A hybrid Bi-LSTM model for data-driven maintenance planning

Abstract

Modern industries dependent on reliable asset operation under constrained resources employ intelligent maintenance methods to maximize efficiency. However, classical maintenance methods rely on assumed lifetime distributions and suffer from estimation errors and computational complexity. The advent of Industry 4.0 has increased the use of sensors for monitoring systems, while deep learning (DL) models have allowed for accurate system health predictions, enabling data-driven maintenance planning. Most intelligent maintenance literature has used DL models solely for remaining useful life (RUL) point predictions, and a substantial gap exists in further using predictions to inform maintenance plan optimization. The few existing studies that have attempted to bridge this gap suffer from having used simple system configurations and non-scalable models. Hence, this paper develops a hybrid DL model using Monte Carlo dropout to generate RUL predictions which are used to construct empirical system reliability functions used for the optimization of the selective maintenance problem (SMP). The proposed framework is used to plan maintenance for a mission-oriented series k-out-of-n:G system. Numerical experiments compare the framework’s performance against prior SMP methods and highlight its strengths. When minimizing cost, maintenance plans are frequently produced that result in mission survival while avoiding unnecessary repairs. The proposed method is usable in large-scale, complex scenarios and various industrial contexts. The method finds exact solutions while avoiding the need for computationally-intensive parametric reliability functions.

Keywords

Deep learning / System prognostics / Selective maintenance / Reliability and maintenance optimization / Information and Computing Sciences / Artificial Intelligence and Image Processing

References

[1] RiceW., CassadyC., NachlasJ.. Optimal maintenance plans under limited maintenance time. Proceedings of the Seventh Industrial Engineering Research Conference, 19981-3

[2] DialloC., VenkatadriU., KhatabA., LiuZ.. Optimal selective maintenance decisions for large serial k-out-of-n: G systems under imperfect maintenance. Reliab. Eng. Syst. Saf., 2018, 175: 234-245.

[3] O’NeilR., KhatabA., DialloC.. Optimal predictive selective maintenance for fleets of mission-oriented systems. Int. J. Prod. Res., 2024, 62248790-8817.

[4] JinZ., ChenC., SyntetosA., LiuY.. Enhanced bearing RUL prediction based on dynamic temporal attention and mixed MLP. Auton. Intell. Syst., 2025, 51. 2

[5] X. Qiao, V.L. Jauw, L.C. Seong, T. Banda, Advances and limitations in machine learning approaches applied to remaining useful life predictions: a critical review. Int. J. Adv. Manuf. Technol., 1–18 (2024)

[6] WuF., WuQ., TanY., XuX.. Remaining useful life prediction based on deep learning: a survey. Sensors, 2024, 2411. 3454

[7] AchourA., KammounM.A., HajejZ.. Towards optimizing multi-level selective maintenance via machine learning predictive models. Appl. Sci., 2023, 141. 313

[8] BasciftciB., AhmedS., GebraeelN.Z., YildirimM.. Stochastic optimization of maintenance and operations schedules under unexpected failures. IEEE Trans. Power Syst., 2018, 3366755-6765.

[9] NoussisA., SaifA., KhatabA., DialloC.. Using a minimalist Bi-LSTM for multi-faceted bearing fault detection. IFAC-PapersOnLine, 2024, 58191258-1263.

[10] SaxenaA., GoebelK., SimonD., EklundN.. Damage propagation modeling for aircraft engine run-to-failure simulation. 2008 International Conference on Prognostics and Health Management, 2008Los AlamitosIEEE1-9

[11] S. Hagmeyer, F. Mauthe, P. Zeiler, Creation of publicly available data sets for prognostics and diagnostics addressing data scenarios relevant to industrial applications. Int. J. Progn. Health Manag. 12(2) (2021)

[12] LiX., DingQ., SunJ.-Q.. Remaining useful life estimation in prognostics using deep convolution neural networks. Reliab. Eng. Syst. Saf., 2018, 172: 1-11.

[13] YangB., LiuR., ZioE.. Remaining useful life prediction based on a double-convolutional neural network architecture. IEEE Trans. Ind. Electron., 2019, 66129521-9530.

[14] Dehghan ShoorkandH., NourelfathM., HajjiA.. A deep learning approach for integrated production planning and predictive maintenance. Int. J. Prod. Res., 2023, 61237972-7991.

[15] SongT., LiuC., WuR., JinY., JiangD.. A hierarchical scheme for remaining useful life prediction with long short-term memory networks. Neurocomputing, 2022, 487: 22-33.

[16] RenL., DongJ., WangX., MengZ., ZhaoL., DeenM.J.. A data-driven auto-cnn-lstm prediction model for lithium-ion battery remaining useful life. IEEE Trans. Ind. Inform., 2020, 1753478-3487.

