(535f) Break

Industry 4.0 has transformed manufacturing, particularly through additive manufacturing (AM) like Fused Deposition Modeling (FDM), which enables cost-effective, customized production with minimal waste. However, FDM's error susceptibility often requires expert oversight to detect and correct defects that compromise quality. Existing automated solutions lack adaptability across different printers, and deep learning models demand large labeled datasets, limiting scalability. To overcome this, we propose a process monitoring framework using pre-trained Large Language Models (LLMs) to detect and address defects in real-time. Validations show that LLM-based agents can autonomously correct common 3D printing issues, matching the performance of experienced engineers.