Harnessing synergy of machine learning and nature-inspired optimization for enhanced compressive strength prediction in concrete

Abstract

Concrete made with additives like slag and fly ash has revolutionized construction by reducing carbon emissions, minimizing waste, lowering labor costs, and enhancing durability and accuracy. Predicting the compressive strength (CS) is vital for achieving optimal performance. Given the nonlinear characteristics of supplementary cement material concrete (SCMC) mixtures, researchers are increasingly turning to machine learning methods. This study assesses nine machine learning models, integrating conventional AI algorithms, such as artificial neural network (ANN), support vector regression (SVR), and random forest (RF) with nature-inspired optimization techniques including chicken swarm optimization (CSO), moth flame optimization algorithm (MFO), and whale optimization algorithm (WOA). By addressing issues related to mechanical property variation, dataset coverage, and model evaluation, the study achieved high prediction accuracy across all nine models. The RF model optimized with CSO, MFO, and WOA consistently performed well across various metrics having R² = 0.98, RMSE = 0.03 during training and R² = 0.87 and RMSE = 0.07 during testing. The visual evidence highlights several advantages, including superior quality control, cost savings, increased safety, and environmental sustainability, which underscore the effectiveness of these models. In addition, feature analysis was performed using SHAP analysis, age and cement are identified as the dominant inputs exacting influence on the CS of SCMC.