Generative AI for Automatic Simulation Model Generation in Factory Planning: A Framework and Prototype

Manual development of simulation models in factory planning is a time-intensive task that requires domain expertise and specialized software knowledge. With recent advancements in generative AI, particularly large language models (LLMs), there is growing interest in automating these processes to support more agile and accessible planning. This paper explores the feasibility of using LLMs to automate the generation and optimization of discrete-event simulation (DES) models from natural language descriptions. We present a framework for automatic simulation model generation (ASMG) that converts user-provided layout descriptions into structured data, validates the output, and instantiates the model in Siemens Plant Simulation via a Python-based interpreter. An iterative feedback loop enables performance-driven optimization using LLMs. A proof-of-concept demonstrates the framework's ability to generate simulation models and refine layouts based on throughput and bottleneck analysis. The framework shows strong potential for reducing manual effort and improving agility in digital factory planning.