Benefit

Automated first-mile transit shuttles could be three times as effective as human-operated vehicles, according to UC Davis research team.

The report, based primarily on transit patterns in San Francisco, finds a wide variety of possible applications for automated vehicles and shuttles, highlighting effective implementations.


9/1/2018


Summary Information

The National Center for Sustainable Transportation, supported by the United States Department of Transportation, released a report outlining several scenarios of automated vehicle adoption in the San Francisco Bay Area. Because of their capabilities, automated vehicles have the capacity to significantly alter transportation patterns across the country. The aim of the report was to understand how different adoption scenarios would affect travel.

The report was made of three parts. First, researchers used the San Francisco Bay Area Metropolitan Transportation Commission's (SFBA MTC) activity-based travel demand model to explore a wide variety of automated vehicle scenarios. Then the authors simulated the introduction of an automated taxi service; finally, they simulated a first-mile transit-access shuttle service. Overall, the intent was to understand where automated vehicles' capabilities might be best applied, and to provide insight into the relative benefits of each service.

Effects of Automated Vehicles in the Bay Area
To examine the effects of automated vehicle introduction in the San Francisco Bay Area, the authors performed a simulation using an application of the MATSim framework that included a dynamic traffic assignment model. This is the same model that is used officially by the SFBA MTC. A total of seven different scenarios of automated vehicle adoption were tested. All scenarios assumed 100 percent market penetration of automated vehicles; they primarily varied in their choice of resulting effectfor instance, increased roadway capacity, or a lowering of the minimum driving age.

Automated Taxis
The next area of investigation was the introduction of an automated taxi service, similar to Uber or Lyft, but using entirely automated vehicles to transport users. Researchers performed a literature review to determine baseline variables; for instance, the model uses an estimated fare based off of the predicted cost of operation of automated vehicles. Several scenarios were run in a MATSim model, each with different taxi and personal vehicle per-mile costs. Each scenario was iterated 50 times, which was indicated to be adequate for the purposes of reaching equilibrium in the simulation.

Automated First-Mile Transit Access
Finally, the researchers simulated a first-mile transit access service using automated shuttles. The MATSim dynamic traffic assignment model was again used. A literature review was performed to understand the factors that would influence users to switch to a first-mile service to access nearby heavy rail. Researchers then examined three different scenarios: single-occupant shuttles, and shared shuttles that either offered door-to-door service or that arranged ad hoc pickup points. All scenarios were tested for both human and automated operation. These were compared to a base case of travelers using their SOV to travel to transit.

RESULTS

Effects of Automated Vehicles in the Bay Area
The scenarios were found to increase VMT from anywhere to 2 percent to 11 percent, and to have significant but varying effects on congestion. The scenarios most effective at reducing vehicle hours of delay (VHD)by up to 84 percentwere those that imputed a significant increase in roadway capacity, with the effect being strongest in one that also significantly raised the operating cost of personal vehicles.

Automated Taxis
In all scenarios, the use of automated taxis was between 4 percent and 6 percent. In a scenario that allows travelers to switch to automated taxis from transit, transit mode share decreases by more than half. However, this shift results in significant increases in congestion owing in part to empty taxi trips, with an overall VMT increase of more than 18 percent. Accordingly, all scenarios consistently showed an increase in total travel time.
The scenarios also show most traveler behavior change in central business areas; scenarios with decreased per-mile taxi costs result in relatively higher shares of users from outlying areas in the region.

Effects of Automated Vehicles in the Bay Area
When reliant on human-operated shuttles, the scenarios were shown to be markedly less cost-efficient, representing on average twice the cost to the system of automated shuttles.

The scenario with ad hoc pickup locations was estimated to halve demand, owing to its lower convenience. On the whole, first-mile access systems impacted as little as 12 percent of potential BART travelers (for the scenario examining a human-operated, shared-pickup shuttle) to as many as 73 percent (for an automated, door-to-door, shared-use shuttle). On the whole, shifting to automated shuttles tripled user share.

Benefit Comments

No comments posted to date

Comment on this Benefit

To comment on this summary, fill in the information below and click on submit. An asterisk (*) indicates a required field. Your name and email address, if provided, will not be posted, but are to contact you, if needed to clarify your comments.



Source

Automated Vehicle Scenarios: Simulation of System-Level Travel Effects Using Agent Based Demand and Supply Models in the San Francisco Bay Area

Published By: National Center for Sustainable Transportation

Source Date: 9/1/2018

URL: https://cloudfront.escholarship.org/dist/prd/content/qt4dk3n531/qt4dk3n531.pdf

Rating

Average User Rating

0 ( ratings)

Rate this Benefit

(click stars to rate)

Goal Areas

Mobility
Efficiency

Typical Deployment Locations

Metropolitan Areas

Keywords

None defined

Benefit ID: 2018-01329