Airline Passenger Forecasting using ARIMA and Artificial Neural Networks Approaches

摘要: Demand uncertainty has been increasing as a result of the rising trend using airplanes transportation mode option in Indonesia over years. This condition results need for ability to accommodate rise airline companies withstand within industry. The forecast accuracy highly determines strategy formulation. Thus, accurate forecasting models are crucially needed. In this study, neural network is proposed create best-fitted model predict future values non-traditional method that already tested predictions. As comparison with traditional model, Autoregressive Integrated Moving Average (ARIMA) applied. study used monthly passenger data from Indonesian airlines, focused on Jakarta-Yogyakarta (CGK-JOG) and Jakarta-Singapore (CGK-SIN) routes, which representatives most profitable route both domestic international flight. Mean Absolute Percentage Error (MAPE) methods were then compared forecasted demand next 12 months calculated. produced better value than ARIMA MAPE 1.29 CGK-JOG 1.66 CGK-SIN route.