WAYS OF BUILDING AN INTELLIGENT AGRICULTURAL COMPANY MANAGEMENT SYSTEM

Abstract

This publication examines the problem of building an intelligent management system for an agricultural company. The intelligent management system of an agricultural enterprise is an important tool for increasing the efficiency and profitability of agriculture. Various approaches can be used to build such a system, such as expert systems, neural networks, and machine learning. In addition, it is possible to create a digital double of an agricultural enterprise, which will allow the use of large volumes of data for weather forecasting, productivity and logistics planning. The main stages of building an intelligent management system include the collection and primary processing of data, their primary analysis and classification into business processes, building models and developing algorithms for decision-making. A weather model can be built based on the analysis of indicators for the last ten years, including temperature, humidity, precipitation and other parameters. Based on this data, a neural network can be developed that can predict the weather with high accuracy. The productivity model can be built on the basis of planned indicators of agricultural crops, such as minimum and maximum productivity, indicators of the chemical composition of the soil, the amount of applied fertilizers and absorption coefficient. Based on this data, a neural network can be developed that can predict yield and assist in production planning. The logistics model can be built on the basis of data on production and transportation of products.