EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

End-to-End Architecture for Real-Time IoT Analytics and Predictive Maintenance Using Stream Processing and ML Pipelines

Ouiam Khattach iD, Omar Moussaoui, Mohammed Hassine

Sensors · 07.05.2025 · 25 (9) · 2945

DOI: 10.3390/s25092945 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 07.05.2025. Verfasst von Ouiam Khattach und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department of Informatics, MATSI Laboratory EST, University Mohammed First, Oujda 60000, Morocco. Die Arbeit wurde bisher 44-mal zitiert (Crossref) und nennt 36 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The rapid proliferation of Internet of Things (IoT) devices across industries has created a need for robust, scalable, and real-time data processing architectures capable of supporting intelligent analytics and predictive maintenance. This paper presents a novel comprehensive architecture that enables end-to-end processing of IoT data streams, from acquisition to actionable insights. The system integrates Kafka-based message brokering for the high-throughput ingestion of real-time sensor data, with Apache Spark facilitating batch and stream extraction, transformation, and loading (ETL) processes. A modular machine-learning pipeline handles automated data preprocessing, training, and evaluation across various models. The architecture incorporates continuous monitoring and optimization components to track system performance and model accuracy, feeding insights to users via a dedicated Application Programming Interface (API). The design ensures scalability, flexibility, and real-time responsiveness, making it well suited for industrial IoT applications requiring continuous monitoring and intelligent decision-making.

Schlagwörter

Stichwörter der Autor:innen

internet of thingsmachine learning pipelinepredictive maintenancereal-time data

Autor:innen und Institutionen

  • Department of Informatics, MATSI Laboratory EST, University Mohammed First, Oujda 60000, Morocco.
  • Tisalabs Limited, T12 Y275 Cork, Ireland.

ORCID-Profile: Ouiam Khattach

Zitieren

APA

Khattach, O., Moussaoui, O., & Hassine, M. (2025). End-to-End Architecture for Real-Time IoT Analytics and Predictive Maintenance Using Stream Processing and ML Pipelines. Sensors, 25(9), 2945. https://doi.org/10.3390/s25092945

Vancouver

Khattach O, Moussaoui O, Hassine M. End-to-End Architecture for Real-Time IoT Analytics and Predictive Maintenance Using Stream Processing and ML Pipelines. Sensors. 2025;25(9):2945. doi:10.3390/s25092945

BibTeX
@article{Khattach2025,
  author = {Khattach, Ouiam and Moussaoui, Omar and Hassine, Mohammed},
  title = {End-to-End Architecture for Real-Time IoT Analytics and Predictive Maintenance Using Stream Processing and ML Pipelines},
  journal = {Sensors},
  year = {2025},
  volume = {25},
  number = {9},
  pages = {2945},
  doi = {10.3390/s25092945}
}
Kostenlos starten →