Home / Current Issue / Paper 1720220
Artificial Intelligence for Predictive Maintenance and Structural Health Monitoring in Aerospace Systems: A State-of-the-Art Review
Subject area: Science,Engineering and Technology · Area of research: Artificial Intelligence, Aerospace
DOI: https://doi.org/10.64388/IREV10I2-1720220
Abstract
AI is moving the aerospace industry toward intelligent and data-driven maintenance practices and the aerospace industry is undergoing a paradigm shift to intelligent and data-driven maintenance systems with AI. This paper provides a comprehensive review of the state-of-the-art of AI methods for predictive maintenance (PdM) and structural health monitoring (SHM) in aerospace systems. We systematically classify and study machine learning (ML) and deep learning (DL) solutions for aircraft engine prognostics, composite damage detection, fatigue life prediction, and impact characterization for supervised, unsupervised and hybrid learning frameworks. A new taxonomy is suggested that correlates the AI techniques to the typical aerospace SHM functions: damage detection, localization, classification and quantification, remaining useful life (RUL) estimation. The review also explores new trends such as physics-informed neural networks (PINNs), digital twins, federated learning, and Explainable AI (XAI) in the context of enabling reliable aerospace SHM. The critical analysis of the TRLs shows that supervised and hybrid methods are at the TRL 5–6 (operational stages), while federated learning and physics-informed deep learning are still in the early experimental stages (TRL 2–4). Data scarcity, environmental and operational variability are identified as key challenges along with mitigation measures, interpretability for certification and real-time onboard processing constraints are identified with mitigation measures. A roadmap to the future of self-learning, intelligent, and certifiable SHM systems that are adaptive and able to diagnose damages autonomously and integrate digital twins without hassle is presented as the conclusion of the paper. This review will be useful to researchers and practitioners in the field of AI-assisted aerospace maintenance and safety.
Keywords
Artificial Intelligence, Predictive Maintenance, Structural Health Monitoring, Aerospace Systems, Aircraft Engine Prognostics, Composite Damage Detection, Physics-Informed Neural Networks, Digital Twins, Federated Learning, Explainable AI.
How to cite this paper
@article{1720220,
author = {Abe Olaoluwa Ojo, Falodun Joshua Mayowa, Babalola Emmanuel Ayooluwa, Ekebafe Godsfavour Onofueh, Omidiji Iyanuoluwa Isaac},
title = {Artificial Intelligence for Predictive Maintenance and Structural Health Monitoring in Aerospace Systems: A State-of-the-Art Review},
journal = {Iconic Research And Engineering Journals},
year = {2026},
volume = {10},
number = {2},
pages = {465-476},
issn = {2456-8880},
url = {https://www.irejournals.com/formatedpaper/1720220.pdf},
abstract = {AI is moving the aerospace industry toward intelligent and data-driven maintenance practices and the aerospace industry is undergoing a paradigm shift to intelligent and data-driven maintenance systems with AI. This paper provides a comprehensive review of the state-of-the-art of AI methods for predictive maintenance (PdM) and structural health monitoring (SHM) in aerospace systems. We systematically classify and study machine learning (ML) and deep learning (DL) solutions for aircraft engine prognostics, composite damage detection, fatigue life prediction, and impact characterization for supervised, unsupervised and hybrid learning frameworks. A new taxonomy is suggested that correlates the AI techniques to the typical aerospace SHM functions: damage detection, localization, classification and quantification, remaining useful life (RUL) estimation. The review also explores new trends such as physics-informed neural networks (PINNs), digital twins, federated learning, and Explainable AI (XAI) in the context of enabling reliable aerospace SHM. The critical analysis of the TRLs shows that supervised and hybrid methods are at the TRL 5–6 (operational stages), while federated learning and physics-informed deep learning are still in the early experimental stages (TRL 2–4). Data scarcity, environmental and operational variability are identified as key challenges along with mitigation measures, interpretability for certification and real-time onboard processing constraints are identified with mitigation measures. A roadmap to the future of self-learning, intelligent, and certifiable SHM systems that are adaptive and able to diagnose damages autonomously and integrate digital twins without hassle is presented as the conclusion of the paper. This review will be useful to researchers and practitioners in the field of AI-assisted aerospace maintenance and safety.},
keywords = {Artificial Intelligence, Predictive Maintenance, Structural Health Monitoring, Aerospace Systems, Aircraft Engine Prognostics, Composite Damage Detection, Physics-Informed Neural Networks, Digital Twins, Federated Learning, Explainable AI.},
month = {August},
doi = {https://doi.org/10.64388/IREV10I2-1720220}
}