M. N. Belozerov1
1 National Research Technological University «MISIS» (Moscow, Russia)

1 mnbelozyorov@gmail.com

Any machine learning system based on statistical data has some inaccuracy, which puts production processes at risk. This risk can lead to financial, reputational and other losses for industrial enterprises. The purpose of the study is to develop the architecture of a model risk management system for forecasting demand for industrial enterprise products. Risk management in creating digital twins of production processes and modeling technological processes in industrial enterprises is based on a conceptual scheme that includes a management system and methodological principles that can be based on various approaches. Given that industrial enterprises specialize in different predictive models, each model requires a separate management system that takes into account a certain set of risk factors. This article is methodological in nature and provides an example of how to integrate machine learning systems and expert knowledge to automate model risk management.

Belozerov M.N. Methodology for managing model risks in predicting demand for the manufacturing company's products. Neurocomputers. 2025. V. 27. № 3. P. 33–39. DOI: https://doi.org/10.18127/j19998554-202503-05 (in Russian)

1. Borisova V.V., Demkina O.V., Savin A.V. Riski tsifrovizatsii promyshlennykh kompanij. Innovatsii i investitsii. 2019. № 12. S. 294–297. (in Russian)
2. Kropachek O.Yu. Issledovanie diagnosticheskikh riskov parametricheski neopredelennoj kvadratichnoj diskriminantnoj funktsii pri diagnostike promyshlennykh ob''ektov. Sistemi upravlіnnya, navіgatsії ta zv'yazku. 2017. № 5. S. 56–59.
3. Bocharov E.P., Aleksentseva O.N., Ermoshin D.V. Otsenka riskov promyshlennykh predpriyatij na osnove imitatsionnogo modelirovaniya. Prikladnaya informatika. 2008. № 1. S. 15–24. (in Russian)
4. Khrustalyov O.E. i dr. Model'noe obosnovanie innovatsionnogo razvitiya naukoemkogo sektora rossijskoj ekonomiki. Dajdzhest-finansy. 2013. № 4. S. 19–29. (in Russian)
5. Rakhlis T.P., Eremenko D.D. Vliyanie tsifrovizatsii na deyatel'nost' krupnykh promyshlennykh predpriyatij: riskovyj aspect. Sb. dokl. «Rossijskie regiony v fokuse peremen» (Ekaterinburg, 18–20 noyabrya 2021 g.). T. 1. Ekaterinburg: UrFU. 2022. S. 476–479. (in Russian)
6. Shablakov A.D., Solov'eva I.A. Otsenka riskov vnedreniya skvoznykh tsifrovykh tekhnologij v promyshlennosti. Vestnik Yuzhno-Ural'skogo gosudarstvennogo universiteta. Seriya: Ekonomika i menedzhment. 2023. T. 17. № 2. S. 133–143. (in Russian)
7. Koksharov V.A. Teoreticheskie aspekty upravleniya energo-effektivnost'yu innovatsionnykh tekhnologij promyshlennogo predpriyatiya. Innovatsii i investitsii. 2018. № 8. S. 3–7. (in Russian)
8. Matveeva L.G., Chernova O.A. Strategicheskij konsortsium kak mekhanizm narashchivaniya innovatsionnogo potentsiala promyshlennosti Yuga Rossii. Journal of Economic Regulation (Voprosy regulirovaniya ekonomiki). 2013. T. 4. № 3. S. 65–72. (in Russian)
9. Pestrikov S.A., Tyul'kina S.Yu. Problemy sozdaniya effektivnoj biznes-sredy v avtomobil'noj promyshlennosti. Shumpeterovskie chteniya. 2013. № 1. S. 66–72. (in Russian)
10. Klochkov V.V. Upravlenie innovatsionnym razvitiem naukoemkoj promyshlennosti: modeli i resheniya. M.: IPU RAN. 2010. (in Russian)
11. Mikulenkov A.S. Osobennosti risk-menedzhmenta predpriyatij v epokhu tsifrovoj transformatsii. Sb. materialov XXI Mezhdunar. nauch.-praktich. konf. «Smirnovskie chteniya–2022». 2022. S. 70–84. (in Russian)
12. Nikitin N.A. Veroyatnostnye metody ucheta model'nykh riskov pri otsenke investitsij v tekhnologii iskusstvennogo intellekta. Innovatsionnoe razvitie ekonomiki. 2023. T. 2. S. 123–134. (in Russian)
13. Minasyan V.B., Ivko D.G. Analiz model'nogo riska ispol'zovaniya tekhnologii mul'tiplikatorov pri otsenke aktsij rossijskikh kompanij. Finansy: teoriya i praktika. 2019. T. 23. № 6. S. 91–116. (in Russian)
14. Kirilyuk I.L. Model'nye riski v finansovoj sfere v usloviyakh ispol'zovaniya iskusstvennogo intellekta i mashinnogo obucheniya. Russian Journal of Economics and Law. 2022. T. 16. № 1. S. 40–50. (in Russian)
15. Moiseev E. i dr. Metod otsenki IT-sostavlyayushchej model'nogo riska i ekonomicheskogo kapitala na ego pokrytie. Journal of Money and Finance. 2022. T. 81. № 3. S. 107–127. (in Russian)
16. Xu T. et al. Risk factors assessment of smart supply chain in intelligent manufacturing services using DEMATEL method with linguistic q-ROF information. Journal of Operations Intelligence. 2024. V. 2. № 1. P. 129–152.
17. Zhu D., Li Z., Mishra A.R. Evaluation of the critical success factors of dynamic enterprise risk management in manufacturing SMEs using an integrated fuzzy decision-making model. Technological Forecasting and Social Change. 2023. V. 186. P. 122137.
18. Radanliev P. et al. Cyber risk at the edge: current and future trends on cyber risk analytics and artificial intelligence in the industrial internet of things and industry 4.0 supply chains. Cybersecurity. 2020. V. 3. P. 1–21.
19. Somer P., Thalmann S. Risk management of AI in industry: a literature review. The 15th Mediterranean Conference on Information Systems (MCIS) and the 6th Middle East & North Africa Conference on digital Information Systems (MENACIS). Madrid. 2023.
20. Zadeh L.A. Fuzzy logic. In the book Granular, Fuzzy, and Soft Computing (Ed. by T.-Y. Lin, Ch.-J. Liau, J. Kacprzyk). New York, NY: Springer US. 2023. P. 19–49.