A.U. Chesalov1

1 CEO Atlansis Software (Tver, Russia)
1 achesalov@mail.ru

A distinctive feature of modern industrial automated systems for predictive or prescriptive maintenance of multi-stage technological processes, decision support systems, analytical, expert, and other production systems is the presence of a large number of uncertainty factors when collecting and analyzing data from a large number of industrial Internet of Things sensors which are operated under various external and internal environmental conditions. One common approach to reducing uncertainty in the collected data and increasing the degree of confidence in it may be to combine methods of intelligent processing of fuzzy information, neural network methods, and Dempster–Shafer evidence theory. In these conditions, it is necessary to develop new models, algorithms, and programs that can solve the problems of reducing data uncertainty factors to improve the performance of predictive models, prepare expert opinions, and optimize maintenance schedules for industrial equipment based on them.

The aim of the research is to develop a mathematical model for reducing the uncertainty of data collected from devices (sensors, gateways, and others) from the industrial Internet of Things and processed in predictive maintenance systems of industrial enterprises based on Dempster–Shafer evidence theory, which will improve the accuracy of forecasts of the operating status of equipment in use, as well as make the necessary adjustments to the schedules and timetables for various types of work required to maintain the equipment in working order.

The results of the study are: a mathematical model for reducing uncertainty in predictive maintenance systems based on Dempster–Shafer evidence theory; an algorithm and its software implementation in the Python programming language. The developed model does not require prior information about the hypothesis. It supports the simulation of conflicts between different data sources. It is well suited for unreliable or ambiguous sensor readings. A distinctive feature of the program is its versatility, which provides the ability to work with a large number of sensors, add new equipment states, and integrate into the operation of industrial predictive maintenance systems.

Practical significance. The results of the research and work can be used to improve the performance of industrial automated predictive or prescriptive maintenance systems, in which multimodal data is collected from different types and kinds of industrial Internet of Things devices operated under various external and internal environmental conditions.

Chesalov A.U. A mathematical model for reducing uncertainty based on Dempster–Shafer theory at the level of data collection and aggregation in automated predictive maintenance systems. Dynamics of complex systems. 2025. V. 19. № 4. P. 62−74. DOI: 10.18127/j19997493-202504-07 (in Russian).

