Results from Phase I of a Pilot Study on Dynamic Ramp Metering in Houston

Abstract

Traffic in Houston, U.S., has continued growing over the past decade. The Houston District of the Texas Department of Transportation (TxDOT) recently began a pilot study to evaluate a dynamic ramp metering system. The project is aimed to convert ramp metering from local control to system-wide dynamic operation. In Phase I of the project, major control parameters and different metering strategies were tested and evaluated in simulation and field settings for a study corridor installed with six ramp meters. The study identified a base metering plan that overall worked well for sites without restrictive queue conditions. This base plan was that average speed of 50 mph or lower in the right-most two mainlanes will call for metering at a constant metering rate of 850 vehicles per hour for at least 4 min, and that queue occupancy of 50% or higher will call for meter shutdown for at least 1 min. Ramp metering coordinated with the downstream intersection performed well by accommodating diverted traffic caused by ramp metering. When operating ramp meters in a group, metering the immediately upstream meter performed best compared with metering further upstream meters. It is evident that ramp metering caused traffic diversion to the frontage road and also caused reduced queue-jumping behaviors on the frontage road at ramp meters with an immediately upstream exit ramp. The coordinated ramp metering strategy can potentially generate a benefit/cost ratio of 117:1 compared with local metering in the District.