SHAP-Based Feature Selection with a Hybrid Convolutional and Recurrent Deep Learning Framework for Remaining Useful Life Prediction of Aero-Engines
##plugins.themes.bootstrap3.article.main##
##plugins.themes.bootstrap3.article.sidebar##
Published
Mar 1, 2026
Naouel Ouafek
Tarek Berghout
Samira Abderrezek
Abdelhabib Bourouis
Abstract
Prognostics and health management for aeroengines is crucial, especially for reducing catastrophic loss of life and minimizing maintenance costs. Accurate Remaining Useful Life (RUL) estimation optimizes schedules, lowers costs, and enhances safety. Physics-based modeling for RUL assessment is challenging due to its inability to fully account for system and environment-related dynamics. Therefore, machine and deep learning approaches are highly recommended. This paper proposes a new method for RUL prediction of turbofan engines using the well-known (CMAPSS) dataset. Generally, data from acquired engines may be corrupted by anomalies due to sensor failures or environmental disturbances, which can affect the accuracy of prediction models. Therefore, this paper combines advanced techniques for reducing irrelevant and redundant sensor signals by applying Shapley Additive Explanations (SHAP) to refine feature selection by quantifying each sensor’s contribution to RUL prediction, ensuring both interpretability and efficiency. Then an anomaly detection and removal followed by RUL prediction with deep learning are used for such complex tasks. Specifically, K-means clustering, autoencoders, Temporal Convolutional Networks (TCN) and Long Short-Term Memory (LSTM) networks. This approach enables effective data preprocessing by detecting and removing anomalies, thus enhancing the quality of the training data. Moreover, uncertainty analysis is performed to assess the reliability of the predictions, including confidence intervals for the results. The experimental results show substantial improvements in predictive accuracy, confirmed by better evaluation metrics, numerical evaluations, and the coefficient of determination, compared to traditional approaches. The TCN-LSTM model achieved an average performance improvement of 24.05% on FD001, 12.44% on FD003, and 4.63% on FD002 (using Monte-Carlo uncertainty estimation). Furthermore, for the FD004, a performance decrease of -4.68
##plugins.themes.bootstrap3.article.details##
Keywords
RUL Prediction, Autoencoder, K-means, SHAP, Uncertainty, CMAPSS dataset, XAI, Outlier removal, Denoising, TCN, LSTM
References
Alomari, Y., Baptista, M., & Ando, M. (2024). Integrating network theory and shap analysis for enhanced rul prediction in aeronautics. PHM Society European Conference, 8, 15. doi: 10.36001/phme.2024.v8i1.4077
Arias Chao, M., Kulkarni, C., Goebel, K., & Fink, O. (2022). Fusing physics-based and deep learning models for prognostics. Reliability Engineering & System Safety, 217, 107961. Retrieved from https://www.sciencedirect.com/science/article/pii/S0951832021004725 doi: 10.1016/j.ress.2021.107961
Arias Chao, M., Kulkarni, C., Goebel, K., & Fink, O. (2021). Aircraft engine run-to-failure dataset under real flight conditions for prognostics and diagnostics. Data, 6(1), 5. Retrieved from https://www.mdpi.com/2306-5729/6/1/5 doi: 10.3390/data6010005
Asif, O., Haider, S. A., Naqvi, S. R., Zaki, J. F.W., Kwak, K.-S., & Islam, S. M. R. (2022). A deep learning model for remaining useful life prediction of aircraft turbofan engine on cmapss dataset. IEEE Access, 10, 95425–95440. doi: 10.1109/ACCESS.2022.3203406
Ayodeji, A., Wang, W., Su, J., Yuan, J., & Liu, X. (2021). An empirical evaluation of attention-based multi-head models for improved turbofan engine remaining useful life prediction. arXiv preprint. doi: 10.48550/arXiv.2109.01761
Berghout, T., & Benbouzid, M. (2022). A systematic guide for predicting remaining useful life with machine learning. Electronics, 11(7), 1125. doi: 10.3390/electronics11071125
Chandola, V., Banerjee, A., & Kumar, V. (2009). Anomaly detection: A survey. ACM Computing Surveys (CSUR), 41(3), 1–58. doi: 10.1145/1541880.1541882
Chen, J., Chen, D., & Liu, G. (2021). Using temporal convolution network for remaining useful lifetime prediction. Engineering Reports, 3(3), e12305. doi: 10.1002/eng2.12305
Deng, S., & Zhou, J. (2024a). Cnn for feature extraction in rul prediction. International Journal of Computational Intelligence Systems.
