МОДЕЛЮВАННЯ ТА ПРОГНОЗУВАННЯ ВАЛОВОГО НАДОЮ МОЛОКА З ВИКОРИСТАННЯМ СУЧАСНИХ МЕТОДІВ АНАЛІЗУ ЧАСОВИХ РЯДІВ

The article considers approaches to modeling and forecasting gross milk yield based on modern methods of time series analysis. The development of forecasting methods in the agro-industrial sector is an important task, since effective management of milk production allows optimizing resources, reducing losses and increasing the profitability of enterprises. Two approaches were used to implement the forecasting models: SARIMAX and Prophet – time series models. A comparative analysis of the accuracy of these models was conducted using evaluation metrics such as RMSE, MAE, MAPE and MASE. The results of the study showed that the Prophet model demonstrated higher forecasting accuracy, especially for long-term trends and seasonal changes. The results obtained can be used to plan production processes in the dairy sector, manage feed stocks, and determine optimal enterprise development strategies.