<![CDATA[Forecasting is an important function of operations management, providing a basis for decision-making in the face of uncertainty. This study aims to compare the performance of the Autoregressive Integrated Moving Average (ARIMA), Moving Average (MA), and Exponential Smoothing (ES) methods in forecasting the BI Rate, while also enriching the body of references on the application of forecasting methods in operations management. A comparative quantitative approach was employed using secondary BI Rate data in the form of a time series. The ARIMA model was developed using the Box–Jenkins procedure, while Moving Average and Exponential Smoothing were employed as comparative methods. Forecasting accuracy was evaluated using Mean Absolute Deviation (MAD), Mean Squared Error (MSE), and Mean Absolute Percentage Error (MAPE). The results show that the three-period Moving Average (MA-3) achieved the highest forecasting accuracy, with a MAD of 0.001689, MSE of 0.0000074, and MAPE of 3.415%. Exponential Smoothing with α = 0.25 yielded a MAD of 0.006682, MSE of 0.0000562, and MAPE of 15.101%, whereas ARIMA(1,1,0) produced a MAD of 0.008929, MSE of 0.0000797, and MAPE of 19.971%. These findings indicate that a more complex method does not necessarily produce greater forecasting accuracy. Given the relatively stable characteristics of the BI Rate data, the Moving Average method provided more accurate forecasts than Exponential Smoothing and ARIMA. This study also demonstrates that the application of ARIMA can serve as an alternative approach for enriching the teaching and study of forecasting in the field of operations management.]]>
