Problems and Models in Strategic Air Traffic Flow Management

Abstract

The thesis comprises of three essays.

The first essay is titled "Do more US airports need slot controls? A welfare based approach to determine slot levels." It analyzes the welfare effects of slot controls on major US airports. We consider the fundamental trade-off between benefits from queuing delay reduction and costs due to simultaneous schedule delay increase to passengers while imposing slot limits at airports. A set of quantitative models and simulation procedures are developed to explore the possible airline scheduling responses through reallocating and trimming flights. We find that, of the 35 major US airports, a more widespread use of slot controls would improve travelers' welfare. The results from our analyses suggest that slot caps at the four airports that currently have slot controls (Washington Reagan, Newark, New York LaGuardia, New York John F. Kennedy) are set too high. Further slot reduction by removing some of the flights at these airports could generate additional benefits to passengers. Slot controls can potentially reduce two thirds of the total system delays caused by congestion. A number of implementation and design issues related to the use of slot controls are also discussed in the paper.

The second essay is titled "Designing the Noah's Ark: A Multi-objective Multi-stakeholder Consensus Building Method." A significant challenge of effective air traffic flow management (ATFM) is to allow for various competing airlines to collaborate with an air navigation service provider (ANSP) in determining flow management initiatives. This challenge has led over the past 15 years to the development of a broad approach to ATFM known as collaborative decision making (CDM). A set of CDM principles has evolved to guide the development of specific tools that support ATFM resource allocation. However, these principles have not been extended to cover the problem of providing strategic advice to an ANSP in the initial planning stages of traffic management initiatives. In the second essay, we describe a mechanism whereby competing airlines provide ``consensus'' advice to an ANSP using a voting mechanism. It is based on the recently developed Majority Judgment voting procedure. The result of the procedure is a consensus real-valued vector that must satisfy a set of constraints imposed by the weather and traffic conditions of the day in question. While we developed and modeled this problem based on specific ATFM features, it appears to be highly generic and amenable to a much broader set of applications. Our analysis of this problem involves several interesting sub-problems, including a type of column generation process that creates candidate vectors for input to the voting process.

The third essay is titled "Strategic Opportunity Analysis in COuNSEL -- A Consensus-Building Mechanism for Setting Service Level Expectations." The consensus-building mechanism described in the second essay has been accepted as a technically viable solution for the stated problem -- although many practical challenges still remain before it may be deployed in operations. A key issue worthy of further investigation is its strong strategy-resistance -- as claimed by the authors of Majority Judgment, the voting procedure embedded in COuNSEL. Using the broad ideas of Nash Equilibria, we characterize the necessary and sufficient conditions for an airline to benefit from unilaterally deviating from truthfully grading one or more candidates. The framework provides the airline with all the other airlines' grades on a set of candidates, and allows it an opportunity to present new grades. The analysis is repeated over multiple instances, and likelihood of beneficial strategic opportunity is presented.