Against the backdrop of the agricultural sector's transition toward a market-oriented and innovation-driven development model, the uncertainty associated with the factors determining its dynamics has increased significantly. Dynamic processes in agriculture are generally described by time series, creating a need for intelligent methods capable of improving the reliability and accuracy of forecasts under conditions of high uncertainty. This paper presents a model and method for forecasting grain production in the Sheki–Zagatala Economic Region of Azerbaijan using fuzzy time series. In the proposed model, grain production is represented as a linguistic variable, where linguistic terms are employed to characterize the modeled process. These linguistic terms are expressed as fuzzy sets, allowing them to be used subsequently in both mathematical modeling and analytical computations. Based on the proposed method, forecasts of grain production for the Sheki–Zagatala Economic Region were calculated, and the effectiveness of the approach was demonstrated. The application of the fuzzy time series model to grain production forecasting enables the processing of uncertain data, overcomes the limitations of conventional statistical methods, and provides more accurate forecasting results (pp.81-87).