Abstract: A knowledge graph generation system includes an input unit and a knowledge graph generator. The input unit receives service reports, each comprising natural language; and provides the service reports to the knowledge graph generator. The knowledge graph generator implements a Large Language Model (LLM), utilizes the LLM to identify entries from the natural language in the service reports, wherein the identified entries comprise names and types of instruments, issues that need to be addressed for those instruments, and actions taken by service technicians to mitigate the issues. The identified entries form rules, each rule comprises a tuple of one instrument type, one issue, and an action step or series of action steps taken to mitigate the issue. The rules are used to generate a plurality of knowledge graph nodes. The knowledge graph generator utilizes the nodes to generate a knowledge graph.