Artificial intelligence approach to predict the dynamic modulus of asphalt concrete mixtures

Abstract

This paper develops an Artificial Neural Network (ANN) model based on 96 experimental data to forecast the dynamic modulus of asphalt concrete mixtures. The accuracy of the models was assessed using numerous performance indexes such as the Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), and coefficient of determination (R2). In addition, this study applied the repeated K-Fold cross-validation technique with 10 folds on the training data set to make the simulation results more reliable and find a model with more general predictive power. According to the research findings, the ANN model accurately predicts the dynamic modulus |E*| of asphalt concrete mixtures. Furthermore, the ANN model could successfully predict the dynamic modulus |E*| of asphalt concrete mixtures with a remarkable R2 = 0.989.