Evaluating the Impact of Data Partitioning and Client Selection on Federated Remaining Useful Life Prediction in Aviation
##plugins.themes.bootstrap3.article.main##
##plugins.themes.bootstrap3.article.sidebar##
https://orcid.org/0000-0002-9023-8295
Mohammed M. Mabkhot
Abstract
Federated Learning (FL) is increasingly explored for Remaining Useful Life (RUL) prediction in aviation, motivated by the distributed nature of operational data across operators and platforms, the need to learn from heterogeneous fleet conditions, and the requirement to preserve data ownership and intellectual property by avoiding raw data sharing. While existing studies report promising results, they rely on subjectively defined benchmarking setups, where non-Independent and Identically Distributed (non-IID) data partitioning, client selection, and comparison criteria are selected without systematic examination of the bias they may introduce. Consequently, it remains unclear whether reported performance differences arise from the learning method itself or from unexamined configuration choices. This paper investigates the bias induced by data partitioning and client selection configurations in FL for aviation RUL prediction. Representative heterogeneity and client selection scenarios, including operating-condition shift, are evaluated under systematic learning settings to isolate their effect on model outcomes. The results show that both partitioning and selection choices can materially influence reported performance independent of the underlying model, demonstrating that selection bias alone can alter fleet-level RUL estimates. These findings highlight the need for harmonised and well-grounded benchmarking practices to support objective comparison and credible evaluation of FL approaches in aviation applications.
How to Cite
##plugins.themes.bootstrap3.article.details##
Federated Learning, Remaining Useful Life, Data Partitioning, Client Selection, Aviation, Prognostics and Health Management
Cho, Y. J., Wang, J., & Joshi, G. (2020). Client selection in federated learning: Convergence analysis and power-of-choice selection strategies.
Dalgkitsis, A., Koufakis, A., Stutterheim, J., Mifsud, A., Atwani, P., Gommans, L., De Laat, C., Papagianni, C., & Oprescu, A. (2024). Secure collaborative model training with dynamic federated learning in multi-domain environments. In Proceedings of SC 2024-W: Workshops of the International Conference for High Performance Computing, Networking, Storage and Analysis (pp. 755–759). doi: https://doi.org/10.1109/SCW63240.2024.00107
Dhada, M., Jain, A. K., Herrera, M., Perez Hernandez, M., & Parlikad, A. K. (2020). Secure and communications-efficient collaborative prognosis. IET Collaborative Intelligent Manufacturing, 2(4), 164–173. doi: https://doi.org/10.1049/iet-cim.2020.0035
Du, N. H., Long, N. H., Ha, K. N., Hoang, N. V., Huong, T. T., & Tran, K. P. (2023). Trans-Lighter: A lightweight federated learning-based architecture for remaining useful lifetime prediction. Computers in Industry, 148, 103888. doi: https://doi.org/10.1016/j.compind.2023.103888
European Union Aviation Safety Agency. (2023). Artificial intelligence roadmap 2.0.
Fu, L., Zhang, H., Gao, G., Zhang, M., & Liu, X. (2023). Client selection in federated learning: Principles, challenges, and opportunities. IEEE Internet of Things Journal, 10(24), 21811–21819. doi: https://doi.org/10.1109/JIOT.2023.3299573
Fu, S., & Avdelidis, N. P. (2023). Prognostic and health management of critical aircraft systems and components: An overview. Sensors, 23(19). doi: https://doi.org/10.3390/s23198124
Jeong, C., Yue, X., & Chung, S. (2026). Fed-Joint: Joint modeling of nonlinear degradation signals and failure events for remaining useful life prediction using federated learning. Reliability Engineering & System Safety, 267, 111833. doi: https://doi.org/10.1016/j.ress.2025.111833
Kabashkin, I. (2024). Integration of foundation models and federated learning in AIoT-based aircraft health monitoring systems. Mathematics, 12(21). doi: https://doi.org/10.3390/math12213428
Kabashkin, I. (2025). Framework for addressing imbalanced data in aviation with federated learning. Information, 16(2), 147. doi: https://doi.org/10.3390/info16020147
Kamei, S., & Taghipour, S. (2023). A comparison study of centralized and decentralized federated learning approaches utilizing the transformer architecture for estimating remaining useful life. Reliability Engineering & System Safety, 233, 109130. doi: https://doi.org/10.1016/j.ress.2023.109130
Kim, N.-H., Dawn, A., & Joo-Ho, C. (2017). Prognostics and health management of engineering systems. Springer International Publishing.
