Abstract
This research communication presents an evaluation strategy for a standardized compressive strength of concrete, identified with an ML-boosted framework model trained on prior published literature datasets to predict the influence of recycled waste. The bootstrap forest model is a specific machine-learning technique employed to capture non-linear strength trends and effects of recycled waste on the validated strength index of recycled fines concrete. This study integrates a machine learning framework with a reviewed database of recycled fines concrete for the reduction of early experimental screening. Random validation for the published datasets was performed with good significance of data reliability, and the variable region was measured through a partial dependence plot. The outcomes revealed by the framework evaluation and strength index interpretations were nonlinear sensitivity, transition, and a potential ML-boosted workflow for performance analysis and decision-making on sustainable concrete production.