Tampa Connected Vehicle Pilot Showed a Nine Percent Decrease in Rate of Forward Collision Conflicts and a Nearly 30 Percent Reduction in Travel Times.

Before-and-After Study of Connected Vehicle Applications Reveals Mobility Benefits in Travel Time, Idle Time and Queue Length, and Safety Benefits in Reduction in Rates of Conflicts and Public Perceptions.

Date Posted
07/27/2021
Identifier
2021-B01583
TwitterLinkedInFacebook

Connected Vehicle Pilot Deployment Program Performance Measurement and Evaluation – Tampa (THEA): Phase 3 Evaluation Report

Summary Information

The Tampa Hillsborough Expressway Authority (THEA) Connected Vehicle (CV) Pilot Deployment aimed to test a connected urban environment and measure the effect and impact of CVs in a busy downtown with tolled expressway, bus and street car service, high pedestrian densities, special events and high dynamic traffic demand over the course of a typical day. THEA deployed Vehicle to Vehicle (V2V) and Vehicle to Infrastructure (V2I) CV solutions within specific areas of downtown Tampa, Florida with the goals of enhancing pedestrian safety, speeding bus operations and reducing conflicts between street cars, pedestrians and passenger cars at locations with high volumes of mixed traffic.

The study featured a robust experimental design involving more than 1,000 private citizens for a period of over 19 months in 2018 – 2020 to experience the functionality and impact of the V2V and V2I CV applications via a Human Machine Interface (HMI) installed in participant's vehicle’s rearview mirror. The performance evaluation of this study focused on measuring, reporting, and assessing the impact and contribution of the CV technology toward enhanced mobility and safety.

Methodology

The study was split into treatment (HMI enabled) and control (HMI disabled) groups using a randomized two-to-one matching (two treatment to one control) stratified by gender, age, income, and education. The analysis relied on data collected from 49 roadside units (RSU) and travel logs stored by the onboard units (OBUs) installed in participants' vehicles.

In order to evaluate the safety benefits, the study analyzed the participants’ behavioral responses to the deployment of the V2V and V2I applications. In the absence of in-vehicle video camera detection and recording, the research team developed data-driven algorithms to detect longitudinal and lateral reactions in response to evasive maneuvers. For each applicable use case, the analysis estimated false positive, false negative, true positive, and true negative rates by using algorithms developed to emulate the applications’ logic to identify and quantify the number of V2V interactions and conflicts for each use case. The mobility impact evaluation relied on a before-after assessment using an interrupted time-series regression analysis. Participant perceptions were collected through a series of surveys based on the participant’s use of the equipment and exposure to the applications. A total of 1,058, 389 and 369 responses were collected for the initial, interim and exit survey. The six use cases that evaluated a number of different V2V and V2I applications are shown in the table below.

Table 1. Use Case Conditions and Planned and/or Deployed Applications

Use Case Number

Condition

Planned and/or Deployed V2V and/or V2I Application

1

Morning Backups

End of Ramp Deceleration Warning (ERDW)-Deployed

Electronic Emergency Brake Light Warning (EEBL)-Deployed

Forward Collision Warning (FCW) -Deployed

Intelligent Traffic Signal System (I-SIG)

2

Wrong-Way Entries

Wrong-Way Entry-Deployed

Intelligent Traffic Signal System (I-SIG)

3

Pedestrian Conflicts

Pedestrian Collision Warning (PCW)-Deployed

4

Transit Signal Priority

Intersection Movement Assist (IMA)

Intelligent Traffic Signal System (I-SIG)

Transit Signal Priority (TSP)

5

Streetcar Conflict

Vehicle Turning Right in Front of Transit Vehicle (VTRFTV)-Deployed

6

Traffic Progression

Electronic Emergency Brake Light Warning (EEBL)-Deployed

 

 

Forward Collision Warning (FCW) -Deployed

 

 

Intersection Movement Assist (IMA)-Deployed

 

 

Intelligent Traffic Signal System (I-SIG)

 

Findings

Mobility Impacts: The broadcasting of speed advisories via the speed harmonization ERDW application contributed to mobility benefits compared with the baseline (before ERDW deployment) conditions. Compared to the baseline (pre-intervention period), the following mobility benefits of ERDW were identified based on an evaluation period of 35 weekdays (Feb 3, 2020 to March 20, 2020):

  • 2.1 percent reduction in mean travel times.
  • 1.8 percent reduction in idle time or time spent traveling at less than one mile per hour.
  • 1.8 percent reduction in queue length.
  • A travel time index (measured as peak hour travel time divided by off-peak travel time) reduction from 2.7 to 1.9 (nearly 30 percent reduction).

 

Safety Impacts: In terms of safety, the following benefits were observed:

  • For Use Case 2, the WWE application correctly warned drivers entering the wrong way of the Reversible Express Lanes (REL) and identified 74 percent of the potentially true conflicts (14 participants of 19 potentially true conflicts) during the PM peak period (3 p.m. to 12 a.m. weekdays). The complexity of the warning delivery system contributed to a false positive rate of about 28 percent for PM and 2.8 percent for AM.
  • FCW and EEBL implemented in Use Case 6 issued 26 warnings classified as true positives of the 64 warnings (41 percent), of which eight were shown to drivers with the HMI enabled (i.e., treatment group).
  • Use Case 6 showed the rates of conflicts per vehicle normalized over time, showed a nine percent decrease for FCW (from 4.6 to 4.2), a 23 percent decrease for EEBL (from 2.2 to 1.7), and an increase in IMA (from 0.1 to 0.5).
  • Six percent of warnings triggered by FCW were true positives (9 out of 150) and 25 percent of warnings triggered by EEBL were true positives (1 out of 4) in Use Case 1. VTRFTV deployed 61 warnings, of which 8 on average (13 percent) were classified as true positives in Use Case 5.

Mixed results were found based on the crash data. Further evaluation is needed as crash analysis often requires longer evaluation periods since crashes are rare events.

Participants' Perceptions: Participant surveys revealed the following findings:

  • Fifty-six percent of the participants were somewhat or very satisfied with their participation in the study.
  • Seventy-six percent of the 130 respondents in the combined interim/final survey determined the ERDW application to be effective in avoiding a conflict. About 51 to 66 percent of the respondents believed the VTRFTV, PCW, IMA, FCW, and EEBL applications to be effective.
  • Both before and after the study, 66 percent of the participants perceived safety as the greatest benefit of CV technology.
  • On average, about 70 percent of the participants confirmed clear administration of the warning and 59-77 percent of the participants were satisfied with the timeliness of the warnings provided by the WWE, ERDW, and FCW applications.
  • The study also showed a 37 percent reduction in concerns about the impact of CV technology on their privacy, and a 65 percent reduction in perceived concern about receiving too many alerts.

Connected Vehicle Pilot Deployment Program Performance Measurement and Evaluation – Tampa (THEA): Phase 3 Evaluation Report

Connected Vehicle Pilot Deployment Program Performance Measurement and Evaluation – Tampa (THEA): Phase 3 Evaluation Report
Source Publication Date
03/05/2021
Author
Sisinnio Concas, Achilleas Kourtellis, Mohsen Kamrani, and Omkar DokurCenter for Urban Transportation Research, University of South Florida
Publisher
University of South Florida's Center for Urban Transportation Research for ITS JPO
Other Reference Number
FHWA-JPO-20-829
Goal Areas
Results Type