[17] LiX., WangL., WangC., MaX., MiaoB., XuD., ChengR.. A method for predicting remaining useful life using enhanced Savitzky-Golay filter and improved deep learning framework. Sci. Rep., 2024, 141. 23983

[18] SuC., LiL., WenZ.. Remaining useful life prediction via a variational autoencoder and a time-window-based sequence neural network. Qual. Reliab. Eng. Int., 2020, 3651639-1656.

[19] LinT., SongL., CuiL., WangH.. Advancing RUL prediction in mechanical systems: a hybrid deep learning approach utilizing non-full lifecycle data. Adv. Eng. Inform., 2024, 61. 102524

[20] FuS., LinL., WangY., GuoF., ZhaoM., ZhongB., ZhongS.. Mca-dtcn: a novel dual-task temporal convolutional network with multi-channel attention for first prediction time detection and remaining useful life prediction. Reliab. Eng. Syst. Saf., 2024, 241. 109696

[21] ZhangH., HeX., YanW., JiangZ., ZhuS.. A machine learning-based approach for product maintenance prediction with reliability information conversion. Auton. Intell. Syst., 2022, 21. 15

[22] ChenC., ShiJ., LuN., ZhuZ.H., JiangB.. Data-driven predictive maintenance strategy considering the uncertainty in remaining useful life prediction. Neurocomputing, 2022, 494: 79-88.

[23] ZhouK.-L., ChengD.-J., ZhangH.-B., HuZ-T., ZhangC.-Y.. Deep learning-based intelligent multilevel predictive maintenance framework considering comprehensive cost. Reliab. Eng. Syst. Saf., 2023, 237. 109357

[24] ShoorkandH.D., NourelfathM., HajjiA.. A hybrid CNN-LSTM model for joint optimization of production and imperfect predictive maintenance planning. Reliab. Eng. Syst. Saf., 2024, 241. 109707

[25] ZhuangL., XuA., WangX.-L.. A prognostic driven predictive maintenance framework based on Bayesian deep learning. Reliab. Eng. Syst. Saf., 2023, 234. 109181

[26] LiuY., HuangH.-Z.. Optimal selective maintenance strategy for multi-state systems under imperfect maintenance. IEEE Trans. Reliab., 2010, 592356-367.

[27] CassadyC.R., MurdockW.P.Jr., PohlE.A.. Selective maintenance for support equipment involving multiple maintenance actions. Eur. J. Oper. Res., 2001, 1292252-258.

[28] SchneiderK., CassadyC.R.. Evaluation and comparison of alternative fleet-level selective maintenance models. Reliab. Eng. Syst. Saf., 2015, 134: 178-187.

[29] ZhangL., ChenX., KhatabA., AnY., FengX.. Joint optimization of selective maintenance and repairpersons assignment problem for mission-oriented systems operating under s-dependent competing risks. Reliab. Eng. Syst. Saf., 2024, 242. 109796

[30] AchamrahF.E., AttajerA.. Multi-objective reinforcement learning-based framework for solving selective maintenance problems in reconfigurable cyber-physical manufacturing systems. Int. J. Prod. Res., 2024, 62103460-3482.

[31] PandeyM., ZuoM.J., MoghaddassR.. Selective maintenance modeling for a multistate system with multistate components under imperfect maintenance. IIE Trans., 2013, 45111221-1234.

[32] KhatabA., DialloC., VenkatadriU., LiuZ., AghezzafE.-H.. Optimization of the joint selective maintenance and repairperson assignment problem under imperfect maintenance. Comput. Ind. Eng., 2018, 125: 413-422.

[33] DialloC., VenkatadriU., KhatabA., LiuZ., AghezzafE.-H.. Optimal joint selective imperfect maintenance and multiple repairpersons assignment strategy for complex multicomponent systems. Int. J. Prod. Res., 2019, 57134098-4117.

[34] Al-JabouriH., SaifA., DialloC., KhatabA.. Distributionally-robust chance-constrained optimization of selective maintenance under uncertain repair duration. Expert Syst. Appl., 2024, 239. 122303

[35] DuanC., DengC., GharaeiA., WuJ., WangB.. Selective maintenance scheduling under stochastic maintenance quality with multiple maintenance actions. Int. J. Prod. Res., 2018, 56237160-7178.

[36] A. Amjadian, R. O’Neil, A. Khatab, J. Chen, U. Venkatadri, C. Diallo, Optimising resource-constrained fleet selective maintenance with asynchronous maintenance breaks. Int. J. Prod. Res., 1–23 (2024)

[37] O’NeilR., KhatabA., DialloC., VenkatadriU.. Optimal joint maintenance and orienteering strategy for complex mission-oriented systems: a case study in offshore wind energy. Comput. Oper. Res., 2023, 149. 106020

[38] Al-JabouriH., SaifA., KhatabA., DialloC., VenkatadriU.. Selective maintenance optimization: a condensed critical review and future research directions. IFAC-PapersOnLine, 2022, 55101213-1218.

