Wenbo Sun

Research Interests

Uncertainty quantification and decision making are increasingly demanded with the development of future technology in engineering and transportation systems. Among the uncertainty quantification problems, I am particularly interested in statistical modelling of engineering system responses with considering the high dimensionality and complicated correlation structure, as well as quantifying the uncertainty from a variety of sources simultaneously, such as the inexactness of large-scale computer experiments, process variations, and measurement noises. I am also interested in data-driven decision making that is robust to the uncertainty. Specifically, I deliver methodologies for anomaly detection and system design optimization, which can be applied to manufacturing process monitoring, distracted driving detection, out-of-distribution object identification, vehicle safety design optimization, etc.

Biography

Wenbo Sun is a research faculty at the University of Michigan Transportation Research Institute (UMTRI) and an affiliated faculty at the Michigan Institute for Data Science (MIDAS). 

He earned his Ph.D. in the Department of Industrial and Operations Engineering (IOE) at the University of Michigan, advised by Professor Judy Jin and Matthew Plumlee. He is interested in methodological research that utilizes artificial intelligence for uncertainty quantification and decision making in engineering applications. He is also interested in collaborating with domain experts to solve data-driven problems. 

Dr. Sun’s dissertation research focused on the methodologies for uncertainty quantification of functional responses, which have been applied to evaluate crash injury risk and improve vehicle design. During postdoctoral research, he has expanded this data analytics research to other application areas through joint projects with different units at the University of Michigan, including UMTRI, the Departments of Mechanical Engineering and Radiation Oncology. He has also worked on collaborative research projects with industry companies, including Honda R&D Americas, FCA, and Samsung. In these projects, he has developed statistical methods that are tailored to specific application problems. He has been lead author on eight journal articles based on this work, where his contributions included problem formulation, statistical modeling, mathematical derivation, coding solutions, and documentation. During his postdoctoral research, he has also developed and submitted four research proposals together with his advisor, Prof. Judy Jin, and other collaborators. One proposal was funded by the UM-Ford Alliance Program and additional proposals will be submitted to NSF. His research activities show an unusually high level of successful collaboration for a junior researcher. Appropriate to UMTRI’s mission, his work has focused on developing methods that have real-world impacts, based on a good understanding of the domain knowledge and the specific analytical needs. His independent work on statistical methodology development has enabled successes in diverse fields. For example, he worked in collaboration with the Department of Radiation Oncology on a precision health project that involved radiation dose determination and delivery. He designed a machine learning algorithm to adaptively adjust the radiation doses based on patients personal information. The work was demonstrated to have real-world impacts and led to a journal paper and a conference presentation.  

Area of Expertise

• Methodology: out-of-distribution learning, Gaussian process modeling, generative models, anomaly detection
• Applications:  computer vision, vehicle safety, assistant driving system, smart manufacturing, health care

Awards

• QCRE Best Student Paper Award in IISE Annual Meeting 2018
• Best Theoretical Paper Award in INFORMS Workshop on Data Mining and Decision Analysis 2021
• DAIS Best Track Paper Award in IISE Annual Conference 2022
• Student Paper Competition Winner at the joint SPES + Q&P session in JSM 2022

Service

Organizer and Chair, Deep neural networks for quality assurance in manufacturing
QSR sponsored session, INFORMS Annual Meeting October 2023

Organizer, Data-driven Decision Making in Transportation 
JSM topic-contributed session, Joint Statistical Meeting August 2023
Organizer and Chair, Machine learning for quality assurance and decision-making 
QSR sponsored session, INFORMS Annual Meeting October 2022
Organizer and Chair, Learning from High-Dimensional Data: Improvement in Efficiency and Interpretability 
DAIS sponsored session, IISE Annual Meeting May 2022
Organizer and Chair,  Deep Learning for Quality Assurance in Manufacturing Systems
QSR sponsored session, INFORMS Annual Meeting November 2021
Organizer and Chair,  Computer Experiments and Decision Making
QSR sponsored session, INFORMS Annual Meeting November 2020
Organizer and Chair,  Data-driven Design and Computer Simulations 
QSR sponsored session, INFORMS Annual Meeting November 2019

Additional Information

Hometown, Tianjin, China

Hobbies: badminton, baking, photography

Website