ABSTRACT
The integration of fly ash into concrete mixtures has been a topic of significant interest in the construction industry owing to its potential to enhance both the mechanical properties and sustainability of concrete structures. The study delves into evaluating the effect of fly ash on concrete mechanical properties through ANN-based predictive modeling. Fly ash's utility as a supplementary cementitious material is crucial for optimizing concrete mix designs, and aligning with sustainable construction goals. The methodology employed in this study involves a systematic approach to data collection, modeling, and evaluation. A total of 1030 data points encompassing various concrete mix parameters, including cement content, fine and coarse aggregate content, water content, superplasticizer, and fly ash specifics, were collected and analyzed. The ANN models were trained, tested, and validated using 70% of the data for training and the remaining 30% for testing and validation. The evaluation process included the calculation of key metrics such as Mean Squared Error (MSE), Mean Absolute Error (MAE), and R-squared score (R2 ) to assess the model's accuracy and reliability in predicting the compressive strength of concrete with varying fly ash content. The study's findings reveal a comprehensive understanding of the influence of fly ash on concrete mechanical properties. The training, validation, and test performance metrics of Mean squared error (MSE), Mean Absolute Error (MAE), and R-squared score (R2 ) yielded favorable values. A lower MSE value of 0.0016 (training), 0.0038 (validation), and 0.0048 (test) indicates better accuracy because it reduces the disparity between the actual and predicted values. The MAE values of 0.0297 (training), 0.0459 (validation), and 0.0509 (test) also indicate better accuracy due to their low values. Higher values of R2 , 0.9644 (training), iv 0.9126 (validation), and 0.8988 (test) closer to 1 (a perfect match) indicate how well the predicted values match the actual values. The models demonstrated robust performance in predicting the compressive strength of concrete with fly ash, showcasing their potential for practical applications in sustainable construction.
