<![CDATA[Indonesia exports many cut flowers to Southeast Asia, mainly to Thailand, Singapore, and Malaysia. However, export amounts fluctuate significantly due to the seasons, weather, and local customs, making it difficult for both producers and decision-makers to predict trends. This study predicts Indonesia’s cut flower exports using the SARIMA method, which helps identify repeating patterns in monthly export data. We used export numbers from April 2018 to August 2024, analyzed with Python, and split the data into a training set (80 percent) and a test set (20 percent). We chose the model’s settings based on patterns found in the data, with the best setup picked using a method called grid search, which aims for the lowest AIC value. The top model was SARIMA(1,0,1)×(2,0,2,12), with an AIC of 876.08. Measures of accuracy showed that SARIMA predicted better than ARIMA(0,1,1), with a MAPE of 0.30 and a MAE of 13.31. These results show SARIMA can effectively handle seasonal changes and make export predictions more reliable. This helps make better plans and policies for flower exports, and can strengthen Indonesia’s position in the market Keywords: Cut flower export, SARIMA, time series forecasting, ARIMA]]>
Copyrights © 2026
