An ANP-Based Decision Framework for ESG-Driven Green Supply Chain Management with Proposed Neural Feature Extraction

This study develops an integrated decision-support framework to advance green supply chain management (GSCM) by systematically linking Environmental, Social, and Governance (ESG) practices, environmental product innovation, corporate performance, and strategic alternatives. Employing the Analytic Network Process (ANP), the proposed model captures complex interdependencies and feedback relationships across life-cycle value chain stages, enabling a holistic evaluation of sustainability-oriented strategies. A Delphi panel comprising 15 experts from academia, industry, and government is used to validate the evaluation criteria and network structure. The empirical results indicate that eco-friendly design, energy and resource efficiency, and carbon–climate management are the most influential drivers shaping green supply chain performance. Moreover, operational and sustainability performance are found to exert greater strategic importance than short-term financial performance, highlighting GSCM as a long-term capability-building approach rather than a cost-centered initiative. To enhance analytical adaptability, this study proposes a conceptual extension integrating neural feature extraction (NFE) signals with ANP-based expert weights. The NFE module is not empirically trained or validated; rather, it illustrates a theoretically consistent mechanism for incorporating data-driven feature signals into structured multi-criteria decision frameworks. Empirical validation of the NFE component is proposed as a future research direction.