MAE Distinguished Speaker Series 2026 Seminar on Pursuing Digital Twins in Semiconductor Manufacturing through AI/ML-based Advancements in Metrology and Process Control

03 Nov 2026 10.00 AM - 11.00 AM LT4 (North Spine, NS4-02-34) Current Students, Public

Professor Dragan Djurdjanovic

Accenture Professor of Manufacturing Systems Engineering

University of Texas at Austin, United States

This seminar will be chaired by Asst Prof Marvin Carl May.

Seminar Abstract

A crucial missing element for enabling Digital Twin (DT) functionalities in semiconductor manufacturing is insufficient spatial and temporal density of wafer metrology. This talk will present opportunities to increase the resolution, precision and speed of semiconductor manufacturing metrology by merging first-principles physics and fundamental mathematics with Machine Learning (ML) and Artificial Intelligence (AI). The talk will discuss AI/ML-based inference of Critical Dimensions (CDs) from scatterometry of large-scale nano-pattern arrays, enabling fast and accurate in-process metrology. AI/ML-based surrogate models facilitate metrology of an entire 300mm wafer within minutes, with Gage Repeatability and Reproducibility (Gage R&R) ratios comparable to Scanning Electron Microscopy. Recent research on Virtual Metrology (VM) will also be presented, using tool sensor readings to estimate electrical and geometric quality characteristics of wafer features. Advances in VM-based process control using statistically tractable and physically interpretable ML concepts represent a significant step towards enabling DT functionalities. Finally, robust optimization algorithms for lithography overlay control will be presented, addressing non-Gaussian noise and spatial variability at sub-nanometer scales. Simulations from a modern 300mm wafer fab show that the novel algorithm could reduce overlay errors by more than 40%.

Speaker's Biography 

Dragan Djurdjanovic obtained his B.S. in Mechanical Engineering and B.S. in Applied Mathematics from the University of Nis, Serbia, in 1997, his M.S. in Mechanical Engineering from Nanyang Technological University, Singapore, in 1999, and his M.S. in Electrical Engineering and Ph.D. in Mechanical Engineering from the University of Michigan, Ann Arbor, in 2002. His research focuses on methodologies for transforming data into useful information and operational decisions, with applications primarily in semiconductor manufacturing. From 2013 to 2023, he held leadership roles at the NSF Industry-University Cooperative Research Center on Intelligent Maintenance Systems and the NSF Engineering Research Center on Nanomanufacturing Systems for Mobile Computing and Mobile Energy Technologies at the University of Texas at Austin. He has co-authored 95 archival journal and more than 50 refereed conference publications, and serves as an Associate Editor for IEEE Transactions on Semiconductor Manufacturing and CIRP Novel Topics in Production Engineering. He is a Fellow of ISEAM and Associate Member of CIRP, and has received several recognitions, including the 2018 August-Wilhelm Scheer Visiting Professorship and the 2006 Outstanding Young Manufacturing Engineer Award from the Society of Manufacturing Engineers (SME).

Categories