A machine learning method for predicting the chloride migration coefficient of concrete

This work adopts a state-of-the-art machine learning algorithm, XGBoost, to predict the chloride migration coefficient (Dnssm) of concrete. An extensive database of experimental data covering various concrete types is created by gathering from research projects and previously published studies. A total of four Dnssm models are developed depending on the number and type of input features. All models are verified with unseen data using four statistical performance indicators and compared to other five tree-based algorithms. The verification results confirm that the XGBoost model predicts the Dnssm with high accuracy. The model has the potential to replace cumbersome, time-consuming and resource-intensive laboratory testing.