AI-ENABLED DIGITAL TWINS FOR PREDICTIVE MAINTENANCE IN INDUSTRY 5.0 MANUFACTURING

Abstract

Industry 5.0 manufacturing systems require maintenance strategies that are not only efficient and intelligent but also human-centric, resilient, and sustainable. This paper examines the role of AI-enabled digital twin predictive maintenance in improving asset reliability, reducing operational downtime, and supporting autonomous decision-making in advanced manufacturing environments. By integrating real-time sensor monitoring, machine learning-based fault prediction, remaining useful life estimation, and virtual asset simulation, the proposed digital twin framework enables continuous assessment of equipment health and proactive maintenance planning. The results indicate that AI-enabled digital twins can significantly improve fault detection accuracy, maintenance response time, energy efficiency, and cost optimization compared with conventional preventive and reactive maintenance approaches. The framework also supports Industry 5.0 priorities by strengthening collaboration between human operators and intelligent systems, enhancing explainability in maintenance decisions, and minimizing unnecessary machine stoppages. Overall, the study demonstrates that predictive maintenance powered by AI and digital twin technology can serve as a critical foundation for smart, adaptive, and sustainable manufacturing systems.