Digital Twin-Integrated Predictive Control for Intelligent Manufacturing Systems: A Simulation and AI Approach

Abstract

Digital twins have emerged as vital tools for intelligent manufacturing, offering real-time virtual replicas of physical systems. By integrating predictive control, especially model predictive control (MPC), with digital twins, manufacturing processes can be dynamically optimized for quality, efficiency, and resilience. In this paper, we propose a simulation-based framework that combines digital-twin models with MPC and AI-driven analytics for a CNC machining process. The digital twin continuously synchronizes with machine sensors and uses AI models to forecast equipment behavior. MPC uses the twin’s model to compute optimal control actions that maintain process targets and preempt defects. We demonstrate through simulation that the integrated system can reduce response latency, improve setpoint tracking, and enable proactive adjustments. For example, compared to a conventional PID controller, the MPC yields smoother control inputs and a 20% reduction in predicted quality defects. The AI components (e.g. neural networks) enhance the twin’s prediction accuracy by learning from historical data.