Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

A holistic approach for quality oriented maintenance planning supported by data mining methods

: Glawar, Robert; Kemeny, Zsolt; Nemeth, Tanja; Matyas, Kurt; Monostori, László; Sihn, Wilfried

Volltext urn:nbn:de:0011-n-4288958 (917 KByte PDF)
MD5 Fingerprint: 2ca3935295612743ed98c1b19ee77dbe
(CC) by-nc-nd
Erstellt am: 13.1.2017

Procedia CIRP 57 (2016), S.259-264
ISSN: 2212-8271
Conference on Manufacturing Systems (CMS) <49, 2016, Stuttgart>
Zeitschriftenaufsatz, Konferenzbeitrag, Elektronische Publikation
Fraunhofer Austria ()
Fraunhofer PMI ()
Instandhaltung; Vorhersage; Qualität; data mining; Qualitätsprüfung; manufacturing system

Appropriate maintenance measures, which are carried out at the right time are a key factor to secure plant availability, product quality and process efficiency in modern manufacturing systems. Established maintenance strategies oftentimes lack in combining these strongly related aspects. They are not capable to anticipate in a holistic way and therefore lead to unnecessarily high maintenance efforts, wasted resources and the occurrence of quality and availability impairments.
In order to realize a holistic and anticipatory approach for maintenance planning, a methodology which consistently compiles and correlates various data via “cause and effect” coherences is depicted. By breaking down the production facilities on component level a basis is set to link condition monitoring data, wear data, quality and production data by using data mining methods. This framework enables the identification of maintenance-critical conditions and the prediction of failure moments and quality deviations.