Analysis, modeling and control of the airport departure process

Abstract

Increased air traffic demand over the past two decades has resulted in significant increases in surface congestion at major airports in the United States. The overall objective of this thesis is to mitigate the adverse effects of airport surface congestion, including increased taxi-out times, fuel burn, and emissions. The thesis tackles this objective in three steps: The first part deals with the analysis of departure operations and the characterization of airport capacity; the second part develops a new model of the departure process; and the third part of the thesis proposes and tests, both on the field and in simulations, algorithms for the control of the departure process. The characterization and estimation of airport capacity is essential for the successful management of congestion. This thesis proposes a new parametric method for estimating the departure capacity of a runway system, the most constrained element of most airports. The insights gained from the proposed technique are demonstrated through a case study of Boston Logan International Airport (BOS). Subsequently, the methodology is generalized to the study of interactions among the three main airports of the New York Metroplex, namely, John F. Kennedy International Airport (JFK), Newark Liberty International Airport (EWR) and LaGuardia Airport (LGA). The individual capacities of the three airports are estimated, dependencies between their operations are identified, and the capacity of the Metroplex as a whole is characterized. The thesis also identifies opportunities for improving the operational capacity of the Metroplex without significant redesign of the airspace. The proposed methodology is finally used to assess the relationship between route availability during convective weather and the capacity of LGA. The second part of the thesis develops a novel analytical model of the departure process. The modeling procedure includes the estimation of unimpeded taxi-out time distributions, and the development of a stochastic and dynamic queuing model of the departure runway(s), based on the transient analysis of D(t)=Ek(t)=1 queuing systems. The parameters of the runway service process are estimated using operational data. Using the aircraft pushback schedule as input, the model predicts the expected runway schedule and the takeoff times. It also estimates the expected queuing delay and its variance for each light, along with the congestion level of the airport, sizes of the departure queues, and the departure throughput. The model is trained using data from EWR in 2011, and is subsequently used to predict taxi-out times at EWR in 2007 and 2010. The final part of this thesis proposes dynamic programming algorithms for controlling the departure process, given the current operating environment. These algorithms, called Pushback Rate Control protocols, predict the departure throughput of the airport, and recommend a rate at which to release pushbacks from the gate in order to control congestion. The thesis describes the design and field-testing of a variant of Pushback Rate Control at BOS in 2011, and the development of a decision-support tool for its implementation. The analysis shows that during 8 four-hour test periods, fuel use was reduced by an estimated 9 US tons (2,650 US gallons), and taxi-out times were reduced by an average of 5.3 min for the 144 flights that were held at the gate. The thesis concludes with simulations of the Pushback Rate Control protocol at Philadelphia International Airport (PHL), one of the most congested airports in the US, and a discussion of the potential benefits and implementation challenges.