Airport passenger estimation with time series analysis method: Balıkesir/Edremit Koca Seyit Airport example

This study presents an analysis of air transportation demand forecasting using monthly passenger data from Edremit Koca Seyit Airport. As the airline industry is highly sensitive to economic fluctuations and seasonal effects, demand forecasting plays a critical role in the operational planning of airline companies. In this context, passenger forecasts were made using the ARIMA model, one of the time series analysis techniques. In the study, monthly passenger data from August 2010 to October 2024 were analyzed. The dataset was transformed into a stationary form, and the most appropriate ARIMA model was selected to carry out the forecasting process. Based on statistical tests, the ARIMA (11,1,6) model was identified as the best-fitting model. This model yielded the lowest AIC and BIC values. Residual analysis showed that the errors were randomly distributed and did not contain autocorrelation, indicating the statistical reliability of the model. The findings of this study contribute to the development of data-driven decision support mechanisms for airport management, airline companies, and transportation planners. Furthermore, it aims to fill a gap in the academic literature by providing evidence supporting the applicability of the ARIMA model in airline demand forecasting.