[39] Al-JabouriH., SaifA., KhatabA., DialloC., VenkatadriU.. A critical review of selective maintenance for mission-oriented systems: challenges and a roadmap for novel contributions. Int. J. Prod. Res., 2024, 62134980-5015.

[40] KhatabA., DialloC., AghezzafE.-H., VenkatadriU.. Condition-based selective maintenance for stochastically degrading multi-component systems under periodic inspection and imperfect maintenance. Proc. Inst. Mech. Eng., Part O: J. Risk Reliab., 2018, 2324447-463

[41] ChenZ., HeY., ZhaoY., HanX., LiuF., ZhouD., WangW.. Mission reliability-oriented selective maintenance optimization for intelligent multistate manufacturing systems with uncertain maintenance quality. IEEE Access, 2019, 7: 109804-109816.

[42] HesabiH., NourelfathM., HajjiA.. A deep learning predictive model for selective maintenance optimization. Reliab. Eng. Syst. Saf., 2022, 219. 108191

[43] ShahrakiA.F., YadavO.P., VogiatzisC.. Selective maintenance optimization for multi-state systems considering stochastically dependent components and stochastic imperfect maintenance actions. Reliab. Eng. Syst. Saf., 2020, 196. 106738

[44] ChoA.D., CarrascoR.A., RuzG.A.. Improving prescriptive maintenance by incorporating post-prognostic information through chance constraints. IEEE Access, 2022, 10: 55924-55932.

[45] Q. Zhang, Y. Liu, B. Zhang, H.-Z. Huang, Selective maintenance optimization under limited maintenance capacities: a machine learning-enhanced approximate dynamic programming. IEEE Trans. Reliab. (2024)

[46] XuY., PiD., WuZ., ChenJ., ZioE.. Hybrid discrete differential evolution and deep q-network for multimission selective maintenance. IEEE Trans. Reliab., 2021, 7141501-1512.

[47] GravesA., LiwickiM., FernándezS., BertolamiR., BunkeH., SchmidhuberJ.. A novel connectionist system for unconstrained handwriting recognition. IEEE Trans. Pattern Anal. Mach. Intell., 2008, 315855-868.

[48] ZhaoR., YanR., WangJ., MaoK.. Learning to monitor machine health with convolutional bi-directional lstm networks. Sensors, 2017, 172. 273

[49] YuW., KimI.Y., MechefskeC.. Remaining useful life estimation using a bidirectional recurrent neural network based autoencoder scheme. Mech. Syst. Signal Process., 2019, 129: 764-780.

[50] ArulmozhiG.. Exact equation and an algorithm for reliability evaluation of k-out-of-n: G system. Reliab. Eng. Syst. Saf., 2002, 78287-91.

[51] VaartA.W.Asymptotic Statistics, 2000CambridgeCambridge University Press3

[52] YuW., KimI.Y., MechefskeC.. An improved similarity-based prognostic algorithm for rul estimation using an rnn autoencoder scheme. Reliab. Eng. Syst. Saf., 2020, 199. 106926

[53] PengC., ChenY., ChenQ., TangZ., LiL., GuiW.. A remaining useful life prognosis of turbofan engine using temporal and spatial feature fusion. Sensors, 2021, 212. 418

[54] ZengJ., LiangZ.. A dynamic predictive maintenance approach using probabilistic deep learning for a fleet of multi-component systems. Reliab. Eng. Syst. Saf., 2023, 238. 109456

[55] LeeJ., MiticiM.. Deep reinforcement learning for predictive aircraft maintenance using probabilistic remaining-useful-life prognostics. Reliab. Eng. Syst. Saf., 2023, 230. 108908

[56] ZhangQ., LiuQ., YeQ.. An attention-based temporal convolutional network method for predicting remaining useful life of aero-engine. Eng. Appl. Artif. Intell., 2024, 127. 107241

[57] XiangF., ZhangY., ZhangS., WangZ., QiuL., ChoiJ.-H.. Bayesian gated-transformer model for risk-aware prediction of aero-engine remaining useful life. Expert Syst. Appl., 2024, 238. 121859

[58] WangL., ZhuZ., ZhaoX.. Dynamic predictive maintenance strategy for system remaining useful life prediction via deep learning ensemble method. Reliab. Eng. Syst. Saf., 2024, 245. 110012

[59] K. Vuckovic, S. Prakash, Remaining useful life prediction using Gaussian process regression model. Annu. Conf. PHM Soc., 14 (2022)

[60] WuX., LiuZ., WangL.. Spatio-temporal degradation model with graph neural network and structured state space model for remaining useful life prediction. Reliab. Eng. Syst. Saf., 2025, 256. 110770

[61] HagmeyerS., ZeilerP.. A comparative study on methods for fusing data-driven and physics-based models for hybrid remaining useful life prediction of air filters. IEEE Access, 2023, 11: 35737-35753.

Funding

Natural Sciences and Engineering Research Council of Canada

Nova Scotia Graduate Scholarship (NSGS)

RIGHTS & PERMISSIONS

The Author(s)