A MapReduce-Based Ensemble Learning Method with Multiple Classifier Types and Diversity for Condition-based Maintenance with Concept Drifts

Chun Cheng Lin, Lei Shu, Der-Jiunn Deng, Tzu Lei Yeh, Yu Hsiang Chen, Hsin Lung Hsieh

Research output: Contribution to specialist publicationArticle

Abstract

Condition-based maintenance (CBM) is to collect a huge amount of production data stream continuously from sensors or IoT devices attached to machines to forecast the time when to maintain machines or replace components. However, as conditions of machines change dynamically with time owing to machine aging or malfunction, the concept of capturing the foresting pattern from the data stream could drift unpredictably so that it is hard to find a robust forecasting method with high precision. Therefore, this work proposes an ensemble learning method with multiple classifier types and diversity for CBM in manufacturing industries. Aside from manipulating data diversity from previous works, the proposed method includes the features of multiple classifier types, dynamic weight adjusting, and data-based adaption to concept drifts for offline learning models, to promote the precision of the forecasting model and precisely detect and adapt to concept drifts. With these features, the proposed method requires powerful computing resources to efficiently respond to practical CBM applications in manufacturing industries. Therefore, furthermore, the implementation of the proposed method based on the MapReduce framework is proposed to increase computational efficiency. Simulation shows that the proposed method can detect and adapt to all concept drifts with a high precision rate.

Original languageEnglish
Specialist publicationIEEE Cloud Computing
DOIs
Publication statusAccepted/In press - 2017 Dec 22

Fingerprint

Classifiers
Computational efficiency
Industry
Aging of materials
Sensors
Internet of things

All Science Journal Classification (ASJC) codes

  • Computer Science (miscellaneous)
  • Software
  • Computer Science Applications
  • Computer Networks and Communications

Cite this

@misc{17f8c45ee4e64b5a84a6f525e89df3db,
title = "A MapReduce-Based Ensemble Learning Method with Multiple Classifier Types and Diversity for Condition-based Maintenance with Concept Drifts",
abstract = "Condition-based maintenance (CBM) is to collect a huge amount of production data stream continuously from sensors or IoT devices attached to machines to forecast the time when to maintain machines or replace components. However, as conditions of machines change dynamically with time owing to machine aging or malfunction, the concept of capturing the foresting pattern from the data stream could drift unpredictably so that it is hard to find a robust forecasting method with high precision. Therefore, this work proposes an ensemble learning method with multiple classifier types and diversity for CBM in manufacturing industries. Aside from manipulating data diversity from previous works, the proposed method includes the features of multiple classifier types, dynamic weight adjusting, and data-based adaption to concept drifts for offline learning models, to promote the precision of the forecasting model and precisely detect and adapt to concept drifts. With these features, the proposed method requires powerful computing resources to efficiently respond to practical CBM applications in manufacturing industries. Therefore, furthermore, the implementation of the proposed method based on the MapReduce framework is proposed to increase computational efficiency. Simulation shows that the proposed method can detect and adapt to all concept drifts with a high precision rate.",
author = "Lin, {Chun Cheng} and Lei Shu and Der-Jiunn Deng and Yeh, {Tzu Lei} and Chen, {Yu Hsiang} and Hsieh, {Hsin Lung}",
year = "2017",
month = "12",
day = "22",
doi = "10.1109/MCC.2017.455160123",
language = "English",
journal = "IEEE Cloud Computing",
issn = "2325-6095",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

A MapReduce-Based Ensemble Learning Method with Multiple Classifier Types and Diversity for Condition-based Maintenance with Concept Drifts. / Lin, Chun Cheng; Shu, Lei; Deng, Der-Jiunn; Yeh, Tzu Lei; Chen, Yu Hsiang; Hsieh, Hsin Lung.

In: IEEE Cloud Computing, 22.12.2017.

Research output: Contribution to specialist publicationArticle

TY - GEN

T1 - A MapReduce-Based Ensemble Learning Method with Multiple Classifier Types and Diversity for Condition-based Maintenance with Concept Drifts

AU - Lin, Chun Cheng

AU - Shu, Lei

AU - Deng, Der-Jiunn

AU - Yeh, Tzu Lei

AU - Chen, Yu Hsiang

AU - Hsieh, Hsin Lung

PY - 2017/12/22

Y1 - 2017/12/22

N2 - Condition-based maintenance (CBM) is to collect a huge amount of production data stream continuously from sensors or IoT devices attached to machines to forecast the time when to maintain machines or replace components. However, as conditions of machines change dynamically with time owing to machine aging or malfunction, the concept of capturing the foresting pattern from the data stream could drift unpredictably so that it is hard to find a robust forecasting method with high precision. Therefore, this work proposes an ensemble learning method with multiple classifier types and diversity for CBM in manufacturing industries. Aside from manipulating data diversity from previous works, the proposed method includes the features of multiple classifier types, dynamic weight adjusting, and data-based adaption to concept drifts for offline learning models, to promote the precision of the forecasting model and precisely detect and adapt to concept drifts. With these features, the proposed method requires powerful computing resources to efficiently respond to practical CBM applications in manufacturing industries. Therefore, furthermore, the implementation of the proposed method based on the MapReduce framework is proposed to increase computational efficiency. Simulation shows that the proposed method can detect and adapt to all concept drifts with a high precision rate.

AB - Condition-based maintenance (CBM) is to collect a huge amount of production data stream continuously from sensors or IoT devices attached to machines to forecast the time when to maintain machines or replace components. However, as conditions of machines change dynamically with time owing to machine aging or malfunction, the concept of capturing the foresting pattern from the data stream could drift unpredictably so that it is hard to find a robust forecasting method with high precision. Therefore, this work proposes an ensemble learning method with multiple classifier types and diversity for CBM in manufacturing industries. Aside from manipulating data diversity from previous works, the proposed method includes the features of multiple classifier types, dynamic weight adjusting, and data-based adaption to concept drifts for offline learning models, to promote the precision of the forecasting model and precisely detect and adapt to concept drifts. With these features, the proposed method requires powerful computing resources to efficiently respond to practical CBM applications in manufacturing industries. Therefore, furthermore, the implementation of the proposed method based on the MapReduce framework is proposed to increase computational efficiency. Simulation shows that the proposed method can detect and adapt to all concept drifts with a high precision rate.

UR - http://www.scopus.com/inward/record.url?scp=85039764265&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85039764265&partnerID=8YFLogxK

U2 - 10.1109/MCC.2017.455160123

DO - 10.1109/MCC.2017.455160123

M3 - Article

AN - SCOPUS:85039764265

JO - IEEE Cloud Computing

JF - IEEE Cloud Computing

SN - 2325-6095

ER -