Predicting the Compressive Strength of Recycled Concrete Using Ensemble Learning Model

The increasing demand for sustainable construction materials has led to significant research on the utilization of recycled concrete aggregates (RCA) in concrete production. This study presents a web-based system designed to predict the compressive strength of recycled concrete using an ensemble learning approach. The system leverages various machine learning models, including CatBoost, XGBoost, Random Forest, and Gradient Boosting Regressor, to predict the compressive strength based on the input features of concrete mix design, including the quantity of materials such as cement, blast furnace slag, fly ash, water, superplasticizer, coarse aggregate, fine aggregate, and the age of the concrete. The system is built using Flask, a lightweight Python framework, with a front-end interface that allows users to input concrete mix details and receive the predicted compressive strength. The ensemble models are evaluated based on key performance metrics such as MAE, MSE, RMSE, and R², providing insights into their accuracy and robustness. The prediction tool facilitates the efficient design and optimization of recycled concrete mixes, promoting the adoption of sustainable construction practices. This approach provides a practical solution to minimize the environmental impact of construction while ensuring the structural integrity of concrete structures.

Keywords:

Recycled Concrete, Compressive Strength, Ensemble Learning, CatBoost, XGBoost, Random Forest, Flask, Machine Learning, Concrete Mix Design, Sustainable Construction.