Application of neural networks and big data technologies in agriculture: increasing the efficiency and sustainability of agricultural production

Abstract

This article is dedicated to exploring the potential of applying neural networks and big data technologies in agriculture to enhance the efficiency and sustainability of agricultural production. Based on a comprehensive analysis of scientific literature and empirical data from real implementation projects, key areas for utilizing these innovative approaches have been identified: precision farming, resource management optimization, crop and livestock condition monitoring, and yield and productivity forecasting. It has been demonstrated that integrating neural network algorithms and big data analysis tools significantly improves the decision-making process at all stages of agricultural production by accounting for numerous factors and identifying non-obvious patterns. A conceptual model of a decision support system for agricultural enterprises has been developed, based on synthesizing machine learning methods and intelligent analysis of heterogeneous data sets. Validation of the model on real datasets showed a 15–20% improvement in yield prediction accuracy and a 10–12% reduction in resource costs compared to traditional approaches. The results lay the foundation for scaling the proposed solutions and adapting them to the specific characteristics of individual agricultural enterprises, aiming for a transition to sustainable and highly productive next-generation agriculture.