Joint Optimization Scheme for the Planning and Operations of Shared Autonomous Electric Vehicle Fleets Serving Mobility on Demand

Joint Optimization Scheme for the Planning and Operations of Shared Autonomous Electric Vehicle Fleets Serving Mobility on Demand

TitleJoint Optimization Scheme for the Planning and Operations of Shared Autonomous Electric Vehicle Fleets Serving Mobility on Demand
Publication TypeJournal Article
Year of Publication2019
AuthorsColin Sheppard, Gordon S Bauer, Brian F Gerke, Jeffery B Greenblatt, Alan T Jenn, Anand R Gopal
JournalTransportation Research Record: Journal of the Transportation Research Board
Pagination036119811983827
Date Published04/2019
ISSN0361-1981
Abstract

As the transportation sector undergoes three major transformations—electrification, shared/on-demand mobility, and automation—there are new challenges to analyzing the impacts of these trends on both the transportation system and the power sector. Most models that analyze the requirements of fleets of shared autonomous electric vehicles (SAEVs) operate at the scale of an urban region, or smaller. A quadratically constrained, quadratic programming problem is formulated, designed to model the requirements of SAEVs at a national scale. The size of the SAEV fleet, the necessary charging infrastructure, the fleet charging schedule, and the dispatch required to serve demand for trips in a region are treated as decision variables. By minimizing both the amortized cost of the fleet and chargers as well as the operational costs of charging, it is possible to explore the coupled interactions between system design and operation. To apply the model at a national scale, key complications about fleet operations are simplified; but a detailed agent-based regional simulation model to parameterize those simplifications is leveraged. Preliminary results are presented, finding that all mobility in the United States (U.S.) currently served by 276 million personally owned vehicles could be served by 12.5 million SAEVs at a cost of $ 0.27/vehicle-mile or $ 0.18/passenger-mile. The energy requirements for this fleet would be 1142 GWh/day (8.5% of 2017 U.S. electricity demand) and the peak charging load 76.7 GW (11% of U.S. power peak). Several model sensitivities are explored, and it is found that sharing is a key factor in the analysis.The transportation sector represents the fastest-growing segment of the world’s greenhouse gas (GHG) emissions, with cars accounting for 8.7% of global energy-related carbon dioxide emissions in 2013, and car sales set to more than double by 2050 (1). In 2017, the transportation sector became the largest emitter of greenhouse gases in the United States (U.S.), overtaking emissions from the electric power industry (2). Transportation, therefore, represents one of the primary challenges to achieving deep decarbonization of the U.S. economy (34).Plug-in electric vehicles (PEVs) have emerged as a market-ready technology with the potential to dramatically reduce the carbon intensity of private transportation (56). Prior research has proven the capability of PEVs to meet the travel needs of the majority of drivers in the U.S. (78). Nine U.S. states (California, Connecticut, Maryland, Massachusetts, New Jersey, New York, Oregon, Rhode Island, and Vermont) have established zero-emission vehicle mandates which combined will lead to deployment of 12 million vehicles, mostly PEVs, in the U.S. by 2030 (911).Simultaneously, other important trends are emerging in the transportation sector. This study attempts to align these trends in a coupled evaluation of electric vehicles with shared autonomous on-demand mobility services. In the remainder of the introduction, future trends in transportation are examined and their potential impact on electrification is discussed. This is followed by an overview of analytical approaches that have been employed to model PEV usage which are drawn upon for this work.

DOI10.1177/0361198119838270
Short TitleTransportation Research Record