Field Experiment of Variable Speed Advisory (VSA) Final Report

This report documents the field test of Variable Speed Advisory (VSA) which is an Active Traffic Management strategy. The test site for the VSA is on State Route 78 Eastbound (SR-78E) from Vista Village Drive (in the City of Vista) to the freeway interchange point of SR-78E and U.S. Route 15 (in the city of Escondido). This test segment is a three-lane freeway with a posted speed limit of 65 mph and it has 10 on-ramps and 10 off-ramps. The project was funded by the California Department of Transportation (Caltrans) Division of Research Innovation and System Information (DRISI) under Contract Number 65A0587. Real-time traffic detector data including flow, speed, and occupancy from pre-existing loop detectors in the field test site, were transmitted via the internet by engineers at Caltrans District 11 (D11) Transportation Management Center (TMC) to a server located in the offices of California Partners for Advanced Transportation Technology (PATH). This data was then aggregated with real time speed data, captured every 30s [seconds] by radar equipment installed along with solar panel powered LED display equipment, for the display of a VSA, at 7 different sites along a 10.8 mile section of SR 78E. These two sources of data were then processed for the estimation of the overall traffic state along the corridor, which was in-turn used to calculate the VSA for each section in order to maximize overall traffic throughput through recurrent bottlenecks on SR-78E. Calculated VSA values were then rounded to multiples of 5 mph and displayed on the VSA signs. Public outreach was conducted by Caltrans D11 Public Information Office (PIO) to educate the public about the VSA test, and encourage their compliance with posted speed advisories. A publicly accessible website was also developed for the real-time display of Google Traffic, traffic state, and VSAs displayed in the field. This site was used extensively by Caltrans management, the project team, and by the public drivers. After different stages of the system development, integration, and installation process were completed, a progressive test procedure was executed to mitigate any potential negative impacts on traffic operation. This procedure included dry-runs (saving data for analysis without roadside display), error detection, system tuning, preliminary testing, and extensive tests for data collection for four weeks. The results of the performance analysis, conducted with an independent PeMS data set, illustrated an improvement in three performance measures for the AM (6-9AM) peak hours: Vehicle Miles Traveled (VMT) increased by 2.72%; Vehicle Hours Traveled (VHT) decreased by 6.28%, and the average speed over the road segment or Q=VMT/VHT increased by 8.71%. In PM peak hours (2-7PM), two of the three performance measures improved: VMT did not have noticeable improvement; VHT decreased by 1.47% on average; and Q increased by 2.80% on average.