Landau, D., de Pater, I., Mitici, M., & Saurabh, N. (2026). Federated learning framework for collaborative remaining useful life prognostics: An aircraft engine case study. Future Generation Computer Systems, 174, 107945. doi: https://doi.org/10.1016/j.future.2025.107945
Lee, J., & Mitici, M. (2023). Deep reinforcement learning for predictive aircraft maintenance using probabilistic remaining useful life prognostics. Reliability Engineering & System Safety, 230, 108908. doi: https://doi.org/10.1016/j.ress.2022.108908
Li, J., Chen, T., & Teng, S. (2024). A comprehensive survey on client selection strategies in federated learning. Computer Networks, 251, 110663. doi: https://doi.org/10.1016/j.comnet.2024.110663
Li, X., Zhang, W., Li, X., & Hao, H. (2024). Partial domain adaptation in remaining useful life prediction with incomplete target data. IEEE/ASME Transactions on Mechatronics, 29(3), 1903–1913. doi: https://doi.org/10.1109/TMECH.2023.3325538
Listou Ellefsen, A., Bjørlykhaug, E., Æsøy, V., Ushakov, S., & Zhang, H. (2019). Remaining useful life predictions for turbofan engine degradation using semi-supervised deep architecture. Reliability Engineering & System Safety, 183, 240–251. doi: https://doi.org/10.1016/j.ress.2018.11.027
Llasag Rosero, R., Silva, C., Ribeiro, B., Albisser, M., Brutsche, M., & Arias Chao, M. (2025). Label synchronization strategies for hybrid federated learning. Reliability Engineering & System Safety, 256, 110751. doi: https://doi.org/10.1016/j.ress.2024.110751
Mabkhot, M. M., Kalawsky, R. S., & Liaqat, A. (2025). Introducing the Manufacturing Digital Passport (MDP): A new concept for realising digital thread data sharing in aerospace and complex manufacturing. Systems, 13(8), 700. doi: https://doi.org/10.3390/systems13080700
Nishio, T., & Yonetani, R. (2019). Client selection for federated learning with heterogeneous resources in mobile edge. In ICC 2019: 2019 IEEE International Conference on Communications (ICC) (pp. 1–7). doi: https://doi.org/10.1109/ICC.2019.8761315
Raouf, I., Kumar, P., Cheon, Y., Tanveer, M., Jo, S.-H., & Kim, H. S. (2025). Advances in prognostics and health management for aircraft landing gear: Progress, challenges, and future possibilities. International Journal of Precision Engineering and Manufacturing-Green Technology, 12(1), 301–320. doi: https://doi.org/10.1007/s40684-024-00646-4
Sharma, R. K. (2024). Framework based on machine learning approach for prediction of the remaining useful life: A case study of an aviation engine. Journal of Failure Analysis and Prevention, 24(3), 1333–1350. doi: https://doi.org/10.1007/s11668-024-01922-w
Smestad, C., & Li, J. (2023). A systematic literature review on client selection in federated learning. ACM International Conference Proceeding Series, 2–11. doi: https://doi.org/10.1145/3593434.3593438
Soltani, B., Haghighi, V., Mahmood, A., Sheng, Q. Z., & Yao, L. (2022). A survey on participant selection for federated learning in mobile networks. In Proceedings of the 17th ACM Workshop on Mobility in the Evolving Internet Architecture (MobiArch 2022) (pp. 19–24). doi: https://doi.org/10.1145/3556548.3559633
Stanton, I., Munir, K., Ikram, A., & El-Bakry, M. (2023). Predictive maintenance analytics and implementation for aircraft: Challenges and opportunities. Systems Engineering, 26(2), 216–237. doi: https://doi.org/10.1002/sys.21651
Vermelin, W. S., Mishra, M., Eng, M. P., Andersson, D., & Kyprianidis, K. (2024). Collaborative training of data-driven remaining useful life prediction models using federated learning. International Journal of Prognostics and Health Management, 15(2). doi: https://doi.org/10.36001/ijphm.2024.v15i2.3821
Zhang, J., Wang, P., Yan, R., & Gao, R. X. (2018). Long short-term memory for machine remaining life prediction. Journal of Manufacturing Systems, 48, 78–86. doi: https://doi.org/10.1016/j.jmsy.2018.05.011
This work is licensed under a Creative Commons Attribution 3.0 Unported License.
The Prognostic and Health Management Society advocates open-access to scientific data and uses a Creative Commons license for publishing and distributing any papers. A Creative Commons license does not relinquish the author’s copyright; rather it allows them to share some of their rights with any member of the public under certain conditions whilst enjoying full legal protection. By submitting an article to the International Conference of the Prognostics and Health Management Society, the authors agree to be bound by the associated terms and conditions including the following:
As the author, you retain the copyright to your Work. By submitting your Work, you are granting anybody the right to copy, distribute and transmit your Work and to adapt your Work with proper attribution under the terms of the Creative Commons Attribution 3.0 United States license. You assign rights to the Prognostics and Health Management Society to publish and disseminate your Work through electronic and print media if it is accepted for publication. A license note citing the Creative Commons Attribution 3.0 United States License as shown below needs to be placed in the footnote on the first page of the article.
First Author et al. This is an open-access article distributed under the terms of the Creative Commons Attribution 3.0 United States License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.