Predicting Failure using Machine Learning and Statistical Based Method: a Production Machine Case Study

Effi Latiffianti, Stefanus Eko Wiratno, Samuel Aditya Christianta

Abstract


This research investigates the applicability of failure detection models based on machine learning and statistical approaches to reduce unplanned downtime in a food production company. Sensor data is utilized to for identifying early failure symptoms. To capture temporal and sequential dependencies in time-series data, we employ one of potential network based method so called the Long Short Term Memory (LSTM) Autoencoder. Furthermore, we contrast the performance of the result with the traditional statistical method, the multivariate Exponentially Weighted Moving Average (EWMA). While both models successfully detected all failures, LSTM-AE demonstrated superior performance by reducing false alarms and providing true alarms with a longer time-to-failure. The findings highlight the potential of leveraging limited data for failure prediction, demonstrating the effectiveness of both models in detecting anomalies while emphasizing their role in enhancing productivity through early failure detection.


Keywords


Anomaly detection; failure; fault; Long Short Term Memory Autoencoder; Multivariate Exponentially Weighted Moving Average

Full Text:

PDF

References


Infraspeak. Maintenance statistics and trends 2025. Infraspeak Team; 2024.

Ensafi M, Thabet W. Challenges and Gaps in Facility Maintenance Practices. In: Leathem T, Perrenoud A, Collins W, editors. ASC 2021. 57th Annual Associated Schools of Construction International Conference, vol. 2 of EPiC Series in Built Environment EasyChair; 2021. p. 237–245. /publications/paper/tZ57.

Latiffianti E, Sheng S, Ding Y. Wind Turbine Gearbox Failure Detection Through Cumulative Sum of Multivariate Time Series Data. Frontiers in Energy Research 2022;10. https://www.frontiersin.org/journals/energy-research/articles/10.3389/fenrg.2022.904622.

Khan PW, Yeun CY, Byun YC. Fault detection of wind turbines using SCADA data and genetic algorithm-based ensemble learning. Engineering Failure Analysis 2023;148:107209. https://www.sciencedirect.com/science/article/pii/S1350630723001632.

Meyer A. 1 - SCADA-based fault detection in wind turbines: data-driven techniques and applications. In: Garcia Marquez FP, Papaelias M, Junior VLJ, editors. Non-Destructive Testing and Condition Monitoring Techniques in Wind Energy Wind Energy Engineering, Academic Press; 2023.p. 1–13. https://www.sciencedirect.com/science/article/pii/B9780323996662000010.

Chen J, Li J, Chen W, Wang Y, Jiang T. Anomaly detection for wind turbines based on the reconstruction of condition parameters using stacked denoising autoencoders. Renewable Energy 2020;147(P1):1469–1480. https://EconPapers.repec.org/RePEc:eee:renene:v:147:y:2020:i:p1:p:1469-1480.

Hermansa M, Kozielski M, Michalak M, Szczyrba K, Wróbel Sikora M. Sensor-Based Predictive Maintenance with Reduction of False Alarms—A Case Study in Heavy Industry. Sensors 2022;22(1). https://www.mdpi.com/1424-8220/22/1/226.

Yadav DK, Kaushik A, Yadav N. Predicting machine failures using machine learning and deep learning algorithms. Sustainable Manufacturing and Service Economics 2024;3:100029. https://www.sciencedirect.com/science/article/pii/S2667344424000124.

Carrasco J, López D, Aguilera-Martos I, García-Gil D, Markova I, García-Barzana M, et al. Anomaly detection in predictive maintenance: A new evaluation framework for temporal unsupervised anomaly detection algorithms. Neurocomputing 2021;462:440–452. https://www.sciencedirect.com/science/article/pii/S0925231221011826.

Di Persio L, Garbelli M, Wallbaum K. Forward-Looking Volatility Estimation for Risk-Managed Investment Strategies during the COVID-19 Crisis. Risks 2021;9(2). https://www.mdpi.com/2227-9091/9/2/33.