Deng, S., & Zhou, J. (2024b). Lstm-based rul prediction for aero-engines. International Journal of Computational Intelligence Systems.
Deng, S., & Zhou, J. (2024c). Prediction of remaining useful life of aero-engines based on cnn-lstm-attention. International Journal of Computational Intelligence Systems, 17, 232. doi: 10.1007/s44196-024-00639-w
Deng, S., & Zhou, J. (2024d). Prediction of remaining useful life of aero-engines based on cnn-lstm-attention. International Journal of Computational Intelligence Systems, 17, 232. doi: 10.1007/s44196-024-00639-w
Fan, Z., Li, W., & Chang, K.-C. (2023). A bidirectional long short-term memory autoencoder transformer for remaining useful life estimation. Mathematics, 11(24), 4972. doi: 10.3390/math11244972
Genane, Y., & Aalah, A. (2023). An explainable artificial intelligence approach for remaining useful life prediction. Aerospace, 10(5), 474. doi: 10.3390/aerospace10050474
Hawkins, D. M. (1980). Identification of outliers. Springer. doi: 10.1007/978-94-015-3994-4
Hong, C. W., Lee, C., Lee, K., Ko, M.-S., Kim, D. E., & Hur, K. (2020). Remaining useful life prognosis for turbofan engine using explainable deep neural networks with dimensionality reduction. Sensors, 20(22), 6626. doi: 10.3390/s20226626
Hu, C., Cheng, F., Ma, L., & Li, B. (2022). State of charge estimation for lithium-ion batteries based on tcn-lstm neural networks. Journal of the Electrochemical Society, 169, 030544. doi: 10.1149/1945-7111/ac5cf2
Hu, J., Jiang, X., Xu, H., & Zhang, K. (2025). Fault prediction of aircraft engine based on adaptive hybrid sampling and bilstm. Scientific Reports, 15, 13726. doi: 10.1038/s41598-025-98756-9
Hu, T., Zhu, X., & Tian, M. (2025). Enhanced soc estimation method for lithium-ion batteries using bayesian-optimized tcn–lstm neural networks. AIP Advances, 15. doi: 10.1063/5.0243811
Kong, X., Chen, Z., & Liu, W. (2025). Deep learning for time series forecasting: a survey. Int. J. Mach. Learn. & Cyber.. doi: 10.1007/s13042-025-02560-w
Lei, Y., Li, N., Guo, L., Yan, T., & Lin, J. (2018). Machinery health prognostics: A systematic review from data acquisition to rul prediction. Mechanical Systems and Signal Processing, 104, 799–834. doi: 10.1016/j.ymssp.2017.11.016
Li, H., Zhang, Z., Li, T., & Si, X. (2024). A review on physics-informed data-driven remaining useful life prediction: Challenges and opportunities. Mechanical Systems and Signal Processing, 209, 111120. Retrieved from https://www.sciencedirect.com/science/article/pii/S0888327024000189 doi: 10.1016/j.ymssp.2024.111120
Liu, L.,Wang, L., & Yu, Z. (2021a). Deep convolutional networks for accurate rul prediction. International Journal of Computational Intelligence Systems.
Liu, L., Wang, L., & Yu, Z. (2021b). Lightgbm model for aircraft engine rul prediction. International Journal of Computational Intelligence Systems.
Liu, L., Wang, L., & Yu, Z. (2021c). Remaining useful life estimation of aircraft engines based on deep convolution neural network and lightgbm combination model. International Journal of Computational Intelligence Systems.
Liu, L., Wang, L., & Yu, Z. (2021d). Xgboost application in rul estimation of aero-engines. International Journal of Computational Intelligence Systems.
