Optimizing Urban Air Mobility Operations in a Corridor Network
Name
mcdonald-mcdonst-dual_sm-AeroAstro_and_Civil-2023-thesis.pdf
Description
Thesis PDF
Size
23.6 MB
Format
Adobe PDF
Checksum (MD5)
be3014c64ce0ff6f11102a158a6232be
Author(s)
McDonald, Spencer T.
Advisor(s)
Jacquillat, Alexandre
Date Issued
June 2023
Publisher
Massachusetts Institute of Technology
Abstract
Electric vertical-takeoff-and-landing vehicles give rise to Urban Air Mobility (UAM) concepts. Initial UAM systems are projected to operate high volumes of flights within a capacitated corridor network. This paper develops a tractable methodology to optimize vehicle dispatching and routing in UAM networks, along with flight trajectories between origin and destination, as well as flow directionality in corridors. We formulate an integer optimization model in a time-space-network that exploits a subpath structure at the flight level. We propose a column generation algorithm to decompose vehicle dispatching decisions in a master problem and flight trajectories in a pricing problem, using a tailored backward label-setting algorithm. This methodology scales to practical instances, with up to 50 vertiports, hundreds of corridor conjunctions, and 1,000 trip requests. We develop a data-driven experimental setup capturing real-world travel demand, air traffic infrastructure and weather patterns. Results demonstrate the benefits of the comprehensive optimization approach developed in this paper, as compared to benchmarks that do not capture flow directionality or operating capacities. This methodology identifies the bottlenecks created by legacy corridors, horizontal separation requirements and adverse weather to inform the design of emerging UAM systems.
MIT Department
Massachusetts Institute of Technology. Department of Aeronautics and Astronautics
Massachusetts Institute of Technology. Department of Civil and Environmental Engineering
Terms of Use
CC0 - Public Domain
Copyright Public Domain
Persistent DSpace Link
