This project, developed by the Postgraduate Program in Civil Engineering (PPGEC) and Computing Department (DECOM) at the Federal Center for Technological Education of Minas Gerais (CEFET-MG), aims to enhance the prediction of the durability of concrete structures, focusing on deterioration processes such as carbonation, through the application of machine learning techniques. The research combines artificial intelligence methods and the development of databases based on mathematical models to analyze the impact of material and environmental variables on structural degradation.

By employing algorithms such as Artificial Neural Networks, Random Forest, and Support Vector Regression, the project investigates methods to identify complex patterns and more accurately predict the progression of phenomena that affect structural integrity. Explainability tools, such as SHAP (SHapley Additive Explanations), are incorporated into the study to provide transparency and reliability to the results.

With this approach, the project aims to promote sustainable and efficient solutions in the planning, constructing, and maintaining concrete structures, supporting more informed and reliable decision-making in civil engineering.