Comparison of Fuzzy Time Series Markov Chain and Fuzzy Time Series Lee in North Sumatra Import Value Forecasting

North Sumatra is a province that carries out import activities in other provinces to complement the shortage of goods in North Sumatra. Every month based on data from the Central Bureau of Statistics (BPS) of North Sumatra, the import value of North Sumatra constantly fluctuates. So a method is needed to show the increase or decrease in the value of North Sumatra in the next period. The method is forecasting, where forecasting is an activity that predicts something in the future, intending to get information in the next period as a decision or decision in the future. Many forecasting methods can be used, but this study uses fuzzy time series Markov Chain and fuzzy time series Lee methods. These two methods aim to obtain the best forecasting method based on the minor mean absolute percentage error (MAPE) and mean square error (MSE). In this study, the data is on the value of imports in North Sumatra taken from the Central Statistics Agency (BPS) of North Sumatra. The test results on forecasting the import value of North Sumatra show that the fuzzy time series Markov Chain is better than the Fuzzy Time series Lee based on the smallest MAPE and MSE values. The MAPE fuzzy time series Markov Chain value is 7.7467%, and the MSE value is 176,748,587. Meanwhile, Lee's fuzzy time series has a MAPE of 10.014% and an MSE value of 2,387,874,804