The prediction of remaining useful life (RUL) is central to Prognostics and Health Management (PHM), and for an accurate prediction, one needs to build an effective prognostics model. However, in building a prognostics model there are some issues to be addressed, such as: assets having more than one fault mode, experiencing unexpected degradation, and some assets possessing more degradation information than others. Previous prognostic studies on assets designed a prediction model to identify fault classifications and another separate model for RUL estimation. We demonstrate the performance and potential advantages of a multi-output neural network prognostics framework, which accounts for more than one fault mode and produces simultaneously fault classifications and RUL predictions for turbofan engines. Using the multi-input-multi-output (MIMO) framework as benchmark results, this multi-output framework contains one final prediction model which could improve the results while being computationally more efficient. For further fault classification and RUL improvements, we introduced cross-teaching between two deep learning models within the multi-output framework. We compared the performance of the cross-teaching multi-output framework with the multi-output neural network framework. The proposed methodologies were applied to the Commercial Modular Aero-Propulsion System Simulation (C-MAPSS)’s FD003, FD004, and the New Commercial Modular Aero-Propulsion System Simulation (N-CMAPSS)’s DS02 datasets. The results indicated that the cross-teaching multi-output neural network framework performed better overall in terms of root mean squared error (RMSE) across all datasets.
BharatheedasanMTKumarawanmidhasTandLA, et al. Enhanced fault diagnosis and remaining useful life predication of rolling bearings using a hybrid multilayer perceptron and LSTM network model. Alex Eng J2025; 115: 355–369.
2.
PandhareVJiaXLeeJ. Collaborative prognostics for machine fleets using a novel federated baseline learner. Annu Conf PHM Soc2021; 13(1): 1–10.
3.
JiangMXingTZioE, et al. A Bayesian data-driven framework for aleatoric and epistemic uncertainty quantification in remaining useful life predictions. IEEE Sens J2024; 24: 42255–42267. https://doi.org/10.1109/JSen.2024.3479079
4.
ZengZDi MaioFZioE, et al. A hierarchical decision-making framework for the assessment of the prediction capability of prognostic methods. J Risk Reliab2017; 231(1): 36–52.
5.
ZhangZSongWLiQ, et al. Multiscale global and local self-attention-based network for remaining useful life prediction. Meas Sci Technol2023; 34: 125154.
6.
RuanDWuYYanJ, et al. Fuzzy-membership-based framework for task transfer learning between fault diagnosis and RUL prediction. IEEE Trans Reliab2023; 72: 989–1002.
7.
JiaoRPengKDongJ, et al. Fault monitoring and remaining useful life prediction framework for multiple fault modes in prognostics. Reliab Eng Syst Saf2020; 203: 107028.
8.
YuTRenZZhangY, et al. A rolling bearing fault diagnosis method based on a new data fusion mechanism and improved CNN. J2024; 238(6): 1156–1169.
9.
XiongJZhouJMaY, et al. Adaptive deep learning-based remaining useful life prediction framework for systems with multiple failure patterns. Reliab Eng Syst Saf2023; 235: 109244.
10.
ParkJ. Efficient ensemble via rotation-based self-supervised learning technique and multi-input-multi-output network. IEEE Access2024; 12: 36135–36147.
11.
ChadhaGSShahSRBSchwungA, et al. Shared temporal attention transformer for remaining useful lifetime estimation. IEEE Access2022; 10: 74244–74258.
12.
WangZZhaoLZhangD, et al. Multi-dimensional failure risk load prediction of electric motor based on physical model and data-driven approach. J2025; 239(5): 966–983.
13.
MoradiRRuiz-Tagle PalazuelosALopez DroguettE, et al. Toward a framework for risk monitoring of complex engineering systems with online operational data: a deep learning-based solution. J Risk Reliab2023; 237(5): 910–921.
14.
LiuLSongXZhouZ. Aircraft engine remaining useful life estimation via a double attention-based data-driven architecture. Reliab Eng Syst Saf2022; 221: 108330.
15.
ZhangZSongWLiQ. Dual-aspect self-attention based on transformer for remaining useful life prediction. IEEE Trans Instrum Meas2022; 71: 2505711.
16.
