With the rapid development of intelligent manufacturing systems, how to achieve low-latency decision-making in complex and ever-changing manufacturing environments has become a key issue. This study proposes a solution based on edge artificial intelligence. By moving computing tasks to edge nodes close to the data source, it significantly reduces data transmission latency and improves the system’s real-time decision-making capabilities. This paper designs an edge computing architecture and optimizes artificial intelligence algorithms to achieve efficient inference in resource-constrained environments. Experimental results show that the proposed framework can effectively reduce decision-making latency in intelligent manufacturing systems while maintaining high decision-making accuracy, providing technical support for the real-time and stability of intelligent manufacturing systems. This paper also explores the potential of collaborative computing at the edge and in the cloud, and proposes future research directions to further improve the application efficiency of edge AI in intelligent manufacturing.