"AI-Powered Predictive Analytics for Concrete Compressive Strength with Material Impact Interpretation"

Abstract: Because it directly affects the quality, longevity, and safety of structures, concrete compressive strength is a crucial factor in the construction sector. Conventional laboratory-based strength testing is expensive, time-consuming, and labor-intensive. This study suggests an AI-powered predictive analytics framework for evaluating concrete compressive strength utilizing machine learning approaches in order to get over these restrictions. A concrete compressive strength dataset that includes mix design factors such cement, water, fly ash, blast furnace slag, superplasticizer, coarse aggregate, fine aggregate, and curing age is used in this work. To find the most effective model for precise strength prediction, several machine learning regression models are put into practice and assessed. Evaluation metrics including Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and R2 are used to evaluate the models' performance. Additionally, explainable AI algorithms are used for material impact interpretation in order to assess how each input information contributes to the prediction of compressive strength. The experimental findings show that sophisticated ensemble-based regression models outperform conventional regression techniques in terms of prediction accuracy. This study provides engineers with an efficient and comprehensible AI solution for strength assessment and concrete mix design optimization.

Keywords: Concrete Compressive Strength, Artificial Intelligence, Machine Learning, Predictive Analytics, Regression Models, Explainable AI, SHAP.