E3S Web Conf.
Volume 351, 202210th International Conference on Innovation, Modern Applied Science & Environmental Studies (ICIES’2022)
|Number of page(s)||6|
|Published online||24 May 2022|
Multi-criteria analysis of diagnostic and prognostic models for predictive maintenance
ENSAM, Hassan II University, Mers Sultan BP 9167, Casablanca, Morocco
* Corresponding author: email@example.com
Predictive maintenance has made considerable progress within the framework of Industry 4.0, making this strategy an effective means of monitoring the proper functioning of industrial systems, which helps to make maintenance operations more environmentally friendly, for example reduction of any kind of failure that causes loss of production and energy. This strategy is implemented through a process of collecting data in online or offline mode of the industrial system whose purpose is to monitor and predict its future state. This article first presents the different single-model and multi-model approaches used for diagnostic and prognostic tasks. An analysis of these models is then carried out, based on a multi-criteria comparison, and highlights the performance of machine learning (ML) models in this context of current digitalization. These ML models can be more efficient by combining with the physicsbased models in multi-model approaches. The relevance of the comparative study is argued by criteria impacting performance, effectiveness, efficiency, the possibility of processing heterogeneous data and mutual cooperation between models. Conclusions are then drawn, in order to give a clear vision for the choice of the diagnostic and prognosis approach of predictive maintenance adapted to the industrial system.
© The Authors, published by EDP Sciences, 2022
This is an Open Access article distributed under the terms of the Creative Commons Attribution License 4.0, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.