DC Field | Value | Language |
dc.contributor.author | FIDMA, MOhamed ABdelillah | - |
dc.date.accessioned | 2023-10-17T07:44:58Z | - |
dc.date.available | 2023-10-17T07:44:58Z | - |
dc.date.issued | 2023 | - |
dc.identifier.uri | https://repository.esi-sba.dz/jspui/handle/123456789/526 | - |
dc.description | Encadré par :M. Benslimane Sidi Mohamed / Co-encadrant : Mme. Hamour Nora /M. Ouchani Samir | en_US |
dc.description.abstract | Abstract :
The advent of Industry 4.0 has brought about significant advancements in manufacturing
processes, leveraging advanced sensing, and data analytics technologies
to optimize efficiency. Within this paradigm, predictive maintenance plays a crucial
role in ensuring the reliability and availability of production systems. However,
the heterogeneous nature of industrial data poses challenges in achieving effective
maintenance decision-making and interoperability across different manufacturing
systems.
This graudation project aims to address these challenges by proposing a hybrid
approach that combines the strengths of ontologies and machine learning
techniques. The research focuses on developing an intelligent system that leverages
standardized knowledge representation and predictive capabilities to enhance
maintenance decision-making in real-time.
This research objective is to explore the use of ontologies as a standardized
knowledge representation approach within the context of smart factories and Cyber-
Physical Systems (CPS). By bridging the semantic gap and improving interoperability,
ontologies facilitate the integration of diverse manufacturing systems,
enabling a holistic view of the maintenance processes.
Furthermore, we aim is to leverage machine learning techniques, such as data
mining and predictive analytics, to detect and predict potential anomalies in manufacturing
processes. By analyzing historical data, machine learning models can
identify patterns and trends that assist in making accurate predictions for maintenance
needs. This approach enables proactive maintenance actions, reducing
downtime and enhancing production efficiency. ***
Résumé :
L’av`enement de l’Industrie 4.0 a entraˆın´e d’importants progr`es dans les processus
de fabrication, en exploitant les technologies avanc´ees de d´etection et d’analyse de
donn´ees pour optimiser l’efficacit´e. Dans ce paradigme, la maintenance pr´edictive
joue un rˆole crucial dans la garantie de la fiabilit´e et de la disponibilit´e des
syst`emes de production. Cependant, la nature h´et´erog`ene des donn´ees industrielles
pose des d´efis pour prendre des d´ecisions de maintenance efficaces et assurer
l’interop´erabilit´e entre diff´erents syst`emes de fabrication.
Ce projet de fin d’´etudes vise `a relever ces d´efis en proposant une approche
hybride qui combine les atouts des ontologies et des techniques d’apprentissage
automatique. La recherche se concentre sur le d´eveloppement d’un syst`eme intelligent
qui exploite la repr´esentation normalis´ee des connaissances et les capacit´es
pr´edictives pour am´eliorer la prise de d´ecision en mati`ere de maintenance en temps
r´eel.
De plus, notre objectif est d’exploiter les techniques d’apprentissage automatique,
telles que l’exploration de donn´ees et l’analyse pr´edictive, pour d´etecter et
pr´edire les anomalies potentielles dans les processus de fabrication. En analysant
les donn´ees historiques, les mod`eles d’apprentissage automatique peuvent identifier
des sch´emas et des tendances qui aident `a faire des pr´edictions pr´ecises pour
les besoins de maintenance.
En combinant les ontologies et les techniques d’apprentissage automatique,
ce projet de recherche vise `a d´evelopper un syst`eme hybride de maintenance
pr´edictive. La validation de l’efficacit´e du syst`eme sera r´ealis´ee par des ´etudes
exp´erimentales, comparant ses performances `a celles des approches existantes.
L’´evaluation portera sur des aspects tels que la pr´ecision de la d´etection des besoins
de maintenance, la r´eduction des temps d’arrˆet et l’am´elioration globale de
l’efficacit´e de production. | en_US |
dc.language.iso | en | en_US |
dc.subject | Industry 4.0 | en_US |
dc.subject | Industrial Cyber-Physical System | en_US |
dc.subject | Predictive Maintenance | en_US |
dc.subject | Ontology | en_US |
dc.subject | Chronicle Mining | en_US |
dc.subject | SWRL Rules | en_US |
dc.title | Hybrid predictive maintenace system based on ontologies and machine learning in industry 4.0 | en_US |
dc.type | Thesis | en_US |
Appears in Collections: | Ingénieur
|