Bonggwon Kang, PhD

Postdoctoral Scholar jointly affiliated with the Department of Industrial & Systems Engineering, University of Washington, and the Research Institute of Intelligent Logistics Big Data, Pusan National University.

E-mail: bonggwon.kang@gmail.com

I'm a postdoctoral scholar in the the Department of Industrial & Systems Engineering at the University of Washington, USA. I received my Ph.D. and B.S. in Industrial Engineering from Pusan National University in 2024 and 2019, respectively. My research focuses on uncertainty quantification for sustainable digital twins in complex manufacturing and material handling systems.

Methodological Frameworks: Discrete-event simulation, Surrogate modeling, Bayesian statistics, Gaussian process regression
Industrial Applications: Semiconductor/display fabs, container terminals, distribution centers, and automobile assembly lines

Highlight I: Simulation Modeling & Analysis

We develop simulation models that capture the underlying mechanisms of real-world operations and provide quantitative insights into complex production and logistics systems. These models serve as a foundational basis for digital twin research, supporting studies in tactical facility design, operational planning, and real-time control, where mathematical optimization and machine learning techniques are integrated for model validation and decision analysis.

Small Size FAB Simulation.mp4

3D SVSRS.mp4

반도체생산시스템.mp4

Highlight II: Simulation Optimization

In semiconductor manufacturing, efficient material handling is critical due to the complex and high-throughput nature of fabrication processes. To address the growing need for intelligent logistics management, I developed a detailed simulation model of an automated material handling system (AMHS) tailored to a semiconductor fab environment. This model was used to evaluate system performance under various configurations and operational scenarios, revealing key bottlenecks and improvement opportunities in vehicle routing and task assignment.

To enhance responsiveness and efficiency, we proposed an optimization-based control approach that enables proactive vehicle dispatching in anticipation of future transport demands. The approach was validated through extensive simulation experiments, demonstrating significant improvements in throughput and reduction in vehicle travel time. These results confirm both the effectiveness and robustness of the method in dynamic environments. This work has been published in the IEEE Transactions on Automation Science and Engineering and presented at the 2023 INFORMS Annual Meeting in Phoenix, AZ, USA.

[3D] FAB AMHS Animation 3.mp4

[3D] FAB AMHS Animation 2.mp4

Highlight III: Simulation Calibration

In modern semiconductor manufacturing, AMHSs play a critical role in transporting jobs with precision across hundreds of processing steps. Given the extreme complexity and sensitivity of these logistics environments, digital twins are increasingly used to simulate and optimize material flow, production planning, and equipment utilization. However, due to various uncertainties, the predictive accuracy of digital twins often deteriorates, leading to gaps between simulation results and actual outcomes, which in turn can cause suboptimal decision-making.

To address this challenge, we developed a Bayesian calibration framework that quantifies and corrects uncertainties within AMHS digital twins using only a small amount of real-world data. This method dramatically enhances model accuracy, particularly under high-traffic conditions where conventional simulations break down. The results, published in the Journal of Manufacturing Systems and presented at the INFORMS Annual Meeting 2024, demonstrated up to 70% improvement in predictive precision. Follow-up research is currently in progress to extend this framework to broader smart manufacturing applications.

Page updated

Google Sites

Report abuse