Boujamza A, Elhaq SL. Attention-based LSTM for Remaining Useful Life Estimation of Aircraft Engines. IFAC-PapersOnLine 2022;https://api.semanticscholar.org/CorpusID:251366314.

Zheng S, Ristovski K, Farahat A, Gupta C. Long Short-Term Memory Network for Remaining Useful Life estimation. In: 2017 IEEE International Conference on Prognostics and Health Management (ICPHM); 2017. p. 88–95.

Haipeng C, Jie Y, Wenxing F. Two-layer fault diagnosis model of aircraft based on LSTM. Aerospace Science and Technology 2025;158:109756. https://www.sciencedirect.com/science/article/pii/S127096382400885X.

Du W, Zhang J, Meng G, Zhang H. Aero-Engine Fault Detection with an LSTM Auto-Encoder Combined with a Self- Attention Mechanism. Machines 2024;12(12). https://www.mdpi.com/2075-1702/12/12/879.

Yang T, Teng J, Li C, Feng Y. Wind turbine fault detection and diagnosis using LSTM neural network. In: 2020 39th Chinese Control Conference (CCC); 2020. p. 4042–4047.

Qin S, Tao J, Zhao Z. Fault diagnosis of wind turbine pitch system based on LSTM with multi-channel attention mechanism. Energy Reports 2023;10:4087–4096. https://www.sciencedirect.com/science/article/pii/S2352484723015081.

Xiang L, Wang P, Yang X, Hu A, Su H. Fault detection of wind turbine based on SCADA data analysis using CNN and LSTM with attention mechanism. Measurement 2021;175:109094. https://www.sciencedirect.com/science/article/pii/S026322412100124X.

Kamat P, Sugandhi R. Anomaly Detection for Predictive Maintenance in Industry 4.0 – A survey. In: E3S Web Conf., vol. 170; 2020. .

Gong L, Chen Y. Machine Learning-enhanced loT and Wireless Sensor Networks for predictive analysis and maintenance in wind turbine systems. International Journal of Intelligent Networks 2024;5:133–144. https://www.sciencedirect.com/science/article/pii/S2666603024000083.

Guo J, Liu G, Zuo Y, Wu J. An Anomaly Detection Framework Based on Autoencoder and Nearest Neighbor. In: 2018 15th International Conference on Service Systems and Service Management (ICSSSM); 2018. p. 1–6.

Chalapathy R, Chawla S. Deep learning for anomaly detection: A survey. arXiv 2019;https://arxiv.org/abs/1901.03407.

Maleki S, Maleki S, Jennings NR. Unsupervised anomaly detection with LSTM autoencoders using statistical data-filtering. Applied Soft Computing 2021;108:107443. https://www.sciencedirect.com/science/article/pii/S1568494621003665.

Demidova G, Privalov D, Semenov D, Lukichev D, Liu Z, Anuchin A. Fault Detection in Electric Drives Based on LSTM Autoencoder Model Machine Learning Approach. In: 2024 IEEE 25th International Conference of Young Professionals in Electron Devices and Materials (EDM); 2024. p. 1670–1675.

Zeng L, Jin Q, Lin Z, Zheng C, Wu Y, Wu X, et al. Dual-attention LSTM autoencoder for fault detection in industrial complex dynamic processes. Process Safety and Environmental Protection 2024;185:1145–1159. https://www.sciencedirect.com/science/article/pii/S0957582024001721.

Lachekhab F, Benzaoui M, Tadjer SA, Bensmaine A, Hamma H. LSTM-Autoencoder Deep Learning Model for Anomaly Detection in Electric Motor. Energies 2024;17(10). https://www.mdpi.com/1996-1073/17/10/2340.

Montgomery DC. Introduction To Statistical Quality Control, 4Th Ed. Wiley India Pvt. Limited; 2007. https://books. google.co.id/books?id=rcu71DvkjX0C.

