Meramal Produksi Padi Nasional: Pendekatan Moving Average dan Triple Exponential Smoothing

Abstract

National rice production is a crucial indicator for maintaining food security in Indonesia. Seasonal fluctuations and annual trends in rice production require accurate forecasting methods to support strategic decision-making. This study aims to compare the forecasting accuracy of national rice production using the Moving Average and Triple Exponential Smoothing methods. Monthly rice production data from the 2020–2024 period were used as the basis of analysis. The forecasting results show that the Moving Average method tends to respond slowly to changes in actual production values, while the Triple Exponential Smoothing method is more responsive in capturing seasonal patterns and trends. Accuracy measurements indicate that Moving Average produced MAPE of 41.39%, MAD of 1,828,830 tons, and MSE of 6.24 , while the Triple Exponential Smoothing method provided better results with MAPE of 18.05%, MAD of 814,216 tons, and MSE of 1.13 . Based on these findings, the Triple Exponential Smoothing method is recommended as a more suitable and effective forecasting technique for national rice production data characterized by seasonal patterns.