A Vision-Driven Architecture for Intelligent Proactive Decision-Making in Industry 5.0 Gigafactories
Year: 2026
Type: conference-paper
Abstract: Intelligent manufacturing systems that integrate sophisticated automation with human-centered decision-making, resilience, and sustainability are being developed as a result of the shift to Industry 5.0. The growing complexity of production processes in battery gigafactories necessitates integrated systems that can convert heterogeneous data into timely and useful operational insights. A vision-driven architecture for proactive decision making in Industry 5.0 gigafactories is proposed in this study. Four interrelated layers visual perception, intelligent data fusion, proactive decision making and human in the loop interaction are integrated into the suggested framework. The architecture allows for continuous monitoring anomaly detection and proactive intervention while preserving human supervision in crucial choices by integrating computer vision, multimodal sensor data, predictive analytics, and operator feedback. A YOLO26-based battery identification model was developed and tested on a unique multi-class battery dataset to determine the viability of the perception layer. The results show that the perception module can deliver trustworthy visual data for further processing. The suggested architecture helps put Industry 5.0 concepts into practice in next-generation gigafactory settings by providing a scalable and human-centric basis for intelligent battery management.
Language: English
Subfield: Industrial and Manufacturing Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W7204073873