UDC: 656.1
https://doi.org/10.25198/2077-7175-2026-3-82
EDN: SUQNTE


viewing

ANALYSIS AND PROSPECTS OF ADAPTATION OF EXPERIENCE IN APPLYING MACHINE LEARNING IN THE SYSTEM OF TECHNICAL MAINTENANCE AND REPAIR OF CITY BUSES

A. N. Strelkov1, I. M. Blyankinshtein2
Saint-Petersburg State University of Architecture and Civil Engineering, Saint-Petersburg, Russia
1 e-mail: tema.strelkoff2016@yandex.ru
2 e-mail: blyankinshtein@mail.ru

Abstract. In the context of accelerated digitalization of automotive transportation, machine learning methods are forming the foundation for transitioning from outdated reactive and preventive maintenance strategies to modern predictive approaches. The aim of this work is to systematize information and analyze the experience of applying machine learning methods in the maintenance and repair systems of vehicles, with the subsequent identification of directions for adapting these methods to the specific challenges of urban bus fleets. The research methodology is based on the analysis of scientific and technical literature using formalized selection criteria, a comparative analysis of the efficiency metrics of ML architectures, and the synthesis of the obtained data in order to assess their applicability and develop a conceptual framework for adapting predictive models to the operating conditions of city buses. A systematic review of scientific publications and industrial solutions on the application of ML models for diagnosing and forecasting the technical condition of vehicles during the 2020–2025 period has been conducted. Particular attention is paid to a comparative assessment of the effectiveness of classical algorithms (Random Forest, XGBoost, LightGBM) and deep architectures (LSTM, CNN, hybrid models) when processing data obtained from onboard systems. It has been established that hybrid CNN-LSTM models demonstrate the highest diagnostic accuracy (up to 99.02%) in predicting electric motor failures, outperforming both classical methods and individual deep architectures. The economic benefits of implementing predictive systems include a reduction in operational costs by 10–40%, a decrease in vehicle downtime by up to 50%, and an improvement in overall fleet reliability. However, the analysis has revealed key limitations to the practical application of ML solutions: imbalanced training datasets due to the rarity of failures, high sensor inaccuracies (exceeding 40% for certain parameters), interpretability issues associated with deep learning “black boxes,” and integration challenges with existing ERP/CMMS systems. The necessity of developing Explainable AI (XAI) methods to increase specialists’ trust in predictions is substantiated. The scientific novelty of the work lies in the development of a conceptual framework for adapting ML models to the cyclic operation mode of buses, which includes incorporating geospatial data, road conditions, and requirements for forecast interpretability. Based on this, directions for future research are proposed to enhance the accuracy of residual life prediction for critical bus components.

Key words: machine learning, predictive maintenance, maintenance and repair, data-driven diagnostics, automotive transport.

Cite as: Strelkov, A. N., Blyankinshtein, I. M. (2026) [Analysis and Prospects of Adaptation of Experience in Applying Machine Learning in the System of Technical Maintenance and Repair of City Buses]. Intellekt. Innovacii. Investicii [Intellect. Innovations. Investments]. Vol. 3, pp. 82–96. – https://doi.org/10.25198/2077-7175-2026-3-82.


