FORECASTING AIR PASSENGER TRAFFIC VOLUME: EVALUATING TIME SERIES MODELS IN LONG-TERM FORECASTING OF KUWAIT AIR PASSENGER DATA

Accurate estimation of air transport demand is vital for airlines, related aviation companies, and government agencies. For example, both short-term long-term business plans airlines require accurate forecasting future traffic flows. This study aims to forecast the volume passengers in Kuwait International Airport (KIA), which state Kuwait. Using monthly data between January 2012 December 2018, this focuses on modelling number KIA. A wide range time series models are considered research, including autoregressive-integrated-moving average model (ARIMA), exponential smoothing with errors term (ETS), Holt-Winters smoothing, neural network autoregression (NNAR), hybrid Bayesian structural (BSTS), a model. The performance these compared using multiple train-test splits where fitted training sets evaluated test sets. mean absolute percentage error (MAPE) used compare various models. Empirical analysis suggests that BSTS compares favorably against other its ability complex series. may be applied problems irregularity.