Peramalan Curah Hujan di Kabupaten Deli Serdang Menggunakan Fuzzy Time Series Cheng

Abstract

Rainfall is one of the important factors affecting various sectors of life, especially in agriculture, the environment, and water resource management. The uncertainty and fluctuations in rainfall data over time require forecasting methods that can effec tively address these uncertainties. This study aims to forecast monthly rainfall in Deli Serdang Regency using the Cheng Fuzzy Time Series method. The data used is secondary monthly rainfall data from January 2020 to December 2023. The stages in this research include the formation of the universe set, determining intervals using the Sturges method, fuzzification of actual data, formation of fuzzy logic relations (FLR), and defuzzification to produce forecast values. Model evaluation is carried out by calculating the Mean Absolute Percentage Error (MAPE) as an accuracy indicator. The forecasting process carried out using the Cheng Fuzzy Time Series method yields a Mean Absolute Percentage Error (MAPE) value of 19.1%, which is classified as good. Thus, this method can be considered a reliable alternative for predicting rainfall, while also providing useful information as a basis for decision-making in various related sectors.