Prediction of Binder Content in Glass Fiber Reinforced Asphalt Mix Using Machine Learning Techniques

Several researchers have reported the results of adding a variety fibers to asphalt concrete described as fiber-reinforced (FRAC). This research paper finds most suitable prediction model for Marshall Stability and optimistic bitumen content useful in glass mix by performing tests further analyzing data consonance with published research. Four machine learning approaches were used find best i.e., Artificial Neural Network, Support Vector Machine, Gaussian Process, Random Forest. Seven statistical metrics evaluate performance applied models Coefficient correlation (CC), Mean absolute-error (MAE), Root mean squared error (RMSE), Relative absolute (RAE), relative (RRSE), Scattering index (SI), Bias. Test testing stage indicated that Machine (SVM_PUK) performs validation amongst all CC values 0.8776 MAE 1.2294, RMSE 1.9653, RAE 38.33%, RRSE 55.22%, SI 1.0648 Bias 0.5005. The Taylor diagram dataset also confirms based on SVM outperforms other models. Results sensitivity analysis show about 5% has significant effect Stability.