The 5th International Workshop on eXtended Reality for Industrial and Occupational Supports (XRIOS)

Manufacturing Physical AI: Enabling Autonomous Factories Through Virtual–Real Connected Intelligence

Abstract:

Physical AI refers to artificial intelligence systems capable of understanding physical interactions, spatial–temporal context, and the dynamics of real-world environments. Among its many potential domains, manufacturing is one of the most promising and impactful application areas.

At KAIST, we are leading a national initiative in Korea to develop autonomous factory AI that coordinates large-scale fleets of heterogeneous robots and automation systems using the principles of Physical AI. The core idea is to train AI in a high-fidelity virtual factory (digital twin), adapt and fine-tune the intelligence for a target manufacturing environment, and then seamlessly transfer this trained intelligence to real factory control systems. This virtual–real connected approach enables safe learning, rapid deployment, and significant improvements in operational performance.

In this keynote, I will introduce the concept of Physical AI for manufacturing, discuss the underlying virtual–real integration technologies, and present real industrial cases where this framework has been successfully applied in Korean factories. I will also outline future directions for XR-enabled digital twins and human–robot–environment interaction in autonomous manufacturing environments.

Bio:

Young Jae Jang received his Ph.D. in Mechanical Engineering from the Massachusetts Institute of Technology (MIT) in 2007, and dual M.S. degrees in Mechanical Engineering and Operations Research from MIT in 2001. He earned his B.S. in Aerospace Engineering from Boston University in 1997. He is currently a Professor in the Department of Industrial and Systems Engineering at the Korea Advanced Institute of Science and Technology (KAIST) and the Founding Director of the Center for KAIST Manufacturing Physical AI, where he leads research on next-generation autonomous manufacturing systems. Since 2025, Prof. Jang has served as the Principal Investigator of a USD 500 million national project on Manufacturing Physical AI, a flagship Korean government initiative aimed at realizing fully autonomous factories through AI-driven robot orchestration and digital twin technologies.