Quantifying the tradeoff between fairness and delay in traffic flow management and planning: A queueing-theory approach
Abstract:
We examine the tradeoff between fairness and delay in air traffic flow management and planning, using an M/D/1 queue bank model. To motivate our flow-based approach to studying fairness, we first give several illustrative examples that characterize fairness (and performance in terms of delay/cost) in the current United State National Airspace System. We then pursue analysis and design of fairness, using the queuebank model. We find that optimal allocations of capacity resources (with respect to average-delay measures) are unfair: much larger delays are imposed on minor ows as compared to major ones at the optimum. Also, the fairness of the optimal allocation is shown to exhibit a dependence, albeit a weak one, on the congestion level. We then explore how much additional delay cost is imposed if a certain fairness level is imposed, and find that significant improvement in fairness can sometimes be achieved with moderate increase in cost. Finally, we briefly note that the fairness issue studied here is analogous with the fairness concerns that arise in bandwidth allocation problems for communication networks, and suggest that further examining these commonalities may be illuminating. Copyright © 2009 by the American Institute of Aeronautics and Astronautics, Inc. All rights reserved.
Año de publicación:
2009
Keywords:
Fuente:

Tipo de documento:
Conference Object
Estado:
Acceso restringido
Áreas de conocimiento:
- Transporte
- Optimización matemática
Áreas temáticas:
- Transporte
- Cultura e instituciones
- Principios generales de matemáticas