[Keynote] Advancing Autonomous Fabs: PM Automation & Standardization Strategy

10:40 am - 11:15 am

The semiconductor manufacturing environment is rapidly evolving toward technology scaling, highly sophisticated equipment, and unprecedented capital intensity. Against this backdrop, preventive maintenance (PM) automation is no longer optional but a fundamental requirement for realizing autonomous fabs. While industry stakeholders unanimously recognize the necessity of PM automation, implementation remains confined to isolated pilot projects. This limitation does not stem from a lack of technology, but from the absence of ecosystem-wide standardization.

Current disparities in equipment architecture, data structures, robotic interfaces, and component and logistics specifications hinder large-scale adoption by increasing redundancy, integration risks, and compatibility concerns across fabs. Survey findings indicate that the industry anticipates an average of 8.23 years to achieve full PM automation, identifying “equipment not designed for automation” (87.3%) and the “resource burden caused by non-standardization” as major obstacles.

The ultimate direction of PM automation lies in integrating predictive maintenance (PdM) capabilities to build intelligent maintenance systems that go beyond robotic task replacement. Real-time data, combined with AI-based analytics, enables fault prediction and optimized maintenance timing. However, non-uniform data formats (such as heterogeneous SVID and log structures) and limited data accessibility significantly impede PdM model reliability.

To address these structural challenges, coordinated, cross-industry standardization efforts led by organizations such as the “SEMI Autonomous Fab Working Group” (Afab WG) and COSAR are essential. This presentation highlights that PM automation represents a holistic systems challenge spanning equipment, data, robotics, and logistics domains. It emphasizes the urgent need for an integrated standardization roadmap to secure interoperability across the semiconductor ecosystem and accelerate the transition toward autonomous mass-production fabs.

Featured Speakers

Youbean Kim (invited)

Assistant Professor, Myongji University

Youbean Kim received the B.S. and M.S. degrees in computer science from Sogang University, Seoul, South Korea, in 2002 and 2004, respectively, the Ph.D. degree in electrical and electronic engineering from Yonsei University, Seoul, in 2009, and the Ph.D. degree in public administration (specialized in Science & Technology Policy) from Hanyang University, Seoul, in 2014.

He was a Senior Engineer with Samsung Electronics and Samsung Display, a Senior Researcher with the Korea Institute of Fusion Energy (KFE), and a research fellow with the National Assembly Futures Institute (NAFI). Since 2022, he has been an Assistant Professor in the Division of Semiconductor Engineering, Myongji University, Yongin-si. His current research interests include predictive maintenance (PdM), semiconductor testing, design for testability (DFT), and high-speed testing, including signal integrity and power integrity.