Subject: Seminar on March 8th,2012(Faculty Candidate)

Time:13:30~14:30, March 8th,2012

Venue：Conference Room (North 510), Shunde Building

Host: Kaibo Wang

Topic: Integrated Modeling of Heterogeneous Information with Applications in Computer Experiments and Failure Prognostics

Speaker: Qiang Zhou, Department of Industrial and Systems Engineering, University of Wisconsin-Madison. (Faculty Candidate)

Abstract:Owing to the advancement in information and sensing technologies, the proliferation of data has been ubiquitous in all aspects of engineering system design, analysis, monitoring, and control. Meanwhile, there is an increasing trend that information gathered for the same system becomes heterogeneous: they may come from different sources (e.g., signals collected by different sensors) and/or may be of different nature (e.g., continuous or discrete). Traditional modeling methods are normally limited to dealing with single sets of homogenous data. By taking advantage of heterogeneous data sources/types, an integrated modeling approach helps to extract useful information to a larger extend for better modeling and prediction performances. In this talk, we will introduce novel data integration algorithms in two areas: (1) computer experiments with both quantitative and qualitative input factors, and (2) failure prognostics based on both degradation signal and time-to-event data. Through examples, the methods have demonstrated to gain superior performances than existing methods. 

Bio: Qiang Zhou is a research associate in quality and manufacturing engineering at the Department of Industrial and Systems Engineering, University of Wisconsin-Madison. He received a B.S. in Automotive Engineering (2005) and a M.S. in Mechanical Engineering (2007) from Tsinghua University, China, a M.S. in Statistics (2010) and a Ph.D. in Industrial Engineering (2011) at University of Wisconsin-Madison. His research interests include modeling and analysis of complex engineering systems for the purpose of design optimization, quality control and productivity improvement. He is the recipient of the Richard S. and Harriet K. Fein Scholarship from ISyE department, UW-Madison and BP research award. He also received the best student paper award from Quality, Statistics and Reliability (QSR) cluster at INFORMS 2010.