1. Palyux B.V., Chesalov A.Yu. Rol` sovremenny`x texnologij iskusstvennogo intellekta v sozdanii i razvitii avtomatizirovanny`x sistem prognoziruemogo i predpisy`vayushhego obsluzhivaniya v promy`shlennosti. Zhurnal Sovremennaya nauka: aktual`ny`e problemy` teorii i praktiki: seriya «Estestvenny`e i Texnicheskie nauki». 2025. № 5. S. 147–155. DOI 10.37882/2223-2966.2025.05.29
2. Palyux B.V., Chesalov A.Yu. Sovremenny`e podxody` k sozdaniyu avtomatizirovanny`x sistem prognoziruemogo obsluzhivaniya mnogostadijny`x texnologicheskix processov v promy`shlennosti. «Inzhiniring predpriyatij i upravlenie znaniyami» (IP&UZ – 2024): sb. nauch. tr. XXVII Rossijskoj nauch. konf. 28–29 noyabrya 2024 g. T. 1 / Pod nauch. red. Yu.F. Tel`nova. M.: FGBOU VO «RE`U im. G.V. Plexanova». 2024. S. 351–357.
3. Chesalov A.Yu. Primenenie prory`vny`x texnologij iskusstvennogo intellekta v promy`shlenny`x e`kosistemax Industrii 4.0. / Sb. materialov IX Sankt-Peterburgskogo mezhdunarodnogo e`konomicheskogo kongressa (SPE`K-2024) «Perspektivny`e integracionny`e processy` v mirovoj e`konomike: noopodxod» / Pod red. S.D. Bodrunova. T. 2. M.: INIR im. S.Yu. Vitte. 2024. S. 176–184.
4. Chesalov A.Yu. Tendencii razvitiya periferijnogo iskusstvennogo intellekta v avtomatizacii texnologicheskix processov. Avtomatizaciya v promy`shlennosti. 2025. № 7. S. 9–14.
5. Sy`tnik A.S. Prediktivnoe obsluzhivanie intellektual`ny`x texnicheskix ob``ektov / Mezhdunar. forum Kazan digital week – 2020. 2020. S. 414–419.
6. Nunes P., Santos J., Rocha E. Challenges in predictive maintenance–A review. CIRP Journal of Manufacturing Science and Technology. 2023. V. 40. P. 53–67.
7. Predictive maintenance: techniques and advantages. [E`lektronny`j resurs]. 2024. URL: https://www.mecalux.com/blog/predictive-maintenance (data obrashheniya: 23.05.2025).
8. Manufacturing: Analytics unleashes productivity and profitability. [E`lektronny`j resurs]. 2017. URL: https://www.mckinsey.com/ capabilities/operations/our-insights/manufacturing-analytics-unleashes-productivity-and-profitability/#/ (data obrashheniya: 03.06.2025).
9. Yadrovskaya M.V., Porksheyan M.V., Sinel`nikov A.A. Perspektivy` texnologii interneta veshhej. Advanced Engineering Research (Rostov-on-Don). 2021. T. 21. № 2. S. 207–217.
10. Markeeva A.V. Internet veshhej (IoT): vozmozhnosti i ugrozy` dlya sovremenny`x organizacij. Obshhestvo: sociologiya, psixologiya, pedagogika. 2016. № 2. S. 42–46.
11. Furstenau L.B. et al. Internet of things: Conceptual network structure, main challenges and future directions. Digital Communications and Networks. 2023. V. 9. № 3. P. 677–687.
12. Chernuxin A.V. i dr. Sistema prediktivnoj analitiki texnicheskogo sostoyaniya e`ksgaustera aglomashiny` s pomoshh`yu metodov iskusstvennogo intellekta. Iskusstvenny`j intellekt i prinyatie reshenij. 2024. № 3. S. 87–103.
13. Vinogradova N.V., Ivanov V.K., Palyux B.V., Sotnikov A.N. Sovremenny`e napravleniya razvitiya i oblasti prilozheniya teorii Dempstera–Shafera (obzor). Iskusstvenny`j intellekt i prinyatie reshenij. 2018. № 4. S. 32–42. DOI: 10.14357/20718594180403 [E`lektronny`j resurs]. 2018. URL: https://www.elibrary.ru/item.asp?id=36643710 (data obrashheniya: 12.05.2025).
14. Postanovlenie Pravitel`stva RF ot 10.09.2009 N 720 (red. ot 15.07.2013, s izm. ot 08.04.2014) «Ob utverzhdenii texnicheskogo reglamenta o bezopasnosti kolesny`x transportny`x sredstv». [E`lektronny`j resurs]. 2014. URL: https://docs.cntd.ru/document/ 902174533 (data obrashheniya: 02.06.2025).
15. Texnicheskij reglament tamozhennogo soyuza TR TS 010/2011 «O bezopasnosti mashin i oborudovaniya» (s izmeneniyami na 24 no­yabrya 2023 goda). [E`lektronny`j resurs]. 2023. URL: https://docs.cntd.ru/document/902307904 (data obrashheniya: 02.06.2025).
16. Hamda N.E.I., Hadjali A., Lagha M. Multisensor data fusion in IoT environments in Dempster–Shafer theory setting: an improved evidence distance-based approach. Sensors. 2023. V. 23. № 11. P. 5141.
17. Wang M. et al. A distributed sensor system based on cloud-edge-end network for industrial internet of things. Future Internet. 2023. V. 15. № 5. Р. 171.
18. Hongchao Wang, Weiting Zhang, Dong Yang, Yuhong Xiang. Deep-learning-enabled predictive maintenance in industrial internet of things: methods, applications, and challenges. IEEE Systems Journal. 2022. V. 17. № 2. Р. 2602–2615. [E`lektronny`j resurs]. 2023. URL: https://ieeexplore.ieee.org/document/9851995 (data obrashheniya: 31.04.2025).
19. Maktoubian J., Taskhiri M.S., Turner P. Intelligent Predictive Maintenance (IPdM) in Forestry: A Review of Challenges and Opportunities. Forests. 2021; 12(11):1495. https://doi.org/10.3390/f12111495
20. Ke Sheng Wang. Key Techniques in Intelligent Predictive Maintenance (IPdM) – A Framework of Intelligent Faults Diagnosis and Prognosis System (IFDaPS). Trans Tech Publications Ltd. Advanced Materials Research. October 20141039:490-505. [E`lektronny`j resurs]. 2014. http://dx.doi.org/10.4028/www.scientific.net/AMR.1039.490 (data obrashheniya: 31.04.2025).
21. Zhe Li, Yi Wang, Ke-Sheng Wang. Intelligent predictive maintenance for fault diagnosis and prognosis in machine centers: Industry 4.0 scenario. Advances in Manufacturing, 2017, 5(4): 377‒387. https://doi.org/10.1007/s40436-017-0203-8
22. Bezerra E.D.C. et al. Dempster–shafer theory for modeling and treating uncertainty in IoT applications based on complex event processing. Sensors. 2021. V. 21. № 5. Р. 1863.
23. Çavdar T., Ebrahimpour N. Decision-making for small industrial Internet of Things using decision fusion. Turkish Journal of Electrical Engineering and Computer Sciences. 2019. V. 27. № 6. Р. 4134–4150. https://doi.org/10.3906/elk-1809-60
24. Lin, Zhiming & Tang, Songping & Peng, Gang & Zhang, Yun & Zhong, Zhenxin. (2017). An artificial neural network model with Yager composition theory for transformer state assessment. 652–655. 10.1109/IAEAC.2017.8054097
25. Wei Y. and Yaowen F. Constructing basic belief assignment from feature data. 2013. Chinese Automation Congress. Changsha, China. 2013. Р. 605–610. doi: 10.1109/CAC.2013.6775807. [E`lektronny`j resurs]. 2013. URL: https://www.researchgate.net/publication/ 271547648_Constructing_basic_belief_assignment_from_feature_data (data obrashheniya: 03.06.2025).
26. Aboshosha A. et al. IoT-based data-driven predictive maintenance relying on fuzzy system and artificial neural networks. Scientific Reports. 2023. V. 13. № 1. Р. 12186.
27. Ansari F., Glawar R., Sihn W. Prescriptive maintenance of CPPS by integrating multimodal data with dynamic Bayesian networks. Machine Learning for Cyber Physical Systems: Selected papers from the International Conference ML4CPS 2017. Berlin, Heidelberg: Springer Berlin Heidelberg. 2019. Р. 1–8.
28. Svidetel`stvo o gosudarstvennoj registracii programm dlya E`VM № 2025667829 Rossijskaya Federaciya. Programma «EUS Model 1 PdM / DST» dlya promy`shlenny`x sistem prognoziruemogo obsluzhivaniya, realizuyushhaya matematicheskuyu model` snizheniya neopredelennosti danny`x / A.Yu. Chesalov (RU); pravoobladatel` OOO «Programmny`e sistemy` Atlansis» (RU). – № 2025664454/69: zayavleno 09.06.2025: opublikovano 09.07.2025. Byul. № 7. 1 s. http://www1.fips.ru/fips_servl/fips_servlet?DB=EVM&DocNumber= 2025667829