본문 바로가기
  • Home

A Study on a Physical AI Application Model for Extending Manufacturing AX

  • Journal of Internet of Things and Convergence
  • Abbr : JKIOTS
  • 2026, 12(4), 7
  • Publisher : The Korea Internet of Things Society
  • Research Area : Engineering > Computer Science > Internet Information Processing
  • Received : August 4, 2026
  • Accepted : August 21, 2026
  • Published : August 31, 2026

Keun-Ho Lee 1

1백석대학교

Accredited

ABSTRACT

A prior study proposed a three-phase manufacturing AX education model centered on Internet of Things (IoT) and artificial intelligence (AI) data analysis, based on a demand survey of 23 manufacturing companies in the Cheonan region. However, that model focused on data collection, analysis, and strategy formulation, and did not extend to the stage of physically executing tasks through robots and automation equipment on the actual manufacturing floor. With recent advances in Vision-Language-Action (VLA) models and foundation robot policies, "Physical AI" — artificial intelligence that perceives, reasons, and acts directly in the physical world rather than merely making judgments on a screen — has emerged as a new axis of manufacturing automation. This study extends the existing IoT-based manufacturing AX education model with Physical AI elements, proposing a four-layer technology stack (Perception-Cognition-Action-Feedback) and a corresponding four-phase education model. By adding physical execution capability and safety/governance indicators to the existing model, the proposed extended model provides a foundation for a sustainable manufacturing AX talent development system that encompasses not only data-driven transformation (DX) but also actual automation execution capability for regional manufacturing industries.

Journal Copyright Policy

No CCL information provided

Citation status

* References for papers published after 2025 are currently being built.