Home // DATA ANALYTICS 2018, The Seventh International Conference on Data Analytics // View article

A Modeling Tool for Equipment Health Estimation Using a System Identification Approach

Authors:
Alexander Dementjev
Ilkay Wunderlich
Klaus Kabitzschh
Germar Schneider

Keywords: Equipment Health; Data-Driven Modeling; System Identification; Predictive Maintenance; Time Series Analysis.

Abstract:
In high-technology manufacturing industries like pharmaceuticals, semiconductors or photovoltaics the best and stable yields are very important. This can only be reached if the appropriate equipment is still operating in the specified working ranges according to the current recipes or process steps. Because of the high complexity of the corresponding equipment and processes, monitoring solutions like fault detection and classification or predictive maintenance are already established. The combination of such techniques with advanced process control leads to a mechanism for avoiding product scraps and, finally, to the maximization of the production efficiency and business competitiveness. The equipment health factor is an important index of the status of a processing equipment and at the same time a key enabler for advanced monitoring and control strategies. Three families of estimation techniques will be briefly explained in this paper: physics-inspired, statistical-based and data-driven. Then, the focus will be set on the system identification-based approach as the most promising and emerging equipment health estimation strategy, explain its backgrounds and its advantages will be illustrated. At the end, a universal tool for investigation and modeling of the equipment health factor will be described and discussed.

Pages: 61 to 66

Copyright: Copyright (c) IARIA, 2018

Publication date: November 18, 2018

Published in: conference

ISSN: 2308-4464

ISBN: 978-1-61208-681-1

Location: Athens, Greece

Dates: from November 18, 2018 to November 22, 2018