Exploring the performance of autonomous mobility on-demand systems under demand uncertainty

Authored by Farid Javanshour, Hussein Dia, Gordon Duncan

Date Published: 2019

DOI: 10.1080/23249935.2018.1528485

Sponsors: No sponsors listed

Platforms: Commuter

Model Documentation: Other Narrative Flow charts Mathematical description

Model Code URLs: Model code not found

Abstract

This paper aims to explore the performance of Autonomous Mobility on-Demand (AMoD) systems under uncertain travel demands in an urban environment using a case study of Melbourne, Australia. The results of the simulation model developed in this study showed that an AMoD system could reduce the current fleet size by 84\% while still meeting the same demand for travel. This, however, comes at a cost of more Vehicle-Kilometres Travelled (VKT). The increase in VKT is significant and amounts to around 77\% for scenarios in which the vehicles are used in car-sharing systems, and 29\% for the scenarios in which vehicles are used as ride-sharing systems. These findings show that the benefits reported in other studies have mainly been overestimated. The current study has also discovered a strong quadratic relationship between AMoD fleet size and VKT.

Tags

agent-based simulation Travel Vehicles Autonomous vehicles Agent-based modelling Autonomous mobility-on-demand systems Disruptive mobility