<![CDATA[Palm oil is a plantation commodity that demonstrates strong resilience during economic crises and serves as one of Indonesia’s major sources of foreign exchange. In North Sumatra Province, palm oil production has shown a steady increase over the past five years, with a cultivated area reaching 490.16 thousand hectares. This study aims to forecast the amount of palm oil production in the province using the fuzzy time series method, which applies the concept of fuzzy sets as its analytical foundation. Several fuzzy time series models, including the Cheng and Lee models, are compared to determine the model that produces the smallest forecasting error based on the criteria of MAE, MSE, MPE, and MAPE. The forecasting results with Cheng's fuzzy time series were 7270.418 thousand tons, while the forecasting results with Lee's fuzzy time series were 7275.425 thousand tons. The results of the analysis of the accuracy level of the MAE value for the Cheng model are 1293.873 and the Lee model is 1463.606, the MSE value for the Cheng model is 5,885,351.768 and the Lee model is 6,304,592.844, the MPE value for the Cheng model is -0.016% and the Lee model is -0.225%, and The MAPE value for the Cheng model is 0.345% while the Lee model is 0.423%. So it was concluded that Cheng's fuzzy time series method was stated to be better for forecasting the amount of palm oil production in North Sumatra province because it produced smaller error values.]]>
Copyrights © 2026
