Comparative study of reliability prediction of compressor system by ANN and TTF time series techniques

P. A. Ozor, C. Mbohwa

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review


A comparative study of reliability prediction of Compressor system by means of Artificial Neural Network (ANN) and Time to failure (TTF) analysis is presented. The data for the study was taken from a specific example namely; CO2 compressor at the Utilities Department of a beverage plant, preferred to be addressed as X plant for the sake of confidentiality. The trained neural network predicted the reliability of the compressor very well, given the very low values of mean square error (1.02564 ×10-3 ) and high value of the regression (0.998742). The result was compared to what obtained in reliability prediction by time-to-failure time series. The later equally gave a considerably high value for the coefficient of determination of the linear regression line between the predicted values and the actual values (0.9824). Yet artificial neural network result gave a better value. The result shows that use of artificial neural network is a good means of predicting reliability of deteriorating systems.

Original languageEnglish
Title of host publicationProceedings of the World Congress on Engineering 2018, WCE 2018
EditorsDavid WL Hukins, Len Gelman, Andrew Hunter, S. I. Ao, A. M. Korsunsky
PublisherNewswood Limited
Number of pages5
ISBN (Electronic)9789881404893
Publication statusPublished - 2018
Event2018 World Congress on Engineering, WCE 2018 - London, United Kingdom
Duration: 4 Jul 20186 Jul 2018

Publication series

NameLecture Notes in Engineering and Computer Science
ISSN (Print)2078-0958


Conference2018 World Congress on Engineering, WCE 2018
Country/TerritoryUnited Kingdom


  • Artificial neural network
  • Deteriorating systems
  • Reliability prediction
  • Time to failure analysis

ASJC Scopus subject areas

  • Computer Science (miscellaneous)


Dive into the research topics of 'Comparative study of reliability prediction of compressor system by ANN and TTF time series techniques'. Together they form a unique fingerprint.

Cite this