Using asphalt mixture beams in the bending beam rheometer: Experimental and numerical approach

The paper is presented to validate the use of asphalt mixture beams in the Bending Beam Rheometer (BBR) test. Experimental work was performed to address the concerns regarding the sources of variability of using asphalt mixture beams for the BBR tests. The dimensional differences on the accuracy of deflections of a beam are evaluated using finite element method and statistical analyses. The effect of air voids on the overall stiffness of asphalt mixtures is found to be less significant at low temperatures through the volumetric measurements on three nominal maximum aggregate sizes NMAS (i.e., 12.5 mm, 9.5 mm, and 4.75 mm) mixture beams. The homoscedasticity of variances among the 12.5 mm, 9.5 mm, and 4.75 mm NMAS mixture groups is statistically validated using the Bartlett's test, so the measurements and comparisons among these NMAS mixture groups are valid for the BBR tests. A linear viscoelastic analysis (LVE) was used to evaluate the thermal properties of the three NMAS mixture beams. The LVE results show that the effect of the aggregate size on the low temperature properties of these three NMAS mixtures is not significant. A band of the confidence interval was established to show highway agencies how to run the Quality Control/Quality Assurance (QC/QA) operations using data collected from the BBR testing. Through the experimental and numerical work, the sources of variability during a BBR test are evaluated and the concerns are addressed. The use of asphalt mixture beams in the BBR tests to control the thermal-induced failure in asphalt pavements is experimentally and numerically validated and properly used for QC/QA.