A method for generating urban flood control emergency plans based on large language models and graph retrieval enhancement

ABSTRACT Urban flood control emergency plans are crucial for addressing flood disasters and are essential for safeguarding lives and property as well as ensuring high-quality urban development. Existing generation methods often rely on shallow retrieval, which can lead to fragmented knowledge; moreover, the generation of long texts lacks structural constraints, frequently resulting in the omission of key elements. To address this, this study proposes a framework for generating urban flood control emergency plans that integrates LLMs with graph retrieval. First, a flood control knowledge graph is constructed based on a domain ontology, providing structured support. Second, we use LLMs to generate candidate questions through backward reasoning and introduce chapter prompts to standardize the output content. Finally, we design a closed-loop process of ‘Scene Analysis – Graph Search – Plan Generation – Quality Assessment – Iterative Optimization’, leveraging the knowledge graph to address the shortcomings of traditional retrieval in complex scenarios. Experiments show that, compared to LLMs and traditional RAG methods, the dual constraint mechanism based on ‘graph retrieval’ and ‘section prompts’ significantly improves the completeness and feasibility of the plans. This study effectively addresses issues such as knowledge fragmentation, content omissions, and structural deficiencies, providing a reliable method for the standardized development of urban flood control emergency plans.