Application of the Fuzzy Time Series Chen and Markov Chain Method with Sturges and Average-Based Intervals in Forecasting the Rupiah Exchange Rate Against the US Dollar

International trade plays a crucial role in meeting the needs of a country’s population, with exchange rates being a key factor influencing cross-border transactions. The instability of exchange rate fluctuations necessitates accurate forecasting to assist governments, exporters, and importers in decision making. As the United States is one of Indonesia’s largest trading partners, predicting the Rupiah to US dollar exchange rate becomes highly essential. Fuzzy Time Series (FTS) is a linguistic-based forecasting method that is effective for data with no discernible patterns. This study compares the accuracy of two FTS developments, namely FTS Chen and FTS Markov Chain, using two interval determination methods: Sturges and Average Based methods. The results show that the FTS Markov chain with average-based intervals provides a better forecasting performance than the other three methods, achieving a MAPE value of 0.3351%. Keywords: Exchange rate, Fuzzy Time Series, Sturges, Average Based, Interval, MAPE.