Analysis of Coordinated Ramp Metering Strategies on I-80 and SR-99 Evaluated Mobility and Reliability Benefits Using Balanced Before and After Data.
Alameda County, California, United States
Contra Costa County, California, United States
Sacramento County, CA
Evaluation of Coordinated Ramp Metering (CRM) Systems in California
Summary Information
Ramp metering has been used extensively to regulate traffic at the entry points of freeway on-ramps, thereby mitigating congestion in freeway merging areas and preserving freeway capacity. Coordinated traffic responsive ramp metering systems (CRM) utilize detector data to adjust ramp meter rates to minimize delay and maximize freeway throughput. Researchers conducted an empirical evaluation of two CRM systems implemented on freeway corridors in California:
- I-80 (eastbound and westbound): a 28-mile-long corridor from the Carquinez Bridge to the MacArthur Maze, containing 44 ramp meters
- SR-99 (northbound): a 13.76-mile-long corridor from Grant Line Road to US-50, containing 13 interchanges
The primary source of field data used in the analysis was gathered from Caltrans Performance Measurement System (PeMS) detector data and vendor-provided peak period travel time and speed data, which were used to establish performance along the freeway mainline.
Methodology
In the “before” period (September-October 2016) the I-80 corridor did not utilize ramp metering. The “after” (September-November 2018) ramp metering system was a CRM algorithm based on fuzzy logic control and was active daily from 5:00 AM to 8:00 PM. This ramp metering algorithm coordinated with its nearby parallel arterial, San Pablo Avenue, to best optimize corridor-level performance in the event of any incidents. SR-99 utilized Local Adaptive Ramp Metering within the “before” period (September-October 2018), and a CRM algorithm developed by University of California, Berkeley in the “after” period (September-November 2019). To account for variations in overall traffic over time, days with similar corridor-level vehicle-miles traveled (VMT) were selected from the before and after periods to enable balanced comparisons. Analysts evaluated the changes in delay using vehicle-hours of travel (VHT) comparisons, and reliability as measured by Planning Time Index (the total travel time that should be planned when an adequate buffer time is used) and Travel Time Index (the ratio of the travel time to the time required to make the same trip at free-flow speeds).
Findings
- Overall, when comparing balanced days in the before versus after periods, the I-80 study corridor showed a three to four percent reduction in AM and PM peak period VHT for eastbound I-80, and a two to nine percent reduction in westbound VHT during peak periods. Peak period reliability as measured by Planning Time Index improved by 7 to 20 percent, and by 2 to 9 percent as measured by Travel Time Index.
- The northbound SR-99 study corridor showed an eight percent reduction in VHT during the AM peak period when comparing balanced before and after data. The improvement in corridor travel time reliability was 28 percent measured by Planning Time Index, and 15 percent measured by Travel Time Index.
- Both CRM strategies improved corridor traffic performance based on the evaluation results. However, due to differences in the before-period operations in each corridor, a comparison to determine which CRM strategy was better was not possible.
