Optimize manufacturing operations with digital twin and deep Q‐network

Abstract The display manufacturing industry is exploring operational optimization strategies using digital twins to address the challenges of complex production lines and rapidly changing market demands. In this study, we explored an optimal solution to release equipment constraints in manufacturing lines using digital twin technology and deep reinforcement learning. As a result, the number of released equipment constraints was reduced by 5% compared to the conventional approach, and production volume increased by 5.6%. To facilitate the interaction between digital twins containing on‐site reference information and deep reinforcement learning, the study adhered to the international ISO23247 standard. This research marks the world's first attempt at a digital transformation study focused on manufacturing line operations using this approach.