Publication detail

Analyzing Machine Performance Using Data Mining

POSPÍŠIL, M. BARTÍK, V. HRUŠKA, T.

Original Title

Analyzing Machine Performance Using Data Mining

English Title

Analyzing Machine Performance Using Data Mining

Type

conference paper

Language

en

Original Abstract

This paper focuses on analysis of machine performance in a manufacturing company. Machine behavior can be complex, because it usually consists of many tasks. Performance of these tasks depends on product attributes, worker's speed, and therefore, analysis is not simple. Performance analysis results can be used for different purposes. Prediction and description are typical products of data mining. Prediction should be used for online monitoring of the manufactory process and as an input for a scheduler. Description can serve as information for managers to know which attributes of products cause problems more frequently. However manufacturing processes are complex, every process is quite unique. Our long term goal is to generalize the most common patterns to build general analyzer. This task is not simple because the lack of real word data and information. Therefore this work may contribute to the other researchers in their understanding of real world manufacturing problems.

English abstract

This paper focuses on analysis of machine performance in a manufacturing company. Machine behavior can be complex, because it usually consists of many tasks. Performance of these tasks depends on product attributes, worker's speed, and therefore, analysis is not simple. Performance analysis results can be used for different purposes. Prediction and description are typical products of data mining. Prediction should be used for online monitoring of the manufactory process and as an input for a scheduler. Description can serve as information for managers to know which attributes of products cause problems more frequently. However manufacturing processes are complex, every process is quite unique. Our long term goal is to generalize the most common patterns to build general analyzer. This task is not simple because the lack of real word data and information. Therefore this work may contribute to the other researchers in their understanding of real world manufacturing problems.

Keywords

Process mining, data mining, manufacturing, performance analysis, simulation, prediction, monitoring, scheduling.

Released

13.07.2016

Publisher

Institute of Electrical and Electronics Engineers

Location

Athens

ISBN

978-1-5090-4239-5

Book

2016 IEEE Symposium on Computational Intelligence and Data Mining

Edition

NEUVEDEN

Edition number

NEUVEDEN

Pages from

1

Pages to

7

Pages count

7

URL

Documents

BibTex


@inproceedings{BUT131008,
  author="Milan {Pospíšil} and Vladimír {Bartík} and Tomáš {Hruška}",
  title="Analyzing Machine Performance Using Data Mining",
  annote="This paper focuses on analysis of machine performance in a manufacturing company.
Machine behavior can be complex, because it usually consists of many tasks.
Performance of these tasks depends on product attributes, worker's speed, and
therefore, analysis is not simple. Performance analysis results can be used for
different purposes. Prediction and description are typical products of data
mining. Prediction should be used for online monitoring of the manufactory
process and as an input for a scheduler. Description can serve as information for
managers to know which attributes of products cause problems more frequently.
However manufacturing processes are complex, every process is quite unique. Our
long term goal is to generalize the most common patterns to build general
analyzer. This task is not simple because the lack of real word data and
information. Therefore this work may contribute to the other researchers in their
understanding of real world manufacturing problems.",
  address="Institute of Electrical and Electronics Engineers",
  booktitle="2016 IEEE Symposium on Computational Intelligence and Data Mining",
  chapter="131008",
  doi="10.1109/SSCI.2016.7849923",
  edition="NEUVEDEN",
  howpublished="electronic, physical medium",
  institution="Institute of Electrical and Electronics Engineers",
  year="2016",
  month="july",
  pages="1--7",
  publisher="Institute of Electrical and Electronics Engineers",
  type="conference paper"
}