Enhancing Concrete Performance: Machine Learning Models for Predicting and Classifying Tensile Strength

Accurate prediction and detection of concrete’s mechanical properties, such as tensile strength are critical for optimizing structural performance. For performing this study, a comprehensive dataset, including cement content, water-cement ratio, aggregate types, and additives were utilized. This study introduces a novel approach that integrates Machine Learning (ML) techniques and novel proposed model to predict and classify the tensile strength of concrete, combining both tasks for a more Comprehensive evaluation. Advanced models, such as XGBoost and ANNs, were fine-tuned using hyperparameter optimization to enhance predictive accuracy. The models demonstrated significant improvements over traditional regression methods, with XGBoost achieving an R<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> of 0.9993 and the lowest MSE (0.0129) and RMSE (0.0013). For classification, the proposed ANN model achieved 99.89% accuracy. This study introduces a novel machine learning framework integrating prediction and classification tasks, Although, the research not only showcases the superior performance of these models but also highlights their potential to reduce the need for extensive physical in concrete engineering. By integrating machine learning with feature engineering, this study paves the way for more efficient, data-driven design processes and broader applications in the construction materials science.