Speaker
Hao Yan
Description
In multistage manufacturing systems, modeling multiple quality indices based on the process sensing variables is important. However, the classic modeling technique predicts each quality variable one at a time, which fails to consider the correlation within or between stages. We propose a deep multistage multi-task learning framework to jointly predict all output sensing variables in a unified end-to-end learning framework according to the sequential system architecture in the MMS. Our numerical studies and real case study have shown that the new model has a superior performance compared to many benchmark methods as well as great interpretability through developed variable selection techniques.
| Keywords | Deep Multitask Learning, Multi-stage Manufacturing, quality prediction |
|---|
Primary authors
Hao Yan
Dr
Nurretin Sergin
(Arizona State University)
William Brenneman
(Procter & Gamble)
Dr
Stephen Lange
(the Procter & Gamble)
Dr
Shan Ba
(LinkedIn)