AI-INTEGRATED DIGITAL TWIN FRAMEWORK FOR PREDICTIVE MAINTENANCE AND OPERATIONAL OPTIMIZATION IN INDUSTRIAL CYBER-PHYSICAL SYSTEM HYBRID EDGE-CLOUD ARCHITECTURE LEVERAGING SELF-SUPERVISED LEARNING AND REAL-TIME SENSOR FUSION FOR RESILIENT SMART MANUFA

In modern industrial environments, the convergence of AI, edge-cloud computing, and digital twins offers unprecedented opportunities for real-time predictive maintenance and operational efficiency. This paper proposes a hybrid edge-cloud digital twin framework, utilizing self-supervised learning and sensor fusion to enhance resiliency in cyber-physical manufacturing systems. By deploying real-time analytics at the edge while maintaining model coordination and historical data storage in the cloud, the proposed architecture balances latency sensitivity and computational scalability. This integrated approach aims to minimize downtime, optimize operations, and foster sustainable smart manufacturing practices.