Quantitative Forecasting of Soybean Commodity Production in Indonesia using POM-QM Software

Food serves not only as a fundamental requirement for human existence but also embodies a significant economic asset, particularly through the agricultural production of food crops. The objective of this research is to project soybean output in Indonesia utilizing Production and Operations Management-Quantitative Method (POM-QM) software. The dataset concerning soybean production spanning from 2000 to 2024 reveals notable variability, characterized by intervals of both scarcity and abundance. Three distinct time-series forecasting methodologies were implemented: Double Moving Average (DMA), Weighted Moving Average (WMA), and Single Exponential Smoothing (SES). The optimal model was determined based on critical precision indicators, which include Mean Absolute Deviation (MAD), Mean Square Error (MSE), and Mean Absolute Percentage Error (MAPE). Among the employed methodologies, WMA emerged as the most precise, demonstrating a MAPE value of 17.89%. The projected soybean production is approximated at 558.3 tons, suggesting an adequate supply to satisfy consumer requirements. It is advisable for the government to leverage these projections to proficiently forecast and strategize for forthcoming national soybean demand. Keywords: Quantitative forecasting, production, soybean commodity, POM-QM