The objective of this study is to find a forecasting model for the value of agriproduct processing types of flour export of Thailand, and to compare suitable forecasting models for predicting the value of agriproduct processing types of flour export of Thailand. The study used export value data of two types of flour products cassava flour and Rice flour on a monthly basis from January 2012 to December 2022, totaling 132 months. The study assessed three individual forecasting methods: Exponential Smoothing Method, Time Series Regression Method and Box-Jenkins Method, along with a combined forecasting using four methods: Least Absolute Value Method (LAV), a combined forecasting Time Series Regression Analysis using Selection Stepwise Regression Variable, MAPE-based combination, and Weighted Inverse to the Square Root Sum of Squared Error (WISRSSE). The forecasts obtained from these methods were compared, considering the method with the lowest Mean Absolute Percentage Error (MAPE) as the most appropriate model. The analysis found that the best-suited forecasting method is the combined forecasting Time Series Regression Analysis using Selection Stepwise Regression Variable. Organizations or businesses interested can use it as a guideline for investment decision-making and for various planning purposes, such as sourcing raw materials, increasing production capacity, and exporting products, both in the present and the future.