References

  1. Safiullin, R. N., et al. (2025) [Models and algorithms for predicting the technical condition of vehicles based on fuzzy logic methods]. Vestnik NCBZhD [Bulletin of the National Centralized Railway Transport]. Vol. 2 (64), pp. 138–152. (In Russ.).
  2. Popov, A. V., Safiullin, R. N. (2023) [Study of the influence of charging mode parameters on the energy performance of lithium-ion batteries of vehicles]. Izvestiya Mezhdunarodnoj akademii agrarnogo obrazovaniya [Bulletin of the International Academy of Agrarian Education]. Vol. 67, pp. 98–103. (In Russ.).
  3. Kamlyuk, V. V., et al. (2024) [Systems analysis of the implementation of digital technologies to ensure the safety and efficiency of vehicles]. Transport. Vzglyad v budushchee – TFV-24: Sbornik nauchnyh statej mezhdunarodnoj nauchno-prakticheskoj konferencii, Sankt-Peterburg, 07–08 noyabrya 2024 goda. – Sankt-Peterburg: Sankt-Peterburgskij gornyj universitet imperatricy Ekateriny II [Transport. Future Vision – TFV-24: Collection of scientific articles from the international scientific and practical conference, St. Petersburg, November 7–8, 2024. St. Petersburg: Empress Catherine II Saint-Petersburg Mining University], pp. 23–29. (In Russ.).
  4. Fedotov, M. V., Grachev, V. V. (2021) [Predictive analytics of the technical condition of diesel locomotive systems using neural network forecast models]. Byulleten’ rezul’tatov nauchnyh issledovanij [Bulletin of scientific research results]. Vol. 3, pp. 102–114. – https://doi.org/10.20295/2223-9987-2021-3-102-114. (In Russ.).
  5. A Comparative Analysis of Advanced Machine Learning Models for Predictive Maintenance in Modern Manufacturing. IoT Digital Twin PLM. Available at: https://iotdigitaltwinplm.com/a-comparative-analysis-of-advanced-machine-learning-models-for-predictive-maintenance-in-modern-manufacturing/ (accessed: 01.11.2025).
  6. Al-Zeyadi, M., et al. (2020) Deep Learning Towards Intelligent Vehicle Fault Diagnosis. 2020 International Joint Conference on Neural Networks (IJCNN). – https://doi.org/10.1109/IJCNN48605.2020.9206972. (In Eng.).
  7. Amellal, A., et al. (2023) Improving Lead Time Forecasting and Anomaly Detection for Automotive Spare Parts with A Combined CNN-LSTM. Approach. Operations and Supply Chain Management: An International Journal. – Vol. 16. No. 2, pp. 265–278. – https://doi.org/10.31387/oscm0530388. (In Eng.).
  8. Aydın, C., Evrentuğ, B. (2025) Evaluation of predictive maintenance efficiency with the comparison of machine learning models in machining production process in brake industry. PeerJ Computer Science. Vol. 11, e2999. – https://doi.org/10.7717/peerj-cs.2999. (In Eng.).
  9. Benhanifia, A., et al. (2025) Systematic review of predictive maintenance practices in the manufacturing sector. Intelligent Systems with Applications. Vol. 26, 200501. – https://doi.org/10.1016/j.iswa.2025.200501. (In Eng.).
  10. Bickelhaupt, S., et al. (2023) Challenges and Opportunities of Future Vehicle Diagnostics in Software-Defined Vehicles. SAE. – https://doi.org/10.4271/2023-01-0847. (In Eng.).
  11. Boucerredj, L, Benalia, N. (2025) A comparative study of machine learning classifiers for intelligent fault diagnosis of electric vehicles based on FMECA data. Advances in Mechanical Engineering. – https://doi.org/10.1177/16878132251342413. (In Eng.).
  12. Chen, F., et al. (2025) Collaborative multiview time series modeling for vehicle maintenance demand prediction. Scientific Reports. Vol. 15, 13058. – https://doi.org/10.1038/s41598-025-96720-1. (In Eng.).
  13. Etukudoh, E. A. (2024) Theoretical framworks of eopfm predictive maintenance (ecopfm) predictive maintenance system. Engineering Science & Technology Journal. Vol. 5. No. 3, pp. 913–923. – https://doi.org/10.51594/estj.v5i3.946. (In Eng.).
  14. Gong, C-S. A., et al. (2022) How to Implement Automotive Fault Diagnosis Using Artificial Intelligence Scheme. Micromachines. Vol. 13. No. 9, 1380. – https://doi.org/10.3390/mi13091380. (In Eng.).
  15. Guo, C. (2023) Fault diagnosis of automobile drive based on a novel deep neural network. Energy Harvesting and Systems. Vol. 11. No. 1, pp. 20230049. – https://doi.org/10.1515/ehs-2023-0049. (In Eng.).
  16. Hazem, A., Alaa, Y., Solayman, M. (2025) Offline Predictive Maintenance for Automotive Engines Using Machine Learning. 2025 Intelligent Methods, Systems, and Applications (IMSA), Giza, Egypt, pp. 610–615. – https://doi.org/10.1109/IMSA65733.2025.11167090. (In Eng.).
  17. Huang, K., Wang, J. (2023) Short-term auto parts demand forecasting based on EEMD–CNN–BiLSTM– Attention–combination model. Journal of Intelligent & Fuzzy Systems: Applications in Engineering and Technology. Vol. 45. No. 4, pp. 5449–5465. – https://doi.org/10.3233/JIFS-224222. (In Eng.).
  18. Kumar, R. S., et al. (2025) Hybrid machine learning framework for predictive maintenance and anomaly detection in lithium-ion batteries using enhanced random forest. Scientific Reports. Vol. 15, 6243. – https://doi.org/10.1038/s41598-025-90810-w. (In Eng.).
  19. Li, P., et al. (2020) A Deep Learning Approach to Detect Real-Time Vehicle Maneuvers Based on Smartphone Sensors. IEEE Transactions on Intelligent Transportation Systems. Vol. 23. No. 4. – https://doi.org/10.1109/TITS.2020.3032055. (In Eng.).
