Harnessing AI-driven modeling to assess the impact of alternative materials on the compressive strength of concrete mix design

Abstract

Concrete, as the most extensively used construction material, contributes significantly to environmental degradation due to the high consumption of natural resources and carbon dioxide emissions. To foster sustainable development, this study investigates the incorporation of alternative materials Fly Ash and Rice Husk Ash as partial replacements for cement in M25 grade concrete. The research evaluates both the compressive strength and workability of these modified mixes. Furthermore, machine learning techniques, including XGBoost, Random Forest, and Support Vector Machine (SVM), were employed to predict the compressive strength based on experimental data. A user-friendly prediction system was developed to enable analysis by selecting either Fly Ash or Rice Husk Ash as the replacement material. Among the models used, XGBoost outperformed the others in terms of predictive accuracy, achieving the highest \(\hbox {R}^{2}\) score and lowest error metrics. The results indicate that these alternative materials can enhance concrete properties at specific replacement levels, and that machine learning models, particularly XGBoost, offer accurate and efficient predictions. This study underscores the potential of integrating sustainable materials with data-driven modeling for eco-friendly and performance-optimized concrete mix designs.