Our website uses cookies. By continuing to use this site, you agree to our use of cookies in accordance with this notice and the Terms of Use.
For authors Editor Contacts News Announcement
Submit manuscript
Journal of Applied Research
Journals / Journal of Applied Research / Issue 5 / Cost-effectiveness of ai/ml application during the transition from regulated maintenance to actual maintenance
UDC 004.85

COST-EFFECTIVENESS OF AI/ML APPLICATION DURING THE TRANSITION FROM REGULATED MAINTENANCE TO ACTUAL MAINTENANCE Cost-effectiveness of ai/ml application during the transition from regulated maintenance to actual maintenance

Published in Journal of Applied Research · Issue 5 · ELocator: 258-264 · Rubric: ECONOMICS. ECONOMIC SCIENCES.
DOI: https://doi.org/10.34755/IROK.2026.89.83.035
Received: 10.08.2026 Accepted: 12.08.2026 Published: 12.08.2026
Authors
OL
Oleg A. Sokolov
AL
Aleksandr A. Andreev
AR
Arseniy Y. Hrustalev
ABSTRACT
The article focuses on finding an economic justification for abandoning outdated maintenance schedules in favor of an AI and ML-based CBM approach. Why is this important? Traditional maintenance strategies have long been criticized for the fact that huge amounts of money (about 30-40% of the budget) are simply wasted due to "scheduled" repairs, while every fifth unplanned failure adds headaches and unplanned expenses to the enterprise. To minimize these losses, it is proposed to use a combination of the LSTM neural network and XGBoost gradient boosting. The tests proved the viability of the approach: maintenance costs were reduced by 24%, equipment downtime by 31%, and the predictability of accidents was increased to the level of 0.89 (Precision). To top it all off, practical guidelines are given for the implementation of smart CBM systems in order to really reduce the cost of operation.
KEYWORDS
artificial intelligence, machine learning, actual condition maintenance, CBM, residual resource forecasting, maintenance and repair, industrial operation, neural networks, gradient boosting, digitalization of production
Text References
Text
Text (PDF)
Read Download

1. Fedotova G. A. Sovremennye problemy standartizacii nadezhnosti v tehnike / G. A. Fedotova, I. A. Sher // Vestnik Irkutskogo gosudarstvennogo tehnicheskogo universiteta. – 2018. – T. 22, № 2(133). – S. 142-152.

2. Oksak I. V. Optimizaciya processov TOIR, strategicheskie podhody i metody povysheniya effektivnosti / I. V. Oksak // Ustoychivoe razvitie Rossii - 2025 : Sbornik statey IV Vserossiyskoy nauchno-prakticheskoy konferencii, Petrozavodsk, 28 avgusta 2025 goda. – Petrozavodsk: Mezhdunarodnyy centr nauchnogo partnerstva "Novaya Nauka" (IP Ivanovskaya I.I.), 2025. – S. 80-99

3. Grishin A. A. Issledovanie effektivnosti algoritmov klassifikacii, ispol'zuyuschih derev'ya resheniy, dlya ekonomicheskogo prognozirovaniya na osnove aposteriornyh dannyh / A. A. Grishin, S. P. Stroev // Vestnik Tehnologicheskogo universiteta. – 2018. – T. 21, № 9. – S. 180-186

4. Korablev Yu. A. Prognozirovanie ostatochnogo sroka poleznogo ispol'zovaniya tehnologicheskogo oborudovaniya metodom glubokogo obucheniya LSTM / Yu. A. Korablev // Izvestiya YuFU. Tehnicheskie nauki. – 2025. – № 5(247). – S. 277-288

5. Sovremennoe sostoyanie metodov mashinnogo obucheniya v prilozheniyah obrabotki odnomernyh signalov / B. I. Valeev, A. I. Sadykova, A. N. Salahutdinova [i dr.] // Elektronika, fotonika i kiberfizicheskie sistemy. – 2022. – T. 2, № 3. – S. 86-95

PDF JATS XML
The contact persons
Deputy Editor-in-Chief
Ol'ga Shatalova
Chief Editor
Raiganat Babaeva
285
downloads
Indexing information will be added later
© jomeam.ru
Agreement Personal data protection and processing policy Support Powered by Editorum, 2026 Instructions
Login or Create
* Forgot your password?