Experience evaluating an intersection collision warning system in Duluth, Minnesota.
Duluth, Minnesota, United States
Advanced LED Warning System for Rural Intersections: Phase 2 (ALERT-2)
Summary Information
This study evaluated the impacts of a dynamic warning system developed to improve safety at intersections having limited sight distance in Duluth, Minnesota. Using vehicle detectors, wireless communications, and signs with LED flashers the system alerted drivers of potential conflicts with opposing traffic at a two-way stop intersection obscured by hilly terrain.
The system dubbed ALERT (Advanced LED Warning System for Rural Intersections) was implemented and evaluated in two phases. Experience gained from the first deployment (ALERT-1) was used to improve the second deployment (ALERT-2). Both applications included Cross Traffic warning signs with blinking LEDs on the main highway, Vehicle Approaching warning signs with LEDs on minor approaches, and a set of non-intrusive sensors and wireless communications needed to activate and deactivate blinking LEDs.
The design of the ALERT-2 system included additional LED blinkers on minor roadway stop signs to further reduce intersection roll-throughs that were frequently observed during no-conflict conditions. These blinkers were activated when a vehicle passed a Stop Ahead sign located upstream from the intersection, and then deactivated when the vehicle arrived at the stop sign.
Field Testing
ALERT-1 and ALERT-2 testing were conducted at two separate locations to eliminate potential skewed results from driver learning effects. This report focused on findings from ALERT-2 but referenced ALERT-1 for comparison purposes. Traffic and video data were collected before and after ALERT-2 was installed. Prior to installation, 42 days of traffic data and 15,893 vehicle movements were recorded from June 26 to August 15, 2011. After installation 258 days of traffic data and 117,854 vehicle movements were recorded from September 5, 2011 to June 26, 2013.
Traveler Surveys
Surveys were mailed to local residents living within a two mile radius of the intersection. Each letter included a short explanation of the purpose of the survey, five short questions, and free space for comments. A pre-stamped self addressed return envelope was included with each survey letter to encourage participation and anonymity. A total of 206 letters were sent out, and 119 (58 percent) surveys were completed and returned.
FINDINGS
Field Testing
- After ALERT-1 and ALERT-2 features were installed, roll-throughs decreased by 22 percent.
- The ALERT-2 mail-in survey found that 92 percent of respondents either "strongly agreed" or "agreed" that the system improved safety at the intersection. Previously, during the ALERT-1 project this number was 72 percent.
- Overall, 87 percent of respondents rated the effectiveness of the warning system as excellent or good.
| 1. On average, how many times a day do you travel through the intersection of Lismore Road and Lakewood Road? | 0 | 1 | 2 | 3+ |
|---|---|---|---|---|
 | 13 (11%) | 24 (20%) | 55 (46%) | 27 (23%) |
| 2. How much do you agree with the following statements: | Strongly Agree | Agree | Disagree | Strongly Disagree |
|---|---|---|---|---|
| 2a) The warning system is easy to understand. | 65 (55%) | 46 (39%) | 6 (5%) | 2 (1%) |
| 2b) The warning system improved the safety of the intersection. | 67 (56%) | 43 (36%) | 1 (1%) | 8 (7%) |
| 2c) The vehicle activated Blinker STOP signs obtain my attention. | 83 (70%) | 34 (28%) | 1 (1%) | 1 (1%) |
| 2d) The warning system could be used at other intersections. | 63 (53%) | 45 (38%) | 6 (5%) | 5 (4%) |
| 3. How would you rate the overall effectiveness of the warning system? | Excellent | Good | Fair | Poor |
|---|---|---|---|---|
| 62 (52%) | 42 (35%) | 10 (9%) | 5 (4%) |
*