LuoXHuMSongT, et al. Semi-supervised medical image segmentation via cross teaching between CNN and transformer. Med Imag Deep Learn2022; 172: 1–14.
17.
YangYZhangLRenL. Semi-supervised medical image segmentation via cross teaching between mobilenet and mobilevit. Image Vis Comput2024; 150(3): 105196.
18.
PengJEstradaGPedersoliM, et al. Deep co-training for semi-supervised image segmentation. Pattern Recognit2020; 107: 107269.
19.
CuiJCaoLZhangT. A two-stage Gaussian process regression model for remaining useful prediction of bearings. J Risk Reliab2024; 238(2): 333–348.
20.
LiangYMaoWWuC. Unsupervised incremental transfer learning with knowledge distillation for online remaining useful life prediction of rotating machinery. J Risk Reliab2025; 239(1): 15–30.
SharifiMGhalebMTaghipourS. Joint parallel-machine scheduling and maintenance planning optimisation with deterioration, unexpected breakdowns, and condition-based maintenance. Int J Syst Sci Oper Logist2023; 10(1): 1–24.
23.
YounessGAalahA. An explainable artificial intelligence approach for remaining useful life prediction. Aerospace2023; 10: 474.
24.
YıldırımUAfşerH. Linear methods for predictive maintenance: the case of NASA C-MAPSS datasets. Appl Sci2025; 15: 9945.
25.
XiaoGJinCBaiJ. Remaining useful life prediction for aero-engines based on multi-scale dilated fusion attention model. Appl Sci2025; 15: 9183.
26.
XiongJFinkOZhouJ, et al. Controlled physics-informed data generation for deep learning-based remaining useful life prediction under unseen operation conditions. Mech Syst Signal Process2023; 197: 110359.
27.
KoutroulisGMutluBKernR. Constructing robust health indicators from complex engineered systems via anticausal learning. Eng Appl Artif Intell2022; 113: 104926.
28.
WahidAYahyaMBreslinJG, et al. Self-attention transformer-based architecture for remaining useful life estimation of complex machines. Procedia Comput Sci2023; 217: 456–464.
29.
LiYChenZHuC, et al. Bearing remaining useful life prediction with an improved CNN-LSTM network using an artificial gorilla troop optimization algorithm. J Risk Reliab2025; 239(1): 55–67.
30.
KameiSTaghipourS. A comparison study of centralized and decentralized federated learning approaches utilizing the transformer architecture for estimating remaining useful life. Reliab Eng Syst Saf2023; 233: 109130.
31.
DuanYLiHZhangN. Mechanical health indicator construction and similarity remaining useful life prediction based on natural language processing model. Meas Sci Technol2022; 33: 094008.
32.
WenJRenJZhaoZ, et al. Residual-based adversarial feature decoupling for remaining useful life prediction of aero-engines under variable operating conditions. Expert Syst Appl2024; 255: 124538.
33.
Forouzandeh ShahrakiAAl-DahidiSRahim TaleqaniA, et al. Using LSTM neural network to predict remaining useful life of electrolytic capacitors in dynamic operating conditions. J2023; 237(1): 16–28.
34.
MalhathkarSThenmozhiS. Deep learning for time series forecasting – with a focus on loss functions and error measures. In: Proceedings of the 2022 IEEE world conference on applied intelligence and computing (AIC), 17–19 August 2022, Sonbhadra, India.
35.
GuoTLuanJGaoJ, et al. Computer-aided diagnosis of pituitary microadenoma on dynamic contrast-enhanced MRI based on spatio-temporal features. Expert Syst Appl2025; 260: 125414.
36.
RamosPMMacedoJBMaiorCB, et al. Combining BERT with numerical variables to classify injury leave based on accident description. J Risk Reliab2024; 238(5): 945–956.
37.
SharmaK. Uncertainty-aware deep learning framework for remaining useful life prediction in turbofan engines with learned aleatoric uncertainty. arXiv:2511.19124, 2025. https://arxiv.org/abs/2511.19124.
38.
KingNHernandezA. Bayesian deep ensembles for reliable remaining useful life estimation under domain shift. Multidiscip Res Comput Inf Syst2025; 5(10): 960–971.
