Statistical and dimensional analysis of hot-mix asphalt mixture characteristics on asphalt pavement analyser rutting behaviour

摘要

This study presents a statistical and dimensional approach as a simple and efficient means to characterise the effects of individual mixture variables on the performance behaviour resulting from the laboratory tests of hot-mix asphalt (HMA) mixtures. The approach is to find the material and/or mixture variables that significantly affect the performance test results and to further assess the level of significance of variables, with the goal of improving the current understanding of HMA performance tests and their practical applications. To demonstrate the approach, two sets of asphalt pavement analyser (APA) test data from Nebraska and one set of data from Kentucky are statistically analysed based on the multiple linear regression technique. A dimensional analysis based on the Buckingham π-theorem is also performed, and the analysis results are compared with the statistical analysis results. Both analyses present comparable results in that the binder stiffness of HMA mixtures is the most significant variable affecting APA performance test results when the mixtures meet volumetric requirements.