Description

Machine learning algorithms are powerful AI tools that have demonstrated strong robustness and reliability in the performance prediction of road infrastructure. In this study, the authors have attempted to develop a prediction model to estimate the transverse cracking in jointed plain cement concrete pavements using three machine learning approaches, namely, decision tree regression (DTR), random forest regression (RFR), and deep neural network (DNN). The results show that the DNN model outperformed DTR and RFR with coefficients of determination (R²) greater than 0.95 in both training and testing data sets. Performance metrics are summarised and presented for all three methods used in this study.