Tools for forecasting the development of marketing systems of the agro-industrial complex of the Arctic zone in the context of digital transformation and volatility

Issue № 5, 2026, article № 2, pages 11-20

Section: Digitalization in agrarian and industrial complex

Language: Russian

Original language title: ИНСТРУМЕНТАРИЙ ПРОГНОЗИРОВАНИЯ РАЗВИТИЯ МАРКЕТИНГОВЫХ СИСТЕМ АГРОПРОМЫШЛЕННОГО КОМПЛЕКСА АРКТИЧЕСКОЙ ЗОНЫ В УСЛОВИЯХ ЦИФРОВОЙ ТРАНСФОРМАЦИИ И ВОЛАТИЛЬНОСТИ

Keywords: AGRO-INDUSTRIAL COMPLEX, ARCTIC ZONE, MARKETING SYSTEMS, FORECASTING, DIGITAL TRANSFORMATION, VOLATILITY, HYBRID MODELS, NEURAL NETWORKS, SATELLITE MONITORING, RISK-BASED PRICING, ARCTIC PREMIUM, NORTHERN DELIVERY

Abstract: The relevance of this study stems from the need to ensure food security and sustainable development of the agro-industrial complex in the Arctic zone of the Russian Federation in the context of digital transformation and high volatility caused by extreme natural and climatic factors and limited transport and logistics accessibility. Traditional marketing forecasting methods, successfully applied in the southern and central regions, demonstrate limited effectiveness in Arctic conditions due to ignoring specific determinants: the state of ice routes, navigation periods along the Northern Sea Route, permafrost dynamics, and the consumer behavior characteristics of isolated Arctic communities. It has been established for the first time that 40 percent of marketing environment volatility in the Arctic zone is determined by transport and climatic factors. A hybrid forecasting model has been developed that integrates SARIMA with neural network analysis of satellite data, improving the accuracy of medium-term sales forecasts by 25-30 percent. The transformation of the pricing model into an Arctic model of "logistics availability - inventory - price" is mathematically proven, and a risk-based pricing methodology with "Arctic premium" coefficients (1.44-1.69) is synthesized, reducing losses from forced sales by 18 percent. A conceptual architecture for a digital platform with a "double loop" principle is proposed, using ethnographic and shift data as proxy variables. This increases the accuracy of short-term demand forecasts from 72 to 89 percent and ensures a 32 percent reduction in losses from overstocking while reducing shortages by 28 percent. The combination of the proposed methods and models forms a methodological basis for the transition of marketing systems in the Arctic agro-industrial complex from reactive supply management to predictive product supply formation, which is critical for ensuring food security and the sustainable socioeconomic development of Arctic territories.

Authors: Eremina Irina Aleksandrovna