Modeling of free flight
Abstract
The goal of this research is to develop parsimonious models suitable for exploring aviation policy issues surrounding free flight. The result is an Oligopoly Model of aviation congestion. Although limited by its assumptions and a lack of data with which to calibrate it, the oligopoly model is confirmed as a valuable tool. The unique features of the Oligopoly Model are its capture of congestion effects over regions, as opposed to network links of classic transportation application, and its capability to produce tractable results across a wide spectrum of assumptions about the control environment by simply treating flight-route decision makers as being organized into distinct groups who optimize their individual interests. It is established that a unique equilibrium congestion pattern will result, under reasonable congestion function assumptions, regardless of the number of groups. With a single group, the limiting system case is obtained, in which all routes are chosen by a single decision maker. This could never be realized in a free-flight environment, because there are too many decisions for any central control. Still, the system equilibrium provides a useful best-case bound. Letting the number of groups grow to infinity, so that each pilot is choosing their own route, leads to the user equilibrium reference solution. While not a worst case, it presents a threshold that any actually implemented control policy should meet or exceed because pilots are taking no advantage of opportunities to gain efficiency through collaboration. What is likely to be the outcome of free flight is an intermediate case with an oligopoly of airlines. Parametric, computational investigations focused on this case. A small example establishes the non-intuitive fact that competition can result in aggregate total congestion even worse than the user case. Larger experiments confirm this phenomenon and show that it persists as free flight introduces new connectivity between different air-space points. Furthermore, studies varying the degree of control individual airlines have over particular origin-destination markets demonstrate that the incremental aggregate cost of oligopoly-equilibrium grows with market competition. These results suggest that free flight's performance may depend significantly on the regulator's ability to guard against monopolistic market behavior.
Degree
Ph.D.
Advisors
Rardin, Purdue University.
Subject Area
Industrial engineering|Operations research
Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server.