Abstract:
This study aims to forecast durian production in Thailand from 2024 to 2030 by comparing two forecasting techniques: the Seasonal Autoregressive Integrated Moving Average with Exogenous Variables (SARIMAX) and the Exponential Smoothing Method. The objective is to identify the most suitable model for accurate predictions. The results indicate that the SARIMAX (1,1,1) model outperforms the Holts Double Exponential Smoothing method, achieving a Mean Absolute Percentage Error (MAPE) of 3.77% and a Root Mean Squared Error (RMSE) of 66,660.08, compared to Double Exponential Smoothing (Holts Method) with a MAPE of 10.74% and an RMSE of 153,732.34. The forecast suggests a continuous increase in durian production, with an estimated 1.56 million tons in 2024, rising to 1.81 million tons by 2030. These findings provide valuable insights for agricultural planning and market management, ensuring a more strategic and data-driven approach to Thailands durian industry.