MacQueen, J. (1967). Some methods for classification and analysis of multivariate observations. Berkeley Symp. on Math. Statist. and Prob., 281–297. doi: 10.48550/arXiv.2110.03585
Naser, M. Z., & Alavi, A. H. (2023). Error metrics and performance fitness indicators for artificial intelligence and machine learning in engineering and sciences. Architecture, Structures and Construction, 3(4), 499–517. doi: 10.1007/s44150-021-00015-8
Sakurada, M., & Yairi, T. (2014). Anomaly detection using autoencoders with nonlinear dimensionality reduction. In Proceedings of the mlsda 2014 2nd workshop on machine learning for sensory data analysis (pp. 4–11). doi: 10.1145/2689746.2689747
Saxena, A., Goebel, K., Simon, D., & Eklund, N. (2008). Damage propagation modeling for aircraft engine run-to-failure simulation. In Proceedings of the international conference on prognostics and health management. doi: 10.1109/PHM.2008.4711414
Sharma, L., Singh, H., & Choudhary, M. P. (2025). Application of deep learning techniques for analysis and prediction of particulate matter at kota city, india. EQA - International Journal of Environmental Quality, 66, 107–115. doi: 10.6092/issn.2281-4485/20687
Tan, W. M., & Teo, T. H. (2021). Remaining useful life prediction using temporal convolution with attention. AI, 2(1), 48–70. Retrieved from https://www.mdpi.com/2673-2688/2/1/5 doi: 10.3390/ai2010005
Wang, S., Ji, B.,Wang,W., Ma, J., & Chen, H.-B. (2022). Remaining useful life prediction of aero-engine based on deep convolutional lstm network. In 6th international conference on system reliability and safety (icsrs) (pp. 494–499). doi: 10.1109/ICSRS56243.2022.10067647
Wang, X., Li, Y., Xu, Y., Liu, X., Zheng, T., & Zheng, B. (2023). Remaining useful life prediction for aero-engines using a time-enhanced multi-head self-attention model. Aerospace. doi: 10.3390/aerospace10010080
Wang, X., Li, Y., Xu, Y., Liu, X., Zheng, T., & Zheng, B. (2024). Improved feature selection for rul prediction using temporal attention mechanisms. Aerospace.
Wang, X., Zhao, Y.,& Pourpanah, F. (2020). Recent advances in deep learning. Int. J. Mach. Learn. & Cyber., 11, 747–750. doi: 10.1007/s13042-020-01096-5
Xin, f. n. m., Wang, Y., Li, Y., Xu, X., Liu, T., & Zheng, B. (2023). Remaining useful life prediction for aero-engines using a time-enhanced multi-head self-attention model. Aerospace, 10. doi: 10.3390/aerospace10010080
Xu, H., Chen, W., Zhao, N., Li, Z., Bu, J., Li, Z., . . . Deng, S. (2019). Deep learning-based anomaly detection in multivariate time series data. In Proceedings of the 28th international joint conference on artificial intelligence (ijcai) (pp. 2325–2331). doi: 10.24963/ijcai.2019/322
Xu, R., & Wunsch, D. (2005). Survey of clustering algorithms. IEEE Transactions on Neural Networks.
