Artificial Intelligence Applications in Sustainable Reverse Supply Chains: A Systematic Review, Gap Analysis, and Future Research Agenda

Abstract

Artificial intelligence (AI) is increasingly reshaping sustainable reverse supply chains (SRSCs) by enhancing product return management, remanufacturing, recycling, recovery, and reintegration processes, while strengthening circular economy implementation and sustainability performance. Escalating regulatory pressures, resource scarcity, climate-related disruptions, and rising reverse logistics complexity have intensified the demand for intelligent, adaptive, and data-driven systems capable of operating effectively under uncertainty. However, despite growing scholarly attention, research on AI-enabled SRSCs remains fragmented across technological, operational, and managerial domains, limiting theoretical integration and a comprehensive understanding of system transformation. This study presents a systematic literature review of peer-reviewed publications from 2000 to early 2026 to synthesize and critically evaluate the evolution of AI applications in SRSCs. Using a structured review protocol and thematic synthesis, it develops a multidimensional taxonomy and a mechanism-based conceptual framework explaining how AI enhances decision intelligence, coordination efficiency, operational resilience, and sustainability performance across reverse supply chain systems. Findings reveal a structural shift from cost-centric reverse logistics toward predictive, adaptive, and sustainability-oriented closed-loop systems characterized by real-time visibility, dynamic optimization, and autonomous decision support. AI significantly improves return flow forecasting accuracy, recovery efficiency, waste reduction, and resilience under uncertainty and disruption. However, the literature remains unevenly developed, particularly in relation to governance structures, cross-functional integration, and methodological standardization. Key research gaps are identified in AI governance, explainability, interoperability, scalability, and social sustainability integration. Importantly, generative AI and large language models (LLMs) emerge as a nascent but largely underexplored frontier with strong potential to transform knowledge-intensive decision-making, coordination mechanisms, and adaptive control in reverse supply chains. In response, the study proposes a future research agenda across six interrelated dimensions to advance intelligent, resilient, and sustainability-oriented closed-loop supply chain systems.