Log-based predictive maintenance in discrete parts manufacturing

Publikation: Beitrag in einer FachzeitschriftKonferenzartikelForschungBegutachtung

Abstract

The performance of discrete parts manufacturing systems is heavily influenced by unplanned machine breakdowns. Predictive maintenance allows for the conversation of unplanned machine breakdowns to scheduled corrective maintenance actions. We present a data-driven approach for estimating the probability of machine breakdown during specified time interval in the future. Machine learning algorithms are utilized for a specific use-case which is based on real-world data-sets including machine log messages, event logs and operational information. The paper describes applied data-mining, feature-extraction and machine learning methods and concludes with results indicating that machine failures can be reliably predicted up to 168 hours in advance.
Originalspracheenglisch
Seiten (von - bis)528-533
Seitenumfang6
FachzeitschriftProcedia CIRP
Jahrgang79
DOIs
PublikationsstatusVeröffentlicht - 2019
VeranstaltungIntelligent Computation in Manufacturing Engineering: Innovative and Cognitive Production Technology and Systems - Island of Ischia, Ischia, Italien
Dauer: 18 Jul 201820 Jul 2018
http://www.icme.unina.it/ICME%2018/ICME_14.htm

Fingerprint

Learning systems
Learning algorithms
Data mining
Feature extraction
Manufacturing
Machine breakdown
Machine learning
Manufacturing systems
Corrective maintenance
Learning methods
Learning algorithm

Schlagwörter

  • Prädiktion
  • Restlebensdauer
  • random forest
  • ensemble prediction
  • Maschinelles Lernen

ASJC Scopus subject areas

  • Artificial intelligence
  • !!Industrial and Manufacturing Engineering
  • !!Management Science and Operations Research
  • !!Control and Systems Engineering

Fields of Expertise

  • Mobility & Production

Dies zitieren

Log-based predictive maintenance in discrete parts manufacturing. / Gutschi, Clemens; Furian, Nikolaus; Suschnigg, Josef; Neubacher, Dietmar; Vössner, Siegfried.

in: Procedia CIRP, Jahrgang 79, 2019, S. 528-533.

Publikation: Beitrag in einer FachzeitschriftKonferenzartikelForschungBegutachtung

@article{ae86ca8635e64a1984b6cc224186173b,
title = "Log-based predictive maintenance in discrete parts manufacturing",
abstract = "The performance of discrete parts manufacturing systems is heavily influenced by unplanned machine breakdowns. Predictive maintenance allows for the conversation of unplanned machine breakdowns to scheduled corrective maintenance actions. We present a data-driven approach for estimating the probability of machine breakdown during specified time interval in the future. Machine learning algorithms are utilized for a specific use-case which is based on real-world data-sets including machine log messages, event logs and operational information. The paper describes applied data-mining, feature-extraction and machine learning methods and concludes with results indicating that machine failures can be reliably predicted up to 168 hours in advance.",
keywords = "Pr{\"a}diktion, Restlebensdauer, random forest, ensemble prediction, Maschinelles Lernen, predictive analytics, Predictive maintenance, random forest, ensemble prediction, Machine learning, Probability of failure estimation, maintenance, Maintenance Management, data mining, feature engineering, feature selection, Remaining useful life, Ensemble prediction, Random forrest",
author = "Clemens Gutschi and Nikolaus Furian and Josef Suschnigg and Dietmar Neubacher and Siegfried V{\"o}ssner",
year = "2019",
doi = "10.1016/j.procir.2019.02.098",
language = "English",
volume = "79",
pages = "528--533",
journal = "Procedia CIRP",
issn = "2212-8271",
publisher = "Elsevier B.V.",

}

TY - JOUR

T1 - Log-based predictive maintenance in discrete parts manufacturing

AU - Gutschi, Clemens

AU - Furian, Nikolaus

AU - Suschnigg, Josef

AU - Neubacher, Dietmar

AU - Vössner, Siegfried

PY - 2019

Y1 - 2019

N2 - The performance of discrete parts manufacturing systems is heavily influenced by unplanned machine breakdowns. Predictive maintenance allows for the conversation of unplanned machine breakdowns to scheduled corrective maintenance actions. We present a data-driven approach for estimating the probability of machine breakdown during specified time interval in the future. Machine learning algorithms are utilized for a specific use-case which is based on real-world data-sets including machine log messages, event logs and operational information. The paper describes applied data-mining, feature-extraction and machine learning methods and concludes with results indicating that machine failures can be reliably predicted up to 168 hours in advance.

AB - The performance of discrete parts manufacturing systems is heavily influenced by unplanned machine breakdowns. Predictive maintenance allows for the conversation of unplanned machine breakdowns to scheduled corrective maintenance actions. We present a data-driven approach for estimating the probability of machine breakdown during specified time interval in the future. Machine learning algorithms are utilized for a specific use-case which is based on real-world data-sets including machine log messages, event logs and operational information. The paper describes applied data-mining, feature-extraction and machine learning methods and concludes with results indicating that machine failures can be reliably predicted up to 168 hours in advance.

KW - Prädiktion

KW - Restlebensdauer

KW - random forest

KW - ensemble prediction

KW - Maschinelles Lernen

KW - predictive analytics

KW - Predictive maintenance

KW - random forest

KW - ensemble prediction

KW - Machine learning

KW - Probability of failure estimation

KW - maintenance

KW - Maintenance Management

KW - data mining

KW - feature engineering

KW - feature selection

KW - Remaining useful life

KW - Ensemble prediction

KW - Random forrest

UR - http://www.scopus.com/inward/record.url?scp=85065420211&partnerID=8YFLogxK

U2 - 10.1016/j.procir.2019.02.098

DO - 10.1016/j.procir.2019.02.098

M3 - Conference article

VL - 79

SP - 528

EP - 533

JO - Procedia CIRP

JF - Procedia CIRP

SN - 2212-8271

ER -