A REAL-TIME AUTOMATIC INTERACTION DYNAMICS NOTATION COMMUNICATION ANALYSIS SYSTEM
Effective teamwork is essential for engineering design, but communication challenges can hinder collaboration. Interaction Dynamics Notation (IDN) provides a structured framework for analyzing team interactions, but its manual coding process is time-consuming and impractical for real-time applications. This study presents an AI-driven system that automatically assigns IDN symbols in real-time using a combination of automatic speech recognition and a large language model (LLM). The system was tested during a NASA design sprint, achieving 80.3% alignment with human-coded IDN classifications—comparable to human inter-rater reliability. Results highlight the model’s strengths in identifying structured conversational patterns and its challenges with context-dependent interactions like humor and idea blocking. The findings demonstrate the feasibility of real-time AI-driven IDN classification, paving the way for AI-facilitated team collaboration and feedback in engineering design teams.
