Shemetov A.I., Ivanov A.S.

The role of technical elements of digitalization in the development of modern animal farming

It is estimated that farmers need to increase production by 70% over the next 50 years to meet the growing global demand for meat and animal products [1]. Since land and other natural resources are limited, more efficient ways of raising more animals per hectare of land will need to be found to meet this growing demand. Today, most animal husbandry methods require manual intervention at some level. This affects production productivity. Previously, digital technologies were expensive and could not be applied on a large scale. Today, sensors, big data, and machine learning algorithms have significant cost advantages over these older detection methods. Currently, the sensors offered by the market are significantly limited in reliable forecasting and disease management in animal husbandry due to continuous automated monitoring in real time. In addition, there are certain technical problems, such as the location of the sensors, what the sampling rate will be, and how the data will be transmitted. All of these considerations affect the accuracy of the algorithms, as well as the scalability and practicality of a solution that ultimately can be used on a livestock farm. In real-time systems, large feature sets can be problematic due to computational complexity and higher storage requirements. In light of the still existing pandemic, when restrictions prevent veterinarians and producers from visiting farms, cowsheds and feed mills (but there is a need for 24/7 information on activities, consumption and production of products in real time), then the current and practically only possible solution is the introduction of digital technologies.

Keywords: LIVESTOCK, SMART FARMING, SENSORS, SENSORS, MACHINE LEARNING