Physics-Grounded Multi-Agent Architecture for Traceable, Risk-Aware Human-AI Decision Support in Manufacturing

Danny Hoang, Ryan Matthiessen, Christopher Miller, Nasir Mannan, Ruby ElKharboutly et al.|May 5, 2026arXiv

Key Takeaway

For safety-critical applications, decompose AI workflows into specialized agents (routing, analysis, retrieval, verification) rather than relying on a single LLM, and enforce physical plausibility constraints before surfacing recommendations to humans.

Summary

A multi-agent system that helps humans make safer decisions in precision manufacturing by combining AI reasoning with physics simulations, inspection data, and verification checks.

agents safety applications

Key Terms

multi-agent-framework knowledge-graph human-in-the-loop digital-twin provenance