Yanto Huang | Argonne National Laboratory

Social Media Handle: 

Presentation Description

Using ride-sharing services has the potential to reduce vehicle ownership and demand for parking (Farhan and Chen, 2018; Zhang et al., 2015). However, ride-sharing services can bring curbside congestion and even increase overall carbon footprint (with low sharing rates), since they bring empty/deadheading vehicle miles traveled (VMT) when vehicles pick up a person.
This paper aims to answer a main question: when a new ride-sharing service is provided by a transportation network company (TNC) to attract travel using personal vehicles, does the TNC fleet need to be electrified in order to maintain the same level of emissions and energy consumption? If the answer is yes, how much the fleet should be electrified for that purpose, and are there any other factors that would impact such a decision? We compare the energy consumption and emissions between scenarios when people use personal cars to directly travel from origins to destinations, to scenarios using ride-sharing services with a certain number of detours likely to happen to each rider.
In this paper, the agent-based end-to-end travel demand simulator POLARIS is used to simulate travel details (Auld et al., 2016). A series of activity models are executed on the synthetic population to generate, schedule, and plan each agent’s travel day, including destination, mode, and route choices. A dynamic ride-sharing algorithm is performed to match vehicles to passenger requests, and electric TNC vehicles will go to charge when reaching a low state of charge. Simulated mesoscopic vehicle trajectories are processed through SVTrip to obtain realistic driving behavior patterns, and the vehicle systems simulator Autonomie is then able to retrieve emissions and energy consumption for the region (Karbowski et al., 2014; Moawad et al., 2016).
A baseline scenario will be set up for all potential ride-sharing users who assume to keep using personal vehicles. As their mode use shifts to ride-sharing services in our scenarios, we look at how total VMT and eVMT change across the network. Moreover, the fleet will be replaced with different EV shares to see those changes, revealed from vehicle charging behavior and their different energy consumptions.
The results of this paper provide insights into how a fleet of new TNC services could be electrified before being introduced in an area, focusing on emission and energy consumption considerations. Sensitivity analysis on fleet size, geo-fence, and study regions will reveal more information on how environmental impacts can vary.

Speaker Biography

Yantao Huang is a computational transportation engineer in Vehicle and Mobility System group at Argonne National Laboratory (ANL). He received his master’s degree in transportation engineering from Imperial College London and his Ph.D. in transportation engineering at the University of Texas at Austin. He also completed a post-doctoral appointment with the University of Texas at Austin and National Renewable Energy Laboratory (NREL). His research has led to journal publications and conference papers that focus on agent-based simulations and travel demand forecasting, including anticipating the impacts of new technologies and sustainable transportation systems, including automated cars and trucks, shared automated vehicle fleets, and electric vehicles.

Co-presenters

Krishna Murthy  Gurumurthy
Argonne National Laboratory

Charbel Mansour
Argonne National Laboratory

Natalia Zuniga-Garcia
Argonne National Laboratory

Joshua Auld
Argonne National Laboratory

Presentation File