PERAMALAN PRODUKSI KARET INDONESIA MENGGUNAKAN FUZZY TIME SERIES DUA FAKTOR ORDE TINGGI RELASI PANJANG BERDASARKAN RASIO INTERVAL
Abstract
The fuzzy time series method for forecasting continues to develop over time. This research discusses fuzzy time series, which considers two factors for high order using interval partitioning based on interval ratio with long relation construction for getting different accuracy in forecasting between combination method and existing method. The first step is the formation of the universe of speech. Second, divide the universe of discourse into several intervals using interval ratios. Third, fuzzification. Fourth, build fuzzy logic relations and fuzzy logic relation groups, and fifth, defuzzification. The previous methods would be compared with the fuzzy logic relation construction result. The simulation used Indonesian rubber production data for 2000-2020. The results and errors were tested using the average forecasting error rate (AFER). AFER value of the forecasting method is 1.863% obtained.
Keywords: Forecasting, fuzzy time series, long relation
MSC2020: 62M10, 62M20, 62M86, 03E72
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.