ProcessLLM-R1: A hypergraph retrieval-augmented and reinforcement learning-driven assembly process planning for commercial aircraft

As a key stage in the intelligent manufacturing of commercial aircraft, process planning plays an important role in improving production efficiency and ensuring product quality. Traditional process planning relies heavily on manual review of scattered, frequently updated technical documents, which is not only time-consuming and labor-intensive but also prone to errors due to limited experience. With the development of large language models (LLMs), retrieval-augmented generation (RAG)-based intelligent process planning methods have achieved significant progress. However, existing RAG-based process planning generation methods typically rely on static one-shot querying and passive filtering. As a result, LLMs lack the ability to autonomously assess knowledge sufficiency, which limits their effectiveness in handling fragmented process knowledge and complex interrelationships. To this end, this paper proposes ProcessLLM-R1, an intelligent method for commercial aircraft assembly process generation based on hypergraph retrieval and reinforcement learning. First, a process knowledge hypergraph is constructed to organize process knowledge, and hypergraph retrieval strategy is designed that can be autonomously invoked by LLM, thereby enhancing the models’ ability to dynamically acquire and organize process knowledge. Second, retrieval is modeled as a multi-round interaction between the agent and knowledge hypergraph, and GRPO is introduced to perform parameter-efficient fine-tuning of the agent, thereby improving its adaptability to process generation tasks. Finally, experimental results on the commercial aircraft machining and assembly process datasets show that ProcessLLM-R1 outperforms the evaluated LLM-based generation and retrieval-augmented baselines across the reported metrics. It is hoped that the research can provide valuable insights and support for the application of LLM-based solutions in the field of aircraft manufacturing.