This research bridges the gap between technical engineering diagrams and natural language by integrating P&IDs with Large Language Models for intuitive interactions.

• Converts P&IDs into graph representations using the DEXPI data model
• Creates knowledge graphs from diagrams using a custom Python package
• Enables natural language queries to extract information from complex technical diagrams
• Provides a more accessible way to work with industrial schematics

This innovation streamlines engineering processes by allowing non-specialists to query and understand complex industrial diagrams, potentially improving safety, efficiency, and knowledge transfer in manufacturing environments.

Talking like Piping and Instrumentation Diagrams (P&IDs)