Artificial Intelligence–Driven Predictive Maintenance for Next-Generation Aircraft Systems
Academic International Journal of Engineering Science · 2025
DOI: 10.59675/e315 ↗Kurz gefasst
Fachartikel in der Zeitschrift Academic International Journal of Engineering Science, 2025. Verfasst von Ahmed E. Naphee. Die Arbeit nennt 48 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
The aviation business encounters a major challenge with regard to the cost of aircraft maintenance, operational time, and assurance of safety in the era of rising complexity and demands of the fleet. The conventional reactive and preventive maintenance strategies have been unable to handle the more demanding needs of the modern aircraft systems that consist of advanced material, elaborate electronic systems as well as intricate assemblies of mechanisms. The paper will discuss the application and the performance of predictive maintenance technologies powered by artificial intelligence and specifically in the field of next generation aircraft systems, the use of machine learning to predict faults and the use of the digital twin technology. An analysis of the performance of the different predictive maintenance frameworks was done on several airplane platforms such as commercial aviation fleets and unmanned aerial vehicles. The study shows that machine learning models, especially ensemble algorithms of a random forest, gradient boosting machine, and deep neural network, have a higher accuracy of fault detection of more than 94% and lower false positive rates by about 67% than standard threshold-based monitoring systems. It was found that the implementation of digital twins allowed the real-time estimation of the state of a system with correlation coefficients over 0.92 between the expected and actual patterns of component degradation. Economic analysis indicates that predictive maintenance with the use of artificial intelligence lowers the number of unscheduled maintenance incidents by 48 per cent, reduces aircraft on-ground time by 35 per cent., and saves the company about 1.2 million dollars a year per plane due to optimized maintenance scheduling and improved component life. Commercial operator case studies show that predictive analytics can help to perform dedicated interventions that avoid disastrous failures but do not lead to preventable maintenance interventions, which helps achieve better safety margins and operational efficiency. It was detected that the combination of Internet of Things sensor networks with cloud-based analytics platforms allows scalable implementation in a wide variety of fleet configurations. Nonetheless, the issues associated with implementation such as data quality assurance, the requirements of model interpretability, and consideration of regulatory requirements demand cautiousness in the development, as well as implementation of the system. The results suggest that predictive maintenance based on artificial intelligence is a radical way of managing aviation systems, as it offers significant gains in safety, reliability, and economic operation as well as contributes to the development of autonomous maintenance decision-making algorithms in the future aviation systems.
Zitieren
APA
Naphee, A. E. (2025). Artificial Intelligence–Driven Predictive Maintenance for Next-Generation Aircraft Systems. Academic International Journal of Engineering Science. https://doi.org/10.59675/e315
Vancouver
Naphee AE. Artificial Intelligence–Driven Predictive Maintenance for Next-Generation Aircraft Systems. Academic International Journal of Engineering Science. 2025. doi:10.59675/e315
BibTeX
@article{Naphee2025,
author = {Naphee, Ahmed E.},
title = {Artificial Intelligence–Driven Predictive Maintenance for Next-Generation Aircraft Systems},
journal = {Academic International Journal of Engineering Science},
year = {2025},
doi = {10.59675/e315}
}
EUVIMED