Circular Economy Transitions in Textile, Apparel, and Fashion: AI-Based Topic Modeling and Sustainable Development Goals Mapping

This study focuses on the shift to a circular economy (CE) in the textile, apparel, and fashion (TAF) sectors, which generate tons of waste annually. Thus, embracing CE practices is essential for contributing to UN Sustainable Development Goals. This study employs a mixed-methods approach, integrating PRISMA for systematic literature selection, BERTopic modeling and AI-driven SDG mapping, and case study analysis to explore emerging CE themes, implemented circular practices, and systemic barriers. Machine-learning-based SDG mapping reveals strong linkages to SDG 9 and SDG 12, emphasizing technological advancements, industrial collaborations, and circular business models. Moderately explored SDGs, namely, SDG 8, SDG 6, and SDG 7, highlight research on labor conditions, water conservation, and clean energy integration. Reviewing 655 peer-reviewed papers, the BERTopic modeling extracted six key themes, including sustainable recycling, circular business models, and consumer engagement, whereas case studies highlighted regulatory frameworks, stakeholder collaboration, and financial incentives as critical enablers. The findings advance institutional theory by demonstrating how certifications, material standards, and regulations drive CE adoption, reinforcing SDG 12 and SDG 16. The natural resource-based view is extended by showing that technological resources alone are insufficiently aligned with SDG 9. Using the Antecedents–Decisions–Outcomes framework, this study presents a structured, AI-driven roadmap for scaling CE in the TAF industry, addressing systemic barriers, and supporting global sustainability goals, highlighting how multistakeholder collaboration, digital traceability, and inclusive governance can improve the impact of CE. The results recommend that CE strategies should be aligned with net-zero targets, carbon credit systems, and block-chain-based supply chains.