Modeling and Managing a Transportation System With Shared Mobility Services

Abstract

This thesis mainly focuses on modeling and managing a transportation system with shared mobility services. Shared mobility services such as ridesharing (or carpooling) and ride-sourcing services have attracted growing attention. However, the impacts of ridesharing on dynamic traffic are not fully clear. Besides, how to propose effective measures to manage city traffic considering shared mobility services is also an urgent issue to be addressed. Furthermore, with the growing number of electric vehicles in ride-sourcing markets, it is important to study how many charging piles to be installed to satisfy the charging demand and how to reduce the impacts of the cruising of electric ride-sourcing vehicles (ERVs) when waiting for trip orders.

The research purpose of the thesis is threefold. (i) In order to investigate the impacts of ridesharing on travelers' choices in a multi-modal transportation system, this thesis examines the ridesharing problem within a doubly dynamical framework with both day-to-day traffic variation and within-day traffic dynamics. And congestion pricing schemes are proposed to reduce network congestion and improve traffic efficiency. (ii) In terms of management strategies for stimulating carpooling to improve transportation system efficiency under the limited road space, this study investigates the road capacity allocation scheme and step tolling-rewarding scheme for carpooling. (iii) From the perspectives of planning and operation, this thesis investigates the provisions of parking spaces and charging piles for electric vehicles in ride-sourcing markets. Some interesting findings are given. Ridesharing may not necessarily reduce traffic congestion in some cases and the proposed dynamic congestion pricing schemes could reduce congestion. Under the proposed joint temporal capacity allocation and step tolling-rewarding scheme, carpooling should not be rewarded on some occasions, and instead should be penalized to avoid overcrowded carpooling flows in order to improve system efficiency. In addition, by providing parking spaces for ERVs, ride-sourcing demand increases, charging demand reduces, and profit and social welfare increase.

Overall, this thesis increases the understanding of shared mobility services in a transportation system, and the insights from the proposed strategies can guide transportation agencies in traffic management and planning as well as transportation companies in fleet operation.