An Automated Vehicle Mobility Service Pilot in Greater Phoenix Found that Users' Safety Perception of the Service Improved By 26 Percent Over the Duration of the Pilot.

Researchers Evaluated a Pilot on Automated Vehicle Mobility on Demand Services for People with Disabilities and Older Adults in Phoenix, Arizona.

Date Posted

An Evaluation of the Valley Metro–Waymo Automated Vehicle RideChoice Mobility on Demand Demonstration, Final Report

Summary Information

As a part of the Federal Transit Administration (FTA) Mobility on Demand (MOD) Sandbox program, Valley Metro and an automated vehicle (AV) company partnered to pilot the use of AVs for Valley Metro’s RideChoice program, a curb-to-curb individual mobility service via taxi or ride-hailing services for people with disabilities and for older adults over the age of 65 and living in Greater Phoenix. This project was a six-month demonstration of AVs SAE J3016 (Level 4, high driving automation) with a vehicle safety operator on-board that occurred between September 15, 2019– March 15, 2020. The objective of the project was to study the potential behavioral impacts of AV MOD services and to understand the potential for AVs to meet the daily needs of otherwise mobility-disadvantaged citizens. The AV rides were offered in a geo-fenced area of about 100 square miles. Three surveys (Prior-, During-, and Post-Pilot) were conducted collecting responses on the safety, convenience, willingness to ride alone or with family/friends, and the likelihood of making new trips for the new AV option from 51 participants. The researchers also analyzed trip data including trip origins and destinations, time of day, cost, travel time, wait time, and distance.

In addition to the surveys and trip data, two focus groups and a policymaker roundtable were held to gain insights on key questions that could further illuminate underlying motivations, attitudes, and perceptions that could drive the future of AV MOD services.


People living near the metro Phoenix region were recruited to participate in the automated vehicle pilot demonstration called RideChoice. Survey samples were not randomly selected as the initial recruits constituted a self-selected sample of people who were eligible for the AV service and who responded to the Expression of Interest survey (a survey conducted before the Prior Survey).

Pilot participants had pre-existing access to subsidized ride hailing using a mobile app; the choice to hail on-demand trips in AVs with a vehicle safety operator on-board was added to the mobile app.

Participants were surveyed before (51 valid responses), during (35 valid responses), and after service availability (39 valid responses). Each survey had three sections including demographics, transportation choices, and “Thoughts About Self-Driving and On-Demand Mobility Services”. The Post-pilot survey also asked questions about employment status and incentive.



  • Based on the survey results, participants’ perceptions of the safety of AV MOD services showed an improvement after the pilot was introduced, with the percentage in agreement (i.e., self-driving cars are safer than human-driven cars) increasing from 49 percent to 61.8 percent in the before and during surveys (a nearly 26 percent increase).
  • A majority of the participants also considered AV MOD services to be safer than traditional RideChoice services. Among participants who used AV and non-AV MOD services during the pilot, 29 percent strongly agreed that traditional RideChoice services were safe, and 70 percent strongly agreed that AV services were safe. Additionally, participants were generally more satisfied with the travel times, costs and reliability for RideChoice Waymo services than for traditional RideChoice services.
  • Among the participants who used AV MOD service during the pilot, 65–85 percent of them agreed that AV MOD service would make travel easier and less stressful. 
  • Trip data showed that participants who used AV MOD services made more trips on AV vehicles as a result of the new AV option. In the During Survey, 59 percent of participants indicated that they were taking more trips in the RideChoice program since AVs were introduced. Participants also reported an expectation of making more trips when AV MOD services become a permanent part of RideChoice options.

An Evaluation of the Valley Metro–Waymo Automated Vehicle RideChoice Mobility on Demand Demonstration, Final Report

An Evaluation of the Valley Metro–Waymo Automated Vehicle RideChoice Mobility on Demand Demonstration, Final Report
Source Publication Date
Stopher, Peter R.; Tassio B. Magassy; Ram M. Pendyala; Devon McAslan; Farah Najar Arevalo; Thaddeus Miller;
Prepared by Arizona State University for US FTA
Other Reference Number
FTA Report No. 0198
Results Type