Parametric programming problem with crop yield forecasting models

The paper considers factor models that allow predicting the yield of agricultural crops. It is shown that the main climatic parameters that affect the effective feature are the air temperature and precipitation during the initial growing season. In this case, the factors of heat supply and moisture supply can represent values for both a month and another interval close to this duration. In addition to air temperature and precipitation, the yield of grain crops is affected by time. Models can reflect the relationship of the effective feature with factors at the level of experimental fields, agricultural organizations, and municipal districts. The presence of significant regression dependencies, which can be linear and nonlinear, reduces the uncertainty of the problem of optimizing agricultural production by reducing random and interval parameters. A model of parametric programming is presented, taking into account the expressions that characterize the relationship between the yield of grain crops and meteorological parameters in two variants, in order to obtain optimal plans for the production of agricultural products by the commodity producer. An example of the implementation of an optimization model for a real economy is considered. The proposed model is designed to support decision-making in conditions of uncertainty. The work is carried out according to statistical data on the yield of wheat, barley and oats in the Usolsky, Cheremkhovsky and Irkutsk districts for 1997-2018; based on the yield of variety plots in the Usolsky, Irkutsk, Bratsky and Nukutsky districts for 2000-2018 according to the data of the State Export Commission; based on the yield of LLC "Sibirskaya Niva" for the period 2005-2018. In addition, daily air temperatures and daily precipitation in the period May–August for 1997-2018 were used for meteorological points: Usolye-Sibirskoye, Cheremkhovo, Irkutsk and Bratsk.