Yang, H., Zhao, F., Jiang, G., Sun, Z., & Mei, X. (2019). A novel deep learning approach for machinery prognostics based on time windows. Applied Sciences, 9(22), 4813. Retriev ed from https://www.mdpi.com/2076-3417/9/22/4813 doi: 10.3390/app9224813
Arias Chao, M., Kulkarni, C., Goebel, K., & Fink, O. (2022). Fusing physics-based and deep learning models for prognostics. Reliability Engineering & System Safety, 217, 107961. Retrieved from https://www.sciencedirect.com/science/article/pii/S0951832021004725 doi: 10.1016/j.ress.2021.107961
Arias Chao, M., Kulkarni, C., Goebel, K., & Fink, O. (2021). Aircraft engine run-to-failure dataset under real flight conditions for prognostics and diagnostics. Data, 6(1), 5. Retrieved from https://www.mdpi.com/2306-5729/6/1/5 doi: 10.3390/data6010005
Asif, O., Haider, S. A., Naqvi, S. R., Zaki, J. F.W., Kwak, K.-S., & Islam, S. M. R. (2022). A deep learning model for remaining useful life prediction of aircraft turbofan engine on cmapss dataset. IEEE Access, 10, 95425–95440. doi: 10.1109/ACCESS.2022.3203406
Ayodeji, A., Wang, W., Su, J., Yuan, J., & Liu, X. (2021). An empirical evaluation of attention-based multi-head models for improved turbofan engine remaining useful life prediction. arXiv preprint. doi: 10.48550/arXiv.2109.01761
Berghout, T., & Benbouzid, M. (2022). A systematic guide for predicting remaining useful life with machine learning. Electronics, 11(7), 1125. doi: 10.3390/electronics11071125
Chandola, V., Banerjee, A., & Kumar, V. (2009). Anomaly detection: A survey. ACM Computing Surveys (CSUR), 41(3), 1–58. doi: 10.1145/1541880.1541882
Chen, J., Chen, D., & Liu, G. (2021). Using temporal convolution network for remaining useful lifetime prediction. Engineering Reports, 3(3), e12305. doi: 10.1002/eng2.12305
Deng, S., & Zhou, J. (2024a). Cnn for feature extraction in rul prediction. International Journal of Computational Intelligence Systems.
Deng, S., & Zhou, J. (2024b). Lstm-based rul prediction for aero-engines. International Journal of Computational Intelligence Systems.
Deng, S., & Zhou, J. (2024c). Prediction of remaining useful life of aero-engines based on cnn-lstm-attention. International Journal of Computational Intelligence Systems, 17, 232. doi: 10.1007/s44196-024-00639-w
Deng, S., & Zhou, J. (2024d). Prediction of remaining useful life of aero-engines based on cnn-lstm-attention. International Journal of Computational Intelligence Systems, 17, 232. doi: 10.1007/s44196-024-00639-w
Fan, Z., Li, W., & Chang, K.-C. (2023). A bidirectional long short-term memory autoencoder transformer for remaining useful life estimation. Mathematics, 11(24), 4972. doi: 10.3390/math11244972
Genane, Y., & Aalah, A. (2023). An explainable artificial intelligence approach for remaining useful life prediction. Aerospace, 10(5), 474. doi: 10.3390/aerospace10050474
Hawkins, D. M. (1980). Identification of outliers. Springer. doi: 10.1007/978-94-015-3994-4
Hong, C. W., Lee, C., Lee, K., Ko, M.-S., Kim, D. E., & Hur, K. (2020). Remaining useful life prognosis for turbofan engine using explainable deep neural networks with dimensionality reduction. Sensors, 20(22), 6626. doi: 10.3390/s20226626
Hu, C., Cheng, F., Ma, L., & Li, B. (2022). State of charge estimation for lithium-ion batteries based on tcn-lstm neural networks. Journal of the Electrochemical Society, 169, 030544. doi: 10.1149/1945-7111/ac5cf2
Hu, J., Jiang, X., Xu, H., & Zhang, K. (2025). Fault prediction of aircraft engine based on adaptive hybrid sampling and bilstm. Scientific Reports, 15, 13726. doi: 10.1038/s41598-025-98756-9
Hu, T., Zhu, X., & Tian, M. (2025). Enhanced soc estimation method for lithium-ion batteries using bayesian-optimized tcn–lstm neural networks. AIP Advances, 15. doi: 10.1063/5.0243811
Kong, X., Chen, Z., & Liu, W. (2025). Deep learning for time series forecasting: a survey. Int. J. Mach. Learn. & Cyber.. doi: 10.1007/s13042-025-02560-w
Lei, Y., Li, N., Guo, L., Yan, T., & Lin, J. (2018). Machinery health prognostics: A systematic review from data acquisition to rul prediction. Mechanical Systems and Signal Processing, 104, 799–834. doi: 10.1016/j.ymssp.2017.11.016
Li, H., Zhang, Z., Li, T., & Si, X. (2024). A review on physics-informed data-driven remaining useful life prediction: Challenges and opportunities. Mechanical Systems and Signal Processing, 209, 111120. Retrieved from https://www.sciencedirect.com/science/article/pii/S0888327024000189 doi: 10.1016/j.ymssp.2024.111120
Liu, L.,Wang, L., & Yu, Z. (2021a). Deep convolutional networks for accurate rul prediction. International Journal of Computational Intelligence Systems.
