American Institute of Aeronautics and Astronautics

The raison d'etre of the Airline Passenger Transportation System (APTS) is the rapid, affordable, and safe transportation of passengers (and cargo). Traditional APTS performance metrics, such as flight delays and cancellation rates, fail to account for the delays to passenger trips caused by cancelled flights, missed connections and diversions. Previous research estimated Passenger Trip Delays statistics for single segment passenger itineraries (i.e. direct itineraries) only. This paper describes the results of analysis of the Passenger Trip Delays for multi-segment itineraries (i.e. direct and connecting itineraries). The results provide a nuanced view of the performance of the APTS, the causes of passenger trip delays, and mitigation strategies. In 2007, 6.8M flights provided service to 299 airports. The network of scheduled flights supported an estimated 51M passenger itineraries. Thirteen percent of the itineraries provided direct service, 87% provided connecting service. Direct itineraries carried an average of 30 passengers, while connecting itineraries carried an average of 4 passengers, resulting in an even split between passengers on direct and connecting itineraries. In terms of performance, 31% of the itineraries were disrupted yielding passenger trip delays totaling 15,841 years. Forty eight percent of the total trip delays were the result of itineraries disrupted by delayed flights, 31% by cancelled flights, 19% by missed connections, and 25% by diverted flights. Connecting itineraries, through hubs or not, were more likely to experience disruptions. The asymmetries in type of itinerary, load factor, and type of disruption provide insights into the inherent structure of the system and explain why a single performance metric cannot incentivize uniform performance improvement. The implications for NextGen and air transportation policy are discussed.