Integrated process-system modelling and control through graph neural network and reinforcement learning

Modern manufacturing systems are becoming increasingly complex, dynamic, and connected, their performance is being affected by not only constituent processes but also system-level interactions. This paper presents an integrated modelling method based on a graph neural network (GNN) multi-agent reinforcement learning (MARL) collaborative control for adjusting individual machining process parameters in response to system- process-level conditions. The structural operational dependencies among machines captured with GNN. Iteratively trained MARL, learn adaptively local parameters, e.g., speed depth of cut, while achieving the global goal improving production yield.