Title: Improving airline schedule management through accurate flight arrival prediction
Abstract: Domestic and international air travel have increased to unprecedented levels, with domestic demand forecast to double over the coming decade. The nation's National Airspace System is now operating so close to its maximum capacity that virtually any disturbance, from a modest weather front to a temporary closure of a major airport runway, causes flight delays throughout the nation. For airlines attempting to maintain their flight operating schedule and myriad of dependencies between flights, the impact of delays are magnified. Effects include delays to departing spoke flights, disrupted ground turn operations including baggage transfer and fueling, and reassignment of flight crews and aircraft equipment. Timely awareness of delayed or unexpectedly late flights can lead to improved operations. An airline has the ability to minimize disruption to its operating schedule when flight arrival deviation information is provided with sufficient time for mitigation. This paper describes a system that blends arrival event predictive information, from both airline and air traffic sources, in determining a blended and best prediction of arrival events. To date, system and data access limitations have prevented airlines from integrating such disparate data elements. The data integration presented incorporates Kalman filtering to account for characteristics of each data source, and an airline user interface that ensures effective use of the improved information. Demonstrations of the prototype system have been well-received by airline operators. * Director, Business Development. Associate Fellow, AIAA. f Director, Technology. Member, AIAA. * Director, Technology. 0 President. Copyright 2001 The American Institute of Aeronautics and Astronautics Inc. All rights reserved. INTRODUCTION World-wide air traffic operations are expected to double from the current 2 million revenue passenger miles to over 4 million in 2010. Some data suggest that the current air traffic system can not adequately handle today's operations, e.g. the number of delayed flights at U.S. airports has increased by 53% over the past two summers [Ref. 1,2]. Such delays produce frustration for a number of groups, spanning the airline passenger to airport personnel. These delays also present great challenges to airline operators, who attempt to maintain their schedule of flight operations despite the unknowns caused by such air traffic delays. An airlines' operational efficiency, and thus economic viability, is largely governed by their ability to maintain their flight operating schedule. Disruption to that flight schedule, even by one delayed aircraft, can cause dramatic downstream effects, including missed passenger connections, incorrect airport gate assignment, and loss of flight crew availability [Ref. 3]. Assistance in identifying delayed, or early, unbound flights can translate to better management of errant flights and the minimization of their impact to maintaining flight schedule. Current airline knowledge of flight arrival time, and the associated management of airline flight schedule using that arrival time estimate, is performed primarily if not exclusively using the airline's historical database of fight operations, with limited enroute pilot updating. Once airborne, the expected time of arrival is predicted by adding the historical enroute flight time for that route to the actual takeoff time. A pilot typically tracks the actual flight against times predicted along waypoints, and then provides a measure of any differences for updating of airline dispatch and ground databases. This approach provides a fairly accurate prediction of arrival time. In today's environment, however, when most airports are confronted with more arriving aircraft than their
Publication Year: 2001
Publication Date: 2001-08-06
Language: en
Type: article
Indexed In: ['crossref']
Access and Citation
Cited By Count: 1
AI Researcher Chatbot
Get quick answers to your questions about the article from our AI researcher chatbot