Anomaly Detection in Aircraft Fuel Systems Using Artificial Neural Networks for Maintenance Decision Support
International Journal of Scientific Multidisciplinary Research · 19.06.2026 · 4 (6) · 831-846
DOI: 10.55927/ijsmr.v4i6.63 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Scientific Multidisciplinary Research, online erschienen am 19.06.2026. Verfasst von Agita Ramadhani und 2 weiteren Autor:innen.
Abstract (Originalsprache)
This study proposes a predictive maintenance framework for aircraft fuel systems using Artificial Neural Networks (ANN) integrated with Failure Mode, Effects, and Criticality Analysis (FMECA). Operational flight recorder data consisting of fuel flow, torque, propeller speed, and outside air temperature parameters were processed through cleaning, normalization, and anomaly analysis before ANN training. The results show that the ANN model successfully detected anomalous operational conditions with stable learning performance and satisfactory classification capability. FMECA identified fuel pumps and fuel metering components as the most critical components affecting operational reliability and safety. The proposed framework supports condition-based maintenance by improving anomaly detection, maintenance prioritization, operational reliability, and aircraft safety in defense aviation environments.
Zitieren
APA
Ramadhani, A., Amperiawan, G., & Manawan, M. (2026). Anomaly Detection in Aircraft Fuel Systems Using Artificial Neural Networks for Maintenance Decision Support. International Journal of Scientific Multidisciplinary Research, 4(6), 831-846. https://doi.org/10.55927/ijsmr.v4i6.63
Vancouver
Ramadhani A, Amperiawan G, Manawan M. Anomaly Detection in Aircraft Fuel Systems Using Artificial Neural Networks for Maintenance Decision Support. International Journal of Scientific Multidisciplinary Research. 2026;4(6):831-846. doi:10.55927/ijsmr.v4i6.63
BibTeX
@article{Ramadhani2026,
author = {Ramadhani, Agita and Amperiawan, Gita and Manawan, Maykel},
title = {Anomaly Detection in Aircraft Fuel Systems Using Artificial Neural Networks for Maintenance Decision Support},
journal = {International Journal of Scientific Multidisciplinary Research},
year = {2026},
volume = {4},
number = {6},
pages = {831--846},
doi = {10.55927/ijsmr.v4i6.63}
}
EUVIMED