Public Policy Model of Delays in a Large Network of Major Airports

Abstract

As more airports in the United States and in Europe become congested, it is increasingly common for delays at one or more airports to spread to other parts of the network. This paper describes an analytical model developed to study this complex phenomenon. The Approximate Network Delays (AND) model computes the delays from local congestion at individual airports and captures the “ripple effect” that leads to the propagation of these delays to other airports. AND can be used to explore at a macroscopic level the implications of a large number of policy alternatives and future scenarios on systemwide delays and associated costs. It has been fully implemented for a network consisting of the 34 busiest airports in the continental United States and for a network of the 19 busiest airports in Europe. An analysis is presented of the estimation of ground slack in the scheduled turnaround times and its strong relation with the spreading of delays. Furthermore, AND shows the effect of delay propagation in an airport network and, for that reason, a comparison is made between the two main hubs in the European and U.S. airport networks, Frankfurt International in Germany and Chicago O'Hare in Illinois. The comparison shows that higher local delays and more short-haul flights in Chicago cause stronger delay propagation than in Frankfurt. Finally, by using AND, the modernization program at Chicago O'Hare could achieve an 80% reduction of delays locally and a 5% reduction in networkwide delays under visual flight rules conditions.