"ECOCHAIN: HARNESSING DATA SCIENCE TO MINIMIZE CARBON FOOTPRINT IN SUPPLY CHAINS" This study introduces a sophisticated data-driven system tailored for optimizing carbon emissions within industrial operations, with a specific emphasis on supply chain management. Leveraging advanced machine learning methodologies— namely decision trees and random forests—the system meticulously evaluates CO2 equivalent (CO2e) contributions originating from various production components. Key functionalities encompass meticulous data pre-processing to calculate precise CO2e values, robust assessment of feature importance to pinpoint pivotal contributors, and the formulation of actionable recommendations. The adopted machine learning algorithms offer distinct advantages: decision trees excel in elucidating complex relationships among production variables, while random forests enhance predictive accuracy by aggregating multiple decision trees. These attributes enable the system to deliver nuanced insights into carbon emissions dynamics within supply chains. Recommendations stemming from the analysis advocate for strategic measures, thereby empowering organizations in the supply chain domain, this system aims to foster substantial reductions in environmental impact while bolstering operational resilience and adherence to sustainability goals