Kyonggi University’s Department of Industrial and Management Engineering is pleased to highlight a recent research achievement by Professor Haejoong Kim, whose study titled “Bayesian Optimization for the Vehicle Dwelling Policy in a Semiconductor Wafer Fab” presents a forward-looking approach to smart manufacturing and semiconductor logistics optimization.
In modern semiconductor wafer fabs, hundreds of automated transport vehicles continuously move wafers between process tools and stockers. A key yet often overlooked factor influencing productivity is the vehicle dwelling policy—where a vehicle should wait when it becomes idle. Traditional heuristics often fail to capture the complex dynamics of real fab operations, leading to unnecessary travel, congestion, and throughput loss.
Professor Kim’s research introduces a significant advancement by applying Bayesian Optimization to automatically determine the most efficient dwelling parameters. Instead of relying on fixed rules, his method uses simulation-based evaluations combined with machine-learning-driven optimization to explore a wide range of possible policies in a highly efficient manner. The study also utilizes Siemens Plant Simulation to construct high-fidelity wafer-fab logistics models, enabling precise evaluation of vehicle behavior under different dwelling strategies.

B. Kang, C. Park, H. Kim, S. Hong. "Bayesian Optimization for the Vehicle Dwelling Policy in a Semiconductor Wafer Fab." IEEE Transactions on Automation Science and Engineering, 12(4), 2024. doi: 10.1109/TASE.2023.3320189