Big Data Mining in eMaintenance : An Overview

Projekt:

JVTC

Sammanfattning:
Maintenance related data are tending to be increasingly huge in
volume, rapid in velocity and vast in variety. Data with these
characteristics bring new challenges with respect to data analysis
and data mining, which requires new approaches and
technologies. In industry, related research and applications, some
contributions have been provided to utilize Big Data technologies
for extraction of information through pattern recognition
mechanisms via eMaintenance solutions. Today, the existing
contributions are not enabling a holistic approach for maintenance
data analysis and therefore are insufficient. However, the
immense value hidden inside the Big Data in eMaintenance is
arousing more and more attention from both academia and
industry. Hence, this paper aims to explore eMaintenance
solutions for maintenance decision-making through utilization of
Big Data technologies and approaches. The paper discusses Big
Data mining in eMaintenance through a general manner by
employing one of the widely accepted frameworks with the name
of Cross Industry Standard Process for Data Mining (CRISPDM).
In addition, the paper outlines features of maintenance data
and investigates six sub-processes (i.e. business understanding,
data understanding, data preparation, modeling, evaluation and
deployment) of data mining applications defined by CRISP-DM
within the domain of eMaintenance.

Länk till publikation i fulltext (pdf-fil, 21 226,9 kB. Öppnas i nytt fönster)

Författare: Liangwei Zhang ; Ramin Karim
Utgivare: Luleå tekniska universitet
Utgivningsdatum: 2014
Diarienummer: TRV 2011/58769
ISBN: 978-91-7439-973-8
Antal sidor: 12
Språk: Engelska
Kontaktperson: Per Olof Larsson Kråik, UHjbs