Deflection Prediction of Rehabilitation Asphalt Pavements through Deep Forest

Abstract

The deep forest is a powerful deep-learning algorithm that has been applied in certain fields. In this study, a deep forest (DF) model was developed to predict the central deflection measured by a falling weight deflectometer (FWD). In total, 11,075 samples containing information related to pavement structure, traffic conditions, and weather conditions were extracted from the LTPP dataset. The performance of the DF model with custom backend settings was compared with that of models random forest (RF), multilayer perceptron (MLP), and DF built on the sklearn backend. All four deep-learning algorithms could identify the complex relationship between central deflection and relevant feature variables with high accuracy and stability. The learning and generalization abilities of DF was stronger than those of MLP and RF. The predictive performance and computation time of DF (custom) were better than those of DF (sklearn), indicating that the custom model was superior to the highly encapsulated model with sklearn as the backend. Feature importance analysis indicated that the drop load of FWD was the key factor influencing deflection. In addition, structural number, annual precipitation, and annual kilo equivalent standard axle load (kESAL) are very important features related with deflection. The feature importance of rehabilitation improvement thickness was less than the drop load, climatic factors, kESAL, structural number, and layer thickness.