EUVIMEDEUVIMED entdecken
Crossref

ASSESSMENT OF PROMISING PARAMETERS OF MILK DEMAND USING A RECURRENT NEURAL NETWORK

Aleksandr Alekseevich Dubovitskii, Elvira Anatolevna Klimentova, Ekaterina Sergeevna Babkina, Vladislav Aleksandrovich SHatskii

Экономика сельского хозяйства России · 01.10.2026 · (10) · 131-140

DOI: 10.32651/2610-131 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Экономика сельского хозяйства России, online erschienen am 01.10.2026. Verfasst von Aleksandr Alekseevich Dubovitskii und 3 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The article is devoted to the assessment of promising parameters of milk demand through the study of factors that have a decisive influence on its formation based on the use of an artificial neural network. As a result of the conducted research, the specifics of the functioning and development trends of dairy cattle breeding, the peculiarities of the formation of demand and prices for raw milk sold by agricultural producers have been identified. In the process of testing two main types of recurrent neural networks (LSTM, GRU), the most acceptable of them for demand forecasting was identified, which was a network based on controlled recurrent blocks with a backpropagation algorithm (GRU), which provides the best metrics for model quality. The factors that have a decisive influence on the formation of the forecast have been identified, including: the volume of milk production, the consumer confidence index, the proportion of the population with monetary incomes below the poverty line, the unemployment rate, the average per capita monetary income of the population, and the price of milk. Within the framework of the constructed model, using the "as in the last year" scenario, the prospective parameters of the demand for milk in the Russian Federation in the amount of 37646.6 thousand tons were calculated, which will exceed the actual level by 391.1 thousand tons in 2025. The constructed model allows making adjustments to the forecast by replacing the actual data of the previous year with its planned/ expert values, which is expected to increase the accuracy of forecasting. It is concluded that reliable demand forecasting contributes to a better understanding by agribusiness entities of the prospects for the development of the milk market, being a prerequisite for successful pricing, laying the foundation for strategic and operational planning.

Zitieren

APA

Dubovitskii, A. A., Klimentova, E. A., Babkina, E. S., & SHatskii, V. A. (2026). ASSESSMENT OF PROMISING PARAMETERS OF MILK DEMAND USING A RECURRENT NEURAL NETWORK. Экономика сельского хозяйства России, 131-140. https://doi.org/10.32651/2610-131

Vancouver

Dubovitskii AA, Klimentova EA, Babkina ES, SHatskii VA. ASSESSMENT OF PROMISING PARAMETERS OF MILK DEMAND USING A RECURRENT NEURAL NETWORK. Экономика сельского хозяйства России. 2026:131-140. doi:10.32651/2610-131

BibTeX
@article{Dubovitskii2026,
  author = {Dubovitskii, Aleksandr Alekseevich and Klimentova, Elvira Anatolevna and Babkina, Ekaterina Sergeevna and SHatskii, Vladislav Aleksandrovich},
  title = {ASSESSMENT OF PROMISING PARAMETERS OF MILK DEMAND USING A RECURRENT NEURAL NETWORK},
  journal = {Экономика сельского хозяйства России},
  year = {2026},
  number = {10},
  pages = {131--140},
  doi = {10.32651/2610-131}
}
Kostenlos starten →