Kadri F, Harrou F, Chaabane S, Sun Y, Tahon C. Seasonal ARMA-based SPC charts for anomaly detection: Application to emergency department systems. Neurocomputing 2016;173:2102–2114. https://www.sciencedirect.com/science/article/pii/S0925231215014654.

Woodall WH, Montgomery DC. Research Issues and Ideas in Statistical Process Control. Journal of Quality Technology 1999;31(4):376–386.

Bersimis S, Psarakis S, Panaretos J. Multivariate Statistical Process Control Charts: An Overview. Quality and Reliability Engineering International 2007 08;23:517 – 543.

Lee JM, Yoo C, Lee IB. Efficient fault detection using multivariate exponentially weighted moving average and independent component analysis. In: Chen B, Westerberg AW, editors. Process Systems Engineering 2003, 8th International Symposium on Process Systems Engineering, vol. 15 of Computer Aided Chemical Engineering Elsevier; 2003.p. 916–921. https:

//www.sciencedirect.com/science/article/pii/S1570794603804248.

Harrou F, Nounou M, Nounou H. A statistical fault detection strategy using PCA based EWMA control schemes. In: 2013 9th Asian Control Conference (ASCC); 2013. p. 1–4.

Marais HL, Zaccaria V, Odlare M. Comparing statistical process control charts for fault detection in wastewater treatment. Water Science & Technology 2022;85(4):1250–1262.

Baklouti I, Mansouri M, Nounou H, Ben Slima M, Ben Hamida A. Fault detection in a wastewater treatment plant. In: 2017 International Conference on Advanced Technologies for Signal and Image Processing (ATSIP); 2017. p. 1–5.

Su J, Zeng Z, Tang C, Liu Z, Li T. A Photovoltaic Fault Diagnosis Method Integrating Photovoltaic Power Prediction and EWMA Control Chart. Energies 2024;17(17). https://www.mdpi.com/1996-1073/17/17/4263.

Mansouri M, Al-khazraji A, Hajji M, Harkat MF, Nounou H, Nounou M. Wavelet optimized EWMA for fault detection and application to photovoltaic systems. Solar Energy 2018;167:125–136. https://www.sciencedirect.com/science/article/pii/S0038092X18303050.

Li H, Deng J, Yuan S, Feng P, Arachchige DDK. Monitoring and Identifying Wind Turbine Generator Bearing Faults Using Deep Belief Network and EWMA Control Charts. Frontiers in Energy Research 2021;9. https://www.frontiersin.org/journals/energy-research/articles/10.3389/fenrg.2021.799039.

Lucas JM, Saccucci MS. Exponentially Weighted Moving Average Control Schemes: Properties and Enhancements. Technometrics 1990;32(1):1–12. https://www.tandfonline.com/doi/abs/10.1080/00401706.1990.10484583.

Jurjevčič B, Senegačnik A, Kuštrin I. A Surveillance of Direct-Firing System for Pulverized-Coal Using Statistically Treated Signals from Intrusive Electrostatic Sensors. Strojniški vestnik - Journal of Mechanical Engineering 2017;63(4):265–274. https://www.sv-jme.eu/article/

a-surveillance-of-direct-firing-system-for-pulverized-coal-using-statistically-treated-signals-from-intrusive-electrostatic-sensors/.

de Amorim LBV, Cavalcanti GDC, Cruz RMO. The choice of scaling technique matters for classification performance. Applied Soft Computing 2023;133:109924. https://www.sciencedirect.com/science/article/pii/S1568494622009735.




DOI: http://dx.doi.org/10.12962%2Fj20882033.v36i1.22501

Refbacks

  • There are currently no refbacks.


Creative Commons License

IPTEK Journal of Science and Technology by Lembaga Penelitian dan Pengabdian kepada Masyarakat, ITS is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Based on a work at https://iptek.its.ac.id/index.php/jts.