Python-Based AI-Assisted Modeling and Computation of Life Cycle Assessment of European Polymeric Waste: Application in Manufacturing and Recycling Industries Regarding Sustainability

Development of sustainability systems for assessment of environmental impacts remains a paramount challenge for green and circular manufacturing of polymers. In this study, a comprehensive life cycle assessment (LCA) framework is developed for European polymeric waste by integrating OpenLCA, Ecoinvent v3.11, and Python-based machine learning (ML) algorithms. Cradle-to-gate, service-life, and cradle-to-grave assessments are performed for representative thermoplastic composite systems, including PP–PET–cotton, HDPE–glass fiber, and PEEK–carbon fiber composites, covering domestic, engineering, and high-performance polymer categories. The results demonstrate that raw material extraction and manufacturing stages dominate environmental impacts, contributing the highest shares to climate change, ecotoxicity, and non-renewable energy consumption. PP-based composite systems exhibit the lowest overall environmental burdens due to lower processing energy and simpler molecular structures, while HDPE-based systems show moderate impacts. PEEK-based composites present the highest impacts per unit mass, driven by energy-intensive synthesis and high processing temperature. Environmental impacts are evaluated using EF v3.1 and ReCiPe methodologies, supported by Monte Carlo simulations and ML-assisted uncertainty quantification. Monte Carlo simulations and ML-assisted LCA provide probabilistic ranges, uncertainty quantification, and predictive insights into impact indicators, enabling the development of a quantitative sustainability system based on probability–impact relationships. A Europe-wide assessment of 57 Mt of polymeric waste highlights that environmental burdens are concentrated in countries with high polymer production and consumption, emphasizing the importance of energy mix, recycling efficiency, and waste management strategies. Overall, this work demonstrates that digitalized LCA coupled with ML offers a powerful decision-support framework for sustainable polymer design, recycling optimization, and circular economy policy development, supporting the transition toward low-carbon and resource-efficient polymer systems in Europe.