Liu, L., Wang, L., & Yu, Z. (2021b). Lightgbm model for aircraft engine rul prediction. International Journal of Computational Intelligence Systems.
Liu, L., Wang, L., & Yu, Z. (2021c). Remaining useful life estimation of aircraft engines based on deep convolution neural network and lightgbm combination model. International Journal of Computational Intelligence Systems.
Liu, L., Wang, L., & Yu, Z. (2021d). Xgboost application in rul estimation of aero-engines. International Journal of Computational Intelligence Systems.
MacQueen, J. (1967). Some methods for classification and analysis of multivariate observations. Berkeley Symp. on Math. Statist. and Prob., 281–297. doi: 10.48550/arXiv.2110.03585
Naser, M. Z., & Alavi, A. H. (2023). Error metrics and performance fitness indicators for artificial intelligence and machine learning in engineering and sciences. Architecture, Structures and Construction, 3(4), 499–517. doi: 10.1007/s44150-021-00015-8
Sakurada, M., & Yairi, T. (2014). Anomaly detection using autoencoders with nonlinear dimensionality reduction. In Proceedings of the mlsda 2014 2nd workshop on machine learning for sensory data analysis (pp. 4–11). doi: 10.1145/2689746.2689747
Saxena, A., Goebel, K., Simon, D., & Eklund, N. (2008). Damage propagation modeling for aircraft engine run-to-failure simulation. In Proceedings of the international conference on prognostics and health management. doi: 10.1109/PHM.2008.4711414
Sharma, L., Singh, H., & Choudhary, M. P. (2025). Application of deep learning techniques for analysis and prediction of particulate matter at kota city, india. EQA - International Journal of Environmental Quality, 66, 107–115. doi: 10.6092/issn.2281-4485/20687
Tan, W. M., & Teo, T. H. (2021). Remaining useful life prediction using temporal convolution with attention. AI, 2(1), 48–70. Retrieved from https://www.mdpi.com/2673-2688/2/1/5 doi: 10.3390/ai2010005
Wang, S., Ji, B.,Wang,W., Ma, J., & Chen, H.-B. (2022). Remaining useful life prediction of aero-engine based on deep convolutional lstm network. In 6th international conference on system reliability and safety (icsrs) (pp. 494–499). doi: 10.1109/ICSRS56243.2022.10067647
Wang, X., Li, Y., Xu, Y., Liu, X., Zheng, T., & Zheng, B. (2023). Remaining useful life prediction for aero-engines using a time-enhanced multi-head self-attention model. Aerospace. doi: 10.3390/aerospace10010080
Wang, X., Li, Y., Xu, Y., Liu, X., Zheng, T., & Zheng, B. (2024). Improved feature selection for rul prediction using temporal attention mechanisms. Aerospace.
Wang, X., Zhao, Y.,& Pourpanah, F. (2020). Recent advances in deep learning. Int. J. Mach. Learn. & Cyber., 11, 747–750. doi: 10.1007/s13042-020-01096-5
Xin, f. n. m., Wang, Y., Li, Y., Xu, X., Liu, T., & Zheng, B. (2023). Remaining useful life prediction for aero-engines using a time-enhanced multi-head self-attention model. Aerospace, 10. doi: 10.3390/aerospace10010080
Xu, H., Chen, W., Zhao, N., Li, Z., Bu, J., Li, Z., . . . Deng, S. (2019). Deep learning-based anomaly detection in multivariate time series data. In Proceedings of the 28th international joint conference on artificial intelligence (ijcai) (pp. 2325–2331). doi: 10.24963/ijcai.2019/322
Xu, R., & Wunsch, D. (2005). Survey of clustering algorithms. IEEE Transactions on Neural Networks.
Yang, H., Zhao, F., Jiang, G., Sun, Z., & Mei, X. (2019). A novel deep learning approach for machinery prognostics based on time windows. Applied Sciences, 9(22), 4813. Retriev ed from https://www.mdpi.com/2076-3417/9/22/4813 doi: 10.3390/app9224813
Section
Technical Papers