Estimation of the Robustness of the XGBoost Regression Model in the Prediction of Pavement’s Mechanical and Geometrical Parameters Based on Static Interpretation of the FWD Test

The FWD is commonly used to conduct a non-destructive evaluation of the capacity of the pavement. The layered pavement is loaded locally by falling weight, and deflection at many points is recorded. Based on these results, knowing the pavement geometry, the mechanical properties of the pavement may be determined using the back-calculation approach. Some methods are used for back calculation, such as analytical, numerical, or ML. An analytical solution for a multi-layered structure leads to non-linear relationships for the thickness or stiffness of each layer, but gives an accurate solution. The other methods, like numerical or ML methods, are just approximation methods with different levels of accuracy. In this paper, the robustness of the XGBoost ML regression model in predicting mechanical and geometrical pavement parameters was estimated. The database was generated from a static analytical solution of an axially symmetrical problem implemented in the form of JPAV software and then explored by training regression models to predict moduli and thickness of pavement layers. Two other databases were created using PCA (Principal Component Analysis) and FDM (Feature Difference Method) to compare models trained with the complete deflection database. The results showed that models trained with the complete deflection database had the best average prediction performance compared to the other two. In contrast, models trained with the database pre-processed by PCA showed predicting performance similar to the previous models, but with a little precision loss. Models trained with the database pre-processed by FDM exhibited excellent prediction on some features but worse on the rest.