Communication Dans Un Congrès Année : 2025

Adaptive learning mode comparison for Predictive Maintenance Systems

Résumé

The industry is increasingly using data-driven predictive maintenance with machine learning to prevent equipment failures. However, manufacturing systems are inherently complex and operate in dynamic, non-stationary environments, leading to the phenomenon of concept drift. Due to edge device limitations, traditional full retraining methods are often impractical, making incremental learning a potential alternative. This study evaluates both approaches through simulated real-world scenarios. Results show that full retraining generally performs better, especially in ideal and degraded conditions. While incremental learning shows acceptable performance with accurate labels and imbalanced datasets, it struggles when dealing with significant labelling inaccuracies.

Fichier principal
Vignette du fichier
Adaptive_learning_mode_comparison_for_Predictive_Maintenance_Systems.pdf (787.26 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Licence

Dates et versions

hal-05162919 , version 1 (15-07-2025)

Licence

Identifiants

  • HAL Id : hal-05162919 , version 1

Citer

Jérémie Marchand, Jannik Laval, Aicha Sekhari, Vincent Cheutet, Jean-Baptiste Danielou. Adaptive learning mode comparison for Predictive Maintenance Systems. Manufacturing Modelling, Management and Control – IFAC MIM2025, Jun 2025, Trondheim, Norway. ⟨hal-05162919⟩
109 Consultations
194 Téléchargements

Partager

  • More