AI-ENABLED ENVIRONMENTAL IMPACT ASSESSMENT FOR SUSTAINABLE FOOD PACKAGING ALTERNATIVES

As global concerns about plastic pollution and carbon emissions intensify, the food packaging industry is under increasing pressure to transition toward sustainable materials. However, assessing the environmental impact of new packaging alternatives remains a complex, time-consuming task that traditionally relies on manual life cycle assessment (LCA). This paper introduces an artificial intelligence (AI)-enabled framework designed to streamline environmental impact assessments of bio-based and biodegradable food packaging options. By integrating machine learning algorithms with LCA datasets, the proposed system rapidly evaluates packaging materials across multiple environmental indicators, including global warming potential (GWP), water usage, eutrophication, and end-of-life scenarios. The study applies supervised learning models trained on historical LCA data to predict the environmental performance of novel packaging materials, enabling faster material selection and iterative eco-design. Results show that AI models can accurately classify high-impact contributors and provide actionable insights for material optimization. This approach not only accelerates sustainability evaluations but also supports more transparent, data-driven decision-making processes in the packaging development lifecycle. The paper concludes with a discussion of challenges and future research directions, including the incorporation of real-time environmental data and model explainability.