From Waste to Wear: A Digital Circular Innovation for Sustainable Industry

Industrial sustainability remains constrained by fragmented data ecosystems and static life-cycle models that fail to connect real-time digital intelligence with circular resource flows. Conventional Life Cycle Assessment (LCA) approaches quantify impacts retrospectively rather than guiding operational decisions. The research responds to the problem of fragmented data architectures and isolated life-cycle models that prevent continuous optimization of energy, material, and environmental performance. It aims to develop a cyber-physical decision framework capable of integrating Artificial Intelligence (AI), Internet of Things (IoT), Blockchain, and Digital Twin technologies into a unified control architecture that enables cross-sector sustainability governance and transforms sustainability indicators into controllable variables through digital-twin intelligence, integrating circular economy principles directly within production and logistics systems. The central research question guiding this work is how these digital technologies can be systematically integrated with circular-economy analytics to achieve real-time, adaptive optimization of sustainability performance across industrial symbiosis networks. Methodologically, the study adopts a hybrid, multi-layered approach beginning with the Hybrid Decision-Support Framework (HDSF), which fuses Analytic Hierarchy Process (AHP), Machine Learning (ML), and IoT-based data streams to identify, prioritize, and quantify sustainability variables. This analytical layer feeds into a Triple Life Cycle Assessment (LCA, LCC, SLCA) model applied to the coffee supply chain, forming the multi-dimensional sustainability baseline of the research. The optimization layer combines Physical Internet (PI), Multi-Agent Systems (MAS), and AI-driven predictive control for logistics and production efficiency, while a digitalization layer built on Blockchain, RFID, and mobile traceability ensures transparent and verifiable data governance. The validated system is extended through Waste Flow Mapping (WFM) and Key Performance Indicator (KPI) analytics to demonstrate industrial symbiosis between the coffee and textile sectors, culminating in a federated Digital Twin that integrates Sustainability Efficiency Index (SEI) computation and feedback-based optimization aligned with ISO 14001, ISO 50001, and UN SDGs 9, 12, and 13. Empirical validation confirms that the DigiCircular Twin-Transition (DT²) model achieves approximately 30% reduction in overall environmental footprint, 18% lower energy use, and 17% less water consumption, with resource efficiency gains of 25–30% and an estimated 38% increase in return on investment (ROI) over a decade of adoption. The results demonstrate that the DT² framework transforms sustainability from a static indicator into a dynamic control parameter within industrial operations, offering a reproducible and policy-ready mechanism for implementing the digital–green twin transition envisioned by Industry 5.0. Nonetheless, full-scale deployment depends on the availability of high-quality, interoperable datasets and cloud infrastructure for real-time monitoring. Social LCA components remain partially qualitative, requiring broader regional datasets for full automation. Simulation environments were validated in controlled industrial pilots rather than multi-plant scale. Future research should extend federated digital twins across additional sectors such as biopolymers, construction, and packaging, enhance AI-based uncertainty modeling, and embed regulatory compliance and carbon-credit monitoring into the SEI for autonomous sustainability governance.