A time-series modelling approach for airport short-term demand forecasting

Abstract

This paper presents a time-series modelling approach for airport short-term demand forecasting. The model assesses how various external factors such as seasonality, fuel price, airline strategies, incidents and financial conditions affect airport activity levels. The quarterly percentile demand change is used to represent airport activity. The modelling approach recommends disaggregating total demand into its local and connecting components due to differences in how local and connecting traffic respond to changes in different factors. The modelling framework is illustrated using the local traffic at Philadelphia International airport as a case study and validated for the largest 100 airports in the domestic US markets. The validation results show that the demand can be predicted with acceptable accuracy even with a simple time-series model.