Dynamic Shared Autonomous Vehicle Fleet Operations with Consideration of Fairness
MetadataShow full item record
AbstractThe future of urban transportation has arrived, and it is moving in the direction of enabling urban mobility platforms to provide shared mobility services, accelerating the shift away from personal vehicle ownership. New companies, like Uber and DiDi, are heavily investing in developing and testing emerging mobility technologies, including shared autonomous vehicles (SAVs). The full implementation of emerging mobility technologies is expected to deliver a transformative wave of urban reform. Besides, emerging mobility technologies could offer promising sustainable solutions that would optimize the usage of limited mobility resources. For instance, shared mobility services are convenient, flexible, cost- and time-efficient, and environment-friendly. Further, fully-autonomous vehicle (AV) technology surpasses human drivers in terms of costs, driving behavior, hours of service, and compliance with the plans of fleet operators. Currently, researchers are extensively studying the operations of SAV fleets that provide on-demand curb-to-curb mobility services. Specifically, they develop traveler assignment and scheduling algorithms that aim to match each traveler with a proper vehicle and plan the schedule of the vehicle simultaneously, including picking-up and dropping-off other travelers, based on a specific fleet objective. This thesis aims to fill an existing gap in the literature regarding introducing “equitable” methods to dynamic ride-sharing (DRS) systems. Thus, to meet the rising concerns of social justice, equity, and fairness in transportation systems, this thesis introduces the proportional fairness concept to DRS systems while considering the passenger heterogeneity in terms of their valuation of in-vehicle travel time. The proportional fairness formulation seeks to balance efficiency and fairness in resource allocation problems. The proportional fairness approach is then compared to two other approaches in a simulation-based environment implemented in MATSim (i.e., an agent-based transport simulator). In a centralized-fleet setting, the first approach aims to maximize traveler utility/satisfaction, while the second approach aims to maximize the total travelers’ utility. Simulation scenarios are tested to quantify the trade-offs between fleet size and vehicle maximum allowable occupancy. The performance of the three approaches is evaluated based on various performance measures from a fleet management perspective [e.g., the ratio of zero-occupant (i.e., empty-vehicle) fleet kilometers traveled to total fleet kilometers traveled], a traveler perspective (e.g., the average traveler wait time), and equity in resource allocation perspective (i.e., the Gini coefficient).
CitationHabib, N. (2021). Dynamic Shared Autonomous Vehicle Fleet Operations with Consideration of Fairness (Unpublished master's thesis). University of Calgary, Calgary, AB.
University of Calgary graduate students retain copyright ownership and moral rights for their thesis. You may use this material in any way that is permitted by the Copyright Act or through licensing that has been assigned to the document. For uses that are not allowable under copyright legislation or licensing, you are required to seek permission.