After receiving his bachelor’s degree in mechanical engineering at Zhejiang University, Dr. Haiyue Wu joined the Laboratory of Sustainable Manufacturing (LSM) for his Ph.D. at Purdue under the supervision of Dr. John W. Sutherland. He worked as a research assistant in the Wabash Heartland Innovation Network (WHIN) project for next generation manufacturing. He led efforts to advance next-generation manufacturing with a particular emphasis on the implementation of internet of things (IoTs) and artificial intelligence (AI) within manufacturing equipment. In particular, he has focused on manufacturing equipment health condition monitoring and predictive maintenance, which employed advanced AI algorithms to detect early machine failures and avoid downtimes. During his Ph.D. program, Dr. Wu has authored and co-authored 7 papers that have appeared in prestigious journals and conference proceedings.

In addition to his contributions in support of the WHIN project, Dr. Wu's research has expanded to embrace additional topics related to prognostic and machine health management. He has developed novel, efficient machine learning (deep learning) methods to perform diagnostic or prognostic analysis related to machine health conditions including anomaly detection, fault classification, and prediction of remaining useful life. Moreover, he has worked on improving our understanding of how to interpret a machine learning model and abstract it to provide more phenomenological interpretations. He has provided new insights into the use of ML models for decision-making. These are all critical issues that need to be addressed to make machine learning more widely applicable. It should be noted that the effort also aligns with the goal of sustainability in manufacturing, as it can reduce energy consumption and materials waste, and extend the useful life of equipment. 

Dr. Wu's dissertation, entitled "Enhancing Interpretability and Adaptability of Manufacturing Equipment Health Models and Establishment of Cost Models for Maintenance Decisions," focuses on developing data-driven machine health models based on deep learning in manufacturing systems and explores three directions related to the practical implementation of prognostic and health management: i) model interpretation, ii) model adaptability and robustness enhancement, and iii) cost-benefit analysis of maintenance strategies. His work provides new insights and methods for improving the reliability of manufacturing equipment and maintenance decision-making, making a significant contribution to the field.