Kajian Metode Fuzzy Time Series-Chen dan Fuzzy Time Series-Markov Chain dan Terapan pada Peramalan Curah Hujan

Abstract

This research uses the Fuzzy Time Series-Chen and Fuzzy Time Series-Markov Chain methods to study and apply both methods to forecasting rainfall in Medan City so that the accuracy of each method is obtained. Fuzzy Time Series is a forecasting method based on fuzzy principles. Forecasting in this method is by using previous data patterns, then these patterns can predict future data. The Fuzzy Time Series (FTS) method is a new approach that combines linguistic variables with an analysis process so that the results of the study and its application are obtained to predict rainfall in Medan City in January 2018-October 2022 seen from the very accurate MAPE value determination. The MAPE value from the results of rainfall forecasting in the city of Medan using the Fuzzy Time Series-Chen method is 1,20% and for forecasting one month ahead it is 264 mm in November 2022 while in the Fuzzy Time Series Markov-Chain it is 1,01% and predictions are made for the next 12 months so that the forecast pattern is most similar to the data pattern in January 2021-December 2021.