Investigations on cooling effects of prepared pavement coatings using the Grubbs method and linear regression analysis

Abstract

The purpose of this study is to improve the evaluation accuracy of the cooling effect of road coatings using a new data processing method based on the Grubbs method and linear regression to exclude abnormal values and improve data accuracy. Five kinds of new pavement cooling coatings are developed as the main objects of this study, whose road performances could fulfil the requirement of the current asphalt pavement. We have also selected the optimal regression variable for the cooling effect of the cooling coatings. It is found in this study that the correlation coefficients of the different cooling effects’ regression formulas remained around 0.99, suggesting that there is significant correlation among the fitting variables. When the temperature of the asphalt pavement reached 60°C, the cooling effect of the new cooling coating is up to nearly 10°C. The data processing method developed in this paper improved the inaccuracy problem of experimental data caused by different environmental factors.