KLASIFIKASI INTENSITAS HUJAN DI SAMARINDA MENGGUNAKAN LOGIKA FUZZY MAMDANI

Abstract

This study aims to classify rainfall intensity in the Samarinda area into three categories: light rain, moderate rain, and heavy rain based on three meteorological variables: temperature (in °C), air pressure (in hPa), and rainfall (in mm) to provide a more adaptive and accurate classification of rainfall intensity based on local weather conditions in Samarinda, which is prone to disasters due to high rainfall intensity. The data used in this study was obtained from the Meteorology, Climatology, and Geophysics Agency (BMKG) Samarinda for the period of October to December 2024. This study implements the Mamdani Fuzzy Logic method, which consists of the stages of fuzzification, rule base application, inference, and defuzzification. Fuzzy logic was chosen due to its ability to handle data that is ambiguous and uncertain, which is common in weather phenomena. Testing results on 50 random weather condition data samples indicate that the developed Mamdani fuzzy model achieved an accuracy of 100% on the test data, demonstrating consistency between the resulting rainfall intensity classification and actual data. Based on these findings, this model can be utilized as a support tool for decision-making, both by individuals and local government agencies, in efforts to monitor and mitigate extreme weather conditions in Samarinda, East Kalimantan.

Keywords : Fuzzy Logic, Mamdani Method, Rainfall Intensity, Classification