Integrating Large Language Model Agents with Digital Twins for Industrial Autonomous Systems

The article presents a dissertation proposing a three-layer framework that integrates large language models (LLMs) with digital twins and automation systems to enhance industrial autonomy. The framework introduces the Task-Process-Service-Resource (TPSR) model, enabling adaptive, goal-oriented behavior through LLM-based reasoning across dynamic environments. Key findings from case studies indicate improved task executability and command accuracy, highlighting the potential for LLM integration to advance industrial automation, although challenges remain regarding digital representation accuracy and computational demands.