ARTIFICIAL INTELLIGENCE-DRIVEN PREDICTIVE ANALYTICS FRAMEWORK FOR SUSTAINABLE GEOPOLYMER CONCRETE USING AGRICULTURAL WASTE MATERIALS

The rapid growth of the construction industry has significantly increased the global demand for conventional concrete materials, resulting in substantial environmental concerns due to excessive cement production and industrial carbon emissions. Traditional Portland cement manufacturing contributes heavily to greenhouse gas emissions and environmental degradation, motivating researchers to investigate sustainable alternatives capable of reducing the environmental impact of infrastructure development. Geopolymer concrete has emerged as a promising sustainable construction material because it allows the incorporation of agricultural and industrial waste materials while maintaining desirable structural and durability characteristics. This research presents an Artificial Intelligence driven predictive analytics framework for sustainable geopolymer concrete utilizing agricultural waste materials including Sugarcane Bagasse Ash (SBA), Banana Peel Ash (BPA), and Fly Ash Type C polymer. The developed system integrates Random Forest Regression models with predictive analytics pipelines that estimate initial setting time,final setting time, compressive strength, and flexural strength across multiple geopolymer compositions. The framework was implemented using Python, Scikit-learn, FastAPI, SQLite databases, and locally hosted predictive services. Experimental evaluation demonstrated approximately 75% predictive accuracy despite limited dataset availability. The proposed framework significantly reduces the time and cost associated with traditional laboratory experimentation while supporting sustainable material optimization and environmentally responsible infrastructure research.