MACHINE LEARNING-BASED COMPARATIVE STUDY FOR COMPRESSIVE STRENGTH PREDICTION OF ADVANCED CONCRETE MATERIALS

Abstract

High-performance concrete (HPC) is widely used in modern construction due to its superior mechanical and durability properties. Accurate prediction of compressive strength is essential for quality control and structural safety. Traditional empirical models often fail to capture the nonlinear behavior of concrete mixtures. In recent years, machine learning (ML) techniques have shown strong potential in predicting concrete properties. This study presents a comparative analysis of multiple machine learning algorithms for predicting the compressive strength of high-performance concrete. Linear Regression, Support Vector Regression, Random Forest, and Artificial Neural Networks are evaluated. Experimental datasets collected from laboratory-tested HPC mixtures are used for model training and validation. Performance is assessed using statistical indicators such as accuracy and root mean square error. Results show that ensemble and neural-based models outperform conventional approaches. The study demonstrates the effectiveness of ML models in improving prediction accuracy and supporting intelligent concrete mix design