ORIGINAL RESEARCH article
Front. Artif. Intell.
Sec. Machine Learning and Artificial Intelligence
Volume 8 - 2025 | doi: 10.3389/frai.2025.1630907
This article is part of the Research TopicTransformative Impact of AI and ML on Modern Manufacturing ProcessesView all 4 articles
FMEA Based Prescriptive Model for Equipment Repair Guidance
Provisionally accepted
- 1Department of Informatics and Computing, Mandume Ya Ndemufayo University, Lubango, Angola
- 2Centre Algoritmi, Department of Information Systems, University of Minho, Guimarães, Portugal
Select one of your emails
You have multiple emails registered with Frontiers:
Notify me on publication
Please enter your email address:
If you already have an account, please login
You don't have a Frontiers account ? You can register here
ABSTRACT abstract Accurately predicting the steps needed to address machine faults is essential for reducing downtime and increasing production in modern manufacturing. This article uses machine failure data and Failure Mode and Effects Analysis to show how machine learning can help maintenance teams select optimal repair methods. The target variables are repair actions, and the input consists of 10 multivariate time-series machine parameters. The prediction task is framed as a classification problem. Two modeling approaches are considered. The first combines multiple time series into a single sequence, allowing the use of Multi-Layer Perceptron, Convolutional Neural Networks, and Fully Convolutional Networks. The second approach retains the time series as 3-D arrays, enabling more advanced applications of MLP, CNN, Multi-Head CNN, and FCN models. These models are evaluated for their ability to predict repair actions, with particular attention to how time-series processing and model design influence classification accuracy. The results identify effective strategies for predicting machine repairs and improving prescriptive maintenance in manufacturing.
Keywords: Multivariate time-series, multi-class classification, FMEA, Corrective maintenance, Quality assurance and control
Received: 18 May 2025; Accepted: 23 Sep 2025.
Copyright: © 2025 Oliveira, A. Brito and Brandão. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
* Correspondence:
Domingos F. Oliveira, dfilipe@umn.edu.ao
Miguel A. Brito, mab@dsi.uminho.pt
Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.