Autonomous Mobility on-Demand in urban areas

Abstract

Due to connectivity problems, the attractiveness of public transport is limited. Policymakers aim to increase the modal share of public transport to protect the accessibility, livability, safety, sustainability and efficiency in the cities of the future. Applying Autonomous Mobility on-Demand (AMoD) systems as a feeder service for public transport hubs can improve the first- and last-mile trip leg, increasing the attractivity of public transport. It is essential for the implementation of AMoD systems to predict the impacts of varying operational strategies on beforehand. From an operators perspective, especially the financial viability of AMoD operations is vital and yet unclear. An existing gravity-based travel demand estimation model built in OmniTRANS is used to predict the AMoD passenger demand. Besides, an agent-based simulation model is developed using the software Anylogic that is connected to the demand-model as an add-on module to simulate the behavior of passengers and AMoD vehicles within an urban environment. The agent-based simulation model is applied to the case study Rotterdam-Zuid, where Station Zuidplein and Station Lombardijen function as an AMoD hub. The simulation outputs show that activating dynamic ridesharing using wireless fast chargers at the stations results in the most financially viable operation. Activating automatic relocation results in the most costly operation. Compared to existing public transport services, carsharing systems and taxi systems, the AMoD system shows to save a large amount of expenses due to the absence of drivers.