Predicting machine failures from industrial time series data

F.J.M. Jansen, M.J. Holenderski, T. Ozcelebi, Paulien Dam, Bas Tijsma

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

2 Citations (Scopus)
1 Downloads (Pure)

Abstract

This paper addresses the problem of predicting machine failures in an industrial manufacturing process based on multivariate time series data. A workflow is presented for cleaning and preprocessing the data, and for training and evaluating a predictive model. Its implementation is modular and extensible to support changes in the underlying produc- tion processes and the gathered data. Two predictive models are presented, based on Convolutional Neural Networks and Recurrent Neural Networks, and evaluated on data from an advanced machining process used for cutting complex shapes into metal pieces.
Original languageEnglish
Title of host publication2018 5th International Conference on Control, Decision and Information Technologies (CoDIT)
Place of PublicationPiscataway
PublisherInstitute of Electrical and Electronics Engineers
Pages1091-1096
Number of pages6
ISBN (Electronic)978-1-5386-5065-3
DOIs
Publication statusPublished - 2018
Event5th International Conference on Control, Decision and Information Technologies (CoDIT 2018) - Thessaloniki, Greece
Duration: 10 Apr 201813 Apr 2018

Conference

Conference5th International Conference on Control, Decision and Information Technologies (CoDIT 2018)
CountryGreece
City Thessaloniki
Period10/04/1813/04/18

Fingerprint Dive into the research topics of 'Predicting machine failures from industrial time series data'. Together they form a unique fingerprint.

  • Cite this

    Jansen, F. J. M., Holenderski, M. J., Ozcelebi, T., Dam, P., & Tijsma, B. (2018). Predicting machine failures from industrial time series data. In 2018 5th International Conference on Control, Decision and Information Technologies (CoDIT) (pp. 1091-1096). [8394915] Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/CoDIT.2018.8394915