Generative artificial intelligence–driven sustainable supply chain management: a UNISONE framework for smart logistics and predictive analytics under Industry 5.0
Kuo-Yi Lin
Abstract
Traditional decision-making frameworks are insufficient for managing the increasingly challenging complexities, uncertainties, and sustainability requirements within global supply chains. Existing decision-making models typically rely on static optimisation or fragmented heuristics, lacking the capacity to adapt to dynamic trade-offs in environmental, operational, and economic dimensions. To address this gap, this study developed UNISONE, a generative artificial intelligence-driven decision-making model integrating multi-criteria reasoning, adaptive learning, and scenario-based simulation to support sustainable supply chain management and smart logistics within the paradigm of Industry 5.0. An empirical case study conducted on Tepla Technology Corporation, demonstrated that this framework improved delivery responsiveness, carbon efficiency, and sourcing stability. Numerical analysis confirmed that GAI, when aligned with structured goal hierarchies and feedback-driven evaluation, enhanced organisational agility and long-term sustainability orientation. This research extends theoretical knowledge on artificial-intelligence-driven supply chain models and provides a practical roadmap for implementing intelligent logistics systems in industrial settings.