Comparative Study of Tree-Based Methods for Concrete Compressive Strength Analysis

This research paper presents a comparative analysis of tree-based machine learning methods for predicting the compressive strength of concrete. Given the critical role of compressive strength in determining the quality and longevity of concrete structures, accurate prediction models are essential for improving construction outcomes and material efficiency. This study evaluates several prominent tree-based algorithms, including Decision Trees, Random Forests, Gradient Boosting Machines (GBM), and Extreme Gradient Boosting (XGBoost), to identify the most effective method for predicting concrete strength based on its composition and curing time. Each model is trained and tested using a dataset comprising various mix designs and curing conditions. Performance metrics such as Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and R² are employed to assess the accuracy and reliability of each model. The findings indicate significant variations in the predictive accuracy of these methods, with XGBoost demonstrating superior performance across most metrics. The paper concludes with a discussion of the practical implications of these results for the construction industry, particularly in optimizing concrete mix designs for desired strength outcomes and sustainability considerations.