indicates that the regression model is highly significant. The complex permittivity of the asphalt specimens is sensitive to changes in water content, bitumen content, aggregate type and pavement thickness. Based on the results, the complex permittivity of asphalt pavements has the potential to serve in non-destructive testing for inspection of new and existing pavements upon obtaining more data over larger ranges of pavement properties and further development of in-situ measurement techniques.
Acknowledgements
The data used in the analysis were measured by Mr J.A. Umana during his study toward a degree of M.E.Sc. The research is funded by the Natural Science and Engineering Research Council of Canada.
Notes
†In this paper, the relative complex permittivity (ε*, unitless) is used throughout, which is defined as the complex permittivity relative to free space with the permittivity of \mgreek{e} _{o} = 8.854 \times 10^{-12} \hspace{0.167em} F/m. The term “relative” is often omitted for abbreviation in the paper.
‡Defined as the ratio of the bulk density of the asphalt specimen and the density of water at 4°C.