Investigation of Artificial Neural Network Models for Predicting the International Roughness Index of Rigid Pavements

AbstractThe International Roughness Index (IRI) is the one of the most important roughness indexes to quantify road surface roughness. In this study, the backpropagation (BP) algorithm and conjugate gradient backpropagation algorithm were used to develop artificial neural network (ANN) model for the prediction of the IRI. A total of 913 samples in the case study of the experimental study was the Vietnamese Highway No.5 between Hanoi and Hai Phong, located in the northern part of Vietnam including 6 inputs and 1 output were collected for training and testing the ANN model. The reliability of ANN model is evaluated by some criteria such as correlation coefficient (R), root mean square error (RMSE) and mean absolute error (MAE). The best ANN architecture could be considered as a new tool for accurate prediction of the IRI for evaluation of quality of road surface roughness.KeywordsInternational roughness index (IRI)Rigid pavementsConcreteArtificial neural networkMachine learning