Developing Generic Flight Schedules for Airport Clusters

Airports have their operational focus on the short term. Nevertheless, a long term view is crucial for planning and developing their infrastructure and business. Airport models are proper for assessments and investigations in the field of airport research whereas such models often focus on single elements only. This scope is not sufficient to address the air transportation system as a whole. Therefore a generic airport model will be implemented that covers operational aspects, e.g. passenger and aircraft movement demand as well as resulting economic changes. The granularity should be as roughly as possible, reducing the complexity and computing time, but accurately enough to model crucial intra-airport relationships. The basis for the analysis build already designed airport clusters, i.e. the airport model does not represent one single airport but rather fits to an airport cluster. 7 clusters were identified and they are characterized by a certain number of attributes like the number of runways, aircraft movements and revenues. In order to assess intra-airport relationships a flight plan, the main model input, is necessary. Therefore a methodology of developing generic flight plans, one for each cluster, will be implemented out of a variety of real flight schedules of the cluster representatives. Statistical and probability distributions are used to determine a suitable daily/weekly distribution of arrival and departure flights and a list of typical flights of such an airport cluster. Every airport-specific data (origin/destination information, specific aircraft type etc.) are transferred into generic categories. Another important aspect to be issued is the future development of the attributes, especially aircraft movements. This is forecasted by an already implemented model called Forecast of Aircraft Movements, short FoAM. Basic approach is the assignment of each flight leg worldwide to a distance, passenger number and aircraft category. For each combination of distance and passenger numbers a typical fleet mix is defined. The forecasted worldwide growth of passenger demand and the empirically determined fleet mix is applied to all legs in order to derive a future scenario. Assuming a certain seat load factor, the growth of the aircraft movements can be deduced from the overall aircraft movements generated by all legs. The methodology will be implemented in JAVA and should be adaptable to category changes and further parameters. According to Figure 1, the objective is to get generic flight plans for future years on the basis of growth of passenger demand. Applying the methodology to an air transport network will show the functionality of the methodology and the possible development of the network in the long term.