Household use of autonomous vehicles with ride sourcing. (April 2021)

• Record Type: Journal Article
• Authors: Khayati, Yashar; Kang, Jee Eun; Karwan, Mark; Murray, Chase
• Is Part Of: Transportation research. Volume 125(2021)
• Publication Date: 2021-04
• Subjects: Shared autonomous vehicles -- Self-driving -- Ride sourcing -- Autonomous vehicles -- Household activity pattern problem; Transportation -- Periodicals; Transportation -- Technological innovations -- Periodicals; 388.011
• DOI: 10.1016/j.trc.2021.102998
• Languages: English
• ISSNs: 0968-090X

Abstract

Highlights: Extended Household Activity Pattern Problem to include use of Private Autonomous Vehicles (PAVs) as well as readily available Shared Autonomous Vehicles (SAVs). New modeling capabilities include carpooling a PAV among family members, ride sourcing to SAVs, PAV waiting for pickup-dropff, PAV parking itself after pickup-dropoff. Adopted the two-stage solution method that solves the household member assignment and scheduling problem followed by the vehicle assignment and scheduling problem. Developed a comprehensive scenario analysis concerning the operational cost of PAVs and costs of SAVs and derived system insights. Abstract: The prospect of autonomous vehicles (AVs) offers the possibility that a household could reduce household owned vehicles to a single vehicle. At the same time, with AVs, a Shared Autonomous Vehicle (SAV) system will rise as a primary mode of serving travel demands. This study aims to model households' activity/travel decisions using both household owned AVs as well as readily available SAVs to perform daily activities. We formulate the Household Activity Pattern Problem with Autonomous Vehicles and Ride Sourcing (HAPPAV-RS), as a mixed integer linear program. The model generates an optimal activity/travel patterns for household members under spatial and temporal constraints. The model can capture driverless operations such as AV pick-up and drop-off, parking availability, empty trips and ridesharing among household members, as well as the use of SAVs with the request waiting time. A decomposition method is used to solve the NP-hard problem HAPPAV-RS. scenarios on using AVs and SAVs with different travel mode availability, AV/SAV cost and SAV waiting time are designed to enable sensitivity analysis. Various travel metrics such as activity pattern feasibility, household's total travel disutility, travel mode VMT and AV-SAV trip coverage are reported.