APPLYING DATA MINING TECHNIQUES TO FORECAST NUMBER OF AIRLINE PASSENGERS IN SAUDI ARABIA (DOMESTIC AND INTERNATIONAL TRAVELS)

This work involves forecasting the number of domestic and international airline passengers in Saudi Arabia. The method used is the neural network technique. Annual data 1975 to 1986 was used and categorized into 16 variables. The forecast was obtained using the Model Quest Miner package, which uses some historical data for developing the model then proceeds to an evaluation phase. The period used for developing the model for the number of passengers was 18 years, while the period used for evaluation was 6 years. Samples of output are presented for each model. Plots of forecasts versus the actual number are shown, together with the percentage of trend fit. Results indicated that the oil gross domestic product, population size and per capita income were found to be the most contributing variables that affect the number of passengers in the Saudi Arabian airline sectors.