  20. Mahiyudin, G., Hussain, M., Dewi, D. D. (2025) A Comprehensive Study on Predicting the Need for Vehicle Maintenance Using Machine Learning. Engineering Proceedings. Vol. 107. No. 1, p. 89. – https://doi.org/10.3390/engproc2025107089. (In Eng.).
  21. Meenakshi, M., Rainu, N. (2023) A framework on driving behavior and pattern using On-Board diagnostics (OBD-II) tool. Materials Today: Proceedings. Vol. 80. No. 3, pp. 3762–3768. – https://doi.org/10.1016/j.matpr.2021.07.376. (In Eng.).
  22. Melnik, Yu. Machine failure prediction using machine learning: Why it is beneficial. InData Labs. Available at: https://indatalabs.com/blog/machine-failure-prediction-machine-learning (accessed: 01.11.2025).
  23. Mishra, D., et al. (2024) Fault detection and diagnosis of electric vehicles using artificial intelligence. International Journal of Applied Power Engineering (IJAPE). Vol. 13. No. 3, p. 653. – https://doi.org/10.11591/ijape. v13.i3.pp653-660. (In Eng.).
  24. Muthukumar, G., Jyosna, P. (2024) CNN-LSTM Hybrid Deep Learning Model for Remaining Useful Life Estimation. – https://doi.org/10.48550/arXiv.2412.15998. (In Eng.).
  25. National Academies of Sciences, Engineering, and Medicine (2024) Implementing and Leveraging Machine Learning at State Departments of Transportation. Washington, DC: The National Academies Press. – https://doi.org/10.17226/27902. (In Eng.).
  26. Pavlopoulos, J., et al. (2023) Automotive fault nowcasting with machine learning and natural language processing. Vol. 113, pp. 843–861. – https://doi.org/10.1007/s10994-023-06398-7. (In Eng.).
  27. Purohit, Sh., Govindarasu, M. (2022). ML-based Anomaly Detection for Intra-Vehicular CAN-bus Networks. IEEE International Conference on Cyber Security and Resilience, pp. 233–238. – https://doi.org/10.1109/CSR54599.2022.9850292. (In Eng.).
  28. Quan, R., Zhang, J., Feng, Z. (2024) Remote Fault Diagnosis for the Powertrain System of Fuel Cell Vehicles Based on Random Forest Optimized with a Genetic Algorithm. Sensors. Vol. 24. No. 4, pp. 1138. – https://doi.org/10.3390/s24041138. (In Eng.).
  29. Sánchez Torres, N. N., et al. (2025) Fault Diagnosis in Internal Combustion Engines Using Artificial Intelligence Predictive Models. Applied System Innovation. Vol. 8. No. 5, pp. 147. – https://doi.org/10.3390/asi8050147. (In Eng.).
  30. Saraswat, A., et al. Predictive Maintenance of Automotive Engines Using Machine Learning and Deep Learning Techniques. IJIRT. Vol. 12. No. 1, pp. 446–455.
  31. Sekar, M. (2025) Machine learning based fault detection and classification for predictive maintenance of gas turbine engines: a comprehensive benchmarking analysis on various models. Aircraft Engineering and Aerospace Technology: An International Journal. – https://doi.org/10.1108/AEAT-04-2025-0160. (In Eng.).
  32. Shah, Ch. (2024) Machine Learning Algorithms for Predictive Maintenance in Autonomous Vehicles. International Journal of Engineering and Computer Science. Vol. 13. No. 1, pp. 26015–26032. – https://doi.org/10.18535/ijecs/v13i01.4786. (In Eng.).
  33. Stewart, A. (2024) Predictive Vehicle Maintenance: Complete 2025 Guide to AI Car Care // dialzara. Available at: https://dialzara.com/blog/ai-predictive-maintenance-in-automotive-guide (accessed: 01.11.2025).
  34. Taheri-Garavand, A., et al. (2022) Application of artificial neural networks for the prediction of performance and exhaust emissions in IC engine using biodiesel-diesel blends containing quantum dot based on carbon doped. Energy Conversion and Management: X. Vol. 16, pp. 100304. – https://doi.org/10.1016/j.ecmx.2022.100304. (In Eng.).
  35. Tunio, N. A., et al. (2025) Performance Comparison Between Deep Learning Models for Fault Classification in Transmission Lines Using Time Series Data. Energy Science & Engineering. Vol. 13. No. 5, pp. 2330–2351. – https://doi.org/10.1002/ese3.70033. (In Eng.).
  36. Wang, J., Chen, J. (2024) Rapid diagnosis of power battery faults in new energy vehicles based on improved boosting algorithm and big data. Energy Informatics. Vol. 7, pp. 139. – https://doi.org/10.1186/s42162-024-00439-8. (In Eng.).
  37. Xin, Yu., et al. (2024) Machine learning based mechanical fault diagnosis and detection methods: a systematic review. Measurement Science and Technology. Vol. 36, pp. 012004. – https://doi.org/10.1088/1361-6501/ad8cf6. (In Eng.).
  38. Yang, D., et al. (2024) A Multivariate Time Series Prediction Method for Automotive Controller Area Network Bus Data. Electronics. Vol. 13. No. 14, pp. 2707. – https://doi.org/10.3390/electronics13142707. (In Eng.).
  39. Yang, Y., Wang, H. (2025). Random Forest-Based Machine Failure Prediction: A Performance Comparison. Applied Sciences. Vol. 15. No. 16, pp. 8841. – https://doi.org/10.3390/app15168841. (In Eng.).

In order for us to provide you with high-quality services, we use cookies that are stored on your computer (location information; ip-address; type, language, OS and browser version; device type and screen resolution; source from which you came to site the user; what pages the user opens and what buttons the user clicks; the same information is used to process statistical data on the use of the site through the Yandex.Metrika Internet service). By clicking the "I agree" button, you confirm that you are informed about the use of cookies on our website. You can disable cookies in your browser settings.