Date of Award

Spring 2015

Document Type

Dissertation

Degree Name

Doctor of Philosophy in Industrial Engineering - (Ph.D.)

Department

Mechanical and Industrial Engineering

First Advisor

Sanchoy K. Das

Second Advisor

Reggie J. Caudill

Third Advisor

Athanassios K. Bladikas

Fourth Advisor

Wenbo Cai

Fifth Advisor

Lazar Spasovic

Abstract

Traditionally, airlines have configured flight operations into a Hub and Spoke network design. Using connecting arrival departure waves at multiple hubs these networks achieve efficient passenger flows. Recently, there has been much growth in the development of global single mega-hub (SMH) flight networks that have a significantly different operating cost structure and schedule design. These are located primarily in the Middle East and are commonly referred to as the ME3. The traditionalist view is that SMH networks are money losers and subsidized by sovereign funds.

This research studies and analyzes SMH networks in an attempt to better understand their flight efficiency drivers. Key characteristics of SMH airports are identified as: (i) There are no peak periods, and flight activity is balanced with coordinated waves (ii) No priority is assigned to arrival/departure times at destinations (selfish strategy) only hub connectivity is considered (iii) There is less than 5% OD traffic at SMH (iv) The airline operates only non-stop flights (v) Passengers accept longer travel times in exchange for economic benefits (vi) Airline and airport owners work together to achieve collaborative flight schedules. This research focuses on the network structure of SMH airports to identify and optimize the operational characteristics that are the source of their advantages. A key feature of SMH airports is that the airline and airport are closely aligned in a partnership. To model this relationship, the Mega-Hub Collaborative Flight Rescheduling (MCFR). Problem is introduced. The MCFR starts with an initial flight schedule developed by the airline, then formulates a cooperative objective which is optimized iteratively by a series of reschedules. Specifically, in a network of iEM cities, the decision variables are i* the flight to be rescheduled, Di* the new departure time of flight to city i* and Hi* the new hold time at the destinatioin city i*. The daily passenger traffic is given by Ni,j and normally distributed with parameters µNi,j and sNi,j.

A three-term MCFR objective function is developed to represent the intersecting scheduling decision space between airlines and airports: (i) Passenger Waiting Time (ii) Passenger Volume in Terminal, and (iii) Ground Activity Wave Imbalance. The function is non-linear in nature and the associated constraints and definitions are also non¬linear. An EXCEL/VBA based simulator is developed to simulate the passenger traffic flows and generate the expected cost objective for a given flight network. This simulator is able to handle up to an M=250 flight network tracking 6250 passenger arcs. A simulation optimization approach is used to solve the MCFR. A Wave Gain Loss (WGL) strategy estimates the impact Zi of flight shift ?i on the objective. The WGL iteratively reschedules flights and is formulated as a non-linear program. It includes functions to capture the traffic affinity driven solution dependency between flights, the relationship between passengers in terminal gradients and flight shifts, and the relationship between ground traffic activity gradients and flight shifts. Each iteration generates a Zi ranked list of flights. The WGL is integrated with the EXCEL/VBA simulator and shown to generate significant costs reduction in an efficient time. Extensive testing is done on a set of 5 flight network problems, each with 3 different passengers flow networks characterized by low, medium and high traffic concentrations.

Share

COinS