View All Events

Modern complex systems, driven by advancements in sensor technology and communication networks, generate vast amounts of data. While the data offers opportunities for deeper insights into system health, it also presents new and significant analytical challenges, including (a) efficient processing of rich and heterogeneous data streams, which may be high-dimensional, incomplete, or contaminated by noise; (b) the extraction of meaningful system knowledge that captures intricate component interactions, evolving dynamics, and underlying uncertainties; and (c) ensuring data privacy while enabling effective knowledge sharing and collaboration across different systems and organizations. This talk explores advanced modeling techniques for Prognostics and Health Management (PHM), focusing on capturing system dynamics, handling incomplete observations, and ensuring privacy-preserving analysis. Statistical-driven stochastic process approaches for real-time system monitoring and degradation modeling will be discussed, followed by an introduction to survival analysis techniques for addressing censored data and gaining deeper insights into system health trends. Finally, federated survival analysis will be presented as a framework that facilitates collaborative modeling across distributed data sources while preserving data confidentiality. Through comprehensive simulations and real-world case studies, the advantages of these approaches over existing techniques will be demonstrated. By integrating these methodologies, the talk highlights key innovations in data-driven PHM, supporting enhanced reliability and maintenance strategies for complex systems.

About the Speaker

Yuxin Wen is an Assistant Professor in the Fowler School of Engineering at Chapman University. She received her B.S. degree in Medical Informatics and Engineering from Sichuan University, Chengdu, China, in 2011, the M.S. degree in Biomedical Engineering from Zhejiang University, Hangzhou, China, in 2014, and the Ph.D. degree in Electrical and Computer Engineering from the University of Texas at El Paso, El Paso, TX, USA, in 2020. Her research focuses on machine learning and statistical modeling for high-dimensional data analytics, with an emphasis on time-to-event modeling in manufacturing and healthcare for health monitoring, system management, and patient outcome prediction. She is a member of INFORMS, IISE, and IEEE.