Open AccessDiagnostic and Prognostic Models for Predictive Maintenance: Multi-Criteria Comparative Analysis
Author(s) -
Mohammed Bouaicha,
Imad El Adraoui,
Nadia Machkour,
Hassan Gziri,
Mourad Zegrari
Publication year - 2021
Publication title -
international journal emerging technology and advanced engineering
Language(s) - English
Resource type - Journals
ISSN - 2250-2459
DOI - 10.46338/ijetae1021_05
Subject(s) - computer science , reliability (semiconductor) , predictive maintenance , process (computing) , relevance (law) , predictive modelling , risk analysis (engineering) , reliability engineering , machine learning , engineering , medicine , power (physics) , physics , quantum mechanics , law , political science , operating system
Predictive maintenance has evolved considerably over the past two decades making this strategy an effective way to monitor the operation of industrial systems, thereby predicting its future states and remaining lifespan. It is therefore developed through a process that begins with the collection of information from the industrial system, the objective of which is its diagnosis or / and its prognosis. This article presents an analysis of single-model and multi-model approaches to the effect of diagnostic and prognostic tasks. This analysis is based on a multi-criteria comparison of the different models in order to provide a clear vision to choose the appropriate approach for predictive maintenance. The relevance of the comparative study is argued by the development of criteria directly impacting performance, reliability, efficiency and mutual cooperation between models. Conclusions are then drawn, in order to identify the appropriate diagnostic and prognostic approach for predictive maintenance.