Prediction of High Temperature Performance of Asphalt Mixture Based on Random Forest

In this paper, the grey relational method is used to analyze the factors that affect the high temperature performance of asphalt mixture, and on this basis, the dynamic stability ofAC13 asphalt mixture model is predicted by random forest. According to the grey correlation analysis, four influencing factors such as 25°C penetration degree, softening point, 135°C rotational viscosity and whey to stone ratio were determined as the input factors of the random forest model, and the random forest model was established and trained, tested and tested. The maximum MAPE in the prediction results is 8.541%, the predicted rut dynamic stability is in good agreement with the measured results, and the random forest prediction model has good accuracy, which can be used to predict the high temperature performance ofAC-13 asphalt mixture.