Programs & Labs
Spring 2001
Fast and fair traffic control
Ed. note: Ramp metering, the practice of regulating the rate at which cars enter an urban freeway by means of traffic signals at entrance ramps, is an increasingly common traffic management technique in metropolitan areas with extensive freeway networks. But despite the broad support that ramp metering in general enjoys among traffic managers and researchers, the benefits and drawbacks of particular metering algorithms continue to be widely debated by academics, politicians, and the press. In the Minneapolis-St. Paul metro area, three research teams studied different aspects of ramp metering during and after an experimental area-wide meter shutoff.
Outside the laboratory, the debate over ramp metering in the Twin Cities has focused more on issues of equity and efficiency than on questions of absolute system-wide performance. To a driver waiting impatiently at a ramp signal while traffic on the adjacent freeway flows smoothly along, measurements of overall system throughput take a back seat to the question of who has to wait, and for how long. These are the questions addressed by assistant professor David Levinson of the Department of Civil Engineering in his current ramp metering research.
Even with extensive data gathering mechanisms in place, it is no easy task to determine the value of ramp metering on a freeway. Multiple entrance and exit ramps serve a wide variety of trip types and numerous vehicle classifications. In addition, there are a multitude of possible measures of effectiveness (MOEs) that could be applied to ramp metering--including traffic volume, level of service to users, equity between different groups of users, number of trips, and costs associated with using the transportation system.
David Levinson (rear) with graduate students Atif Sheikh, Shantanu Das, and Lei Zhang
Levinson analyzed data gathered by pavement-embedded loop detectors and by direct observation of ramp queues. From this data he was able to calculate the amount of delay on ramps. Levinson also calculated trip time between pairs of entrance and exit ramps, termed origin-destination (O-D) pairs.
On a macro scale, Levinson says, ramp metering as it is implemented in the Twin Cities benefits drivers making long trips at the expense of those making shorter trips—as the system's designers intended. The rationale for this design decision is based on the original design goals of the interstate freeway system.
The city freeway networks that carry huge volumes of local traffic are components of an interstate highway system originally designed to carry passengers and freight on long-haul trips between widely separated urban areas. Since then, rapid suburban growth and the corresponding increase in commuter and business traffic has transformed the role of city freeways.
But this change in the way freeways are used has not always been matched by changes in the way ramp meters control traffic. The overall goal has remained one of maximizing mainline traffic flow through bottleneck areas, without regard to the effects on traffic trying to enter the freeway from streets nearer the urban core. There are, however, other possible objectives or design constraints, such as ensuring that traffic delays are distributed as evenly as possible to all users of the freeway system.
Levinson points out the example of traffic signals at the intersection of a major arterial road and a minor side street. In terms of minimum total delay, he says, it might be most efficient to give 100 percent of the green signal time to the major road and none to the minor street. Of course, traffic lights alternate so that no user is delayed for more than a few minutes, and Levinson wonders if this kind of limit on individual delay—even at the expense of overall freeway efficiency--may be necessary for ramp meters to satisfy equity considerations.
Now that metropolitan freeways have been re-tasked as high-volume commuter routes serving the still-growing suburbs, Levinson asks, what are the policy implications for transportation managers and land use planners? Additional research into the value of time spent under different traffic conditions, and data collection systems to measure ramp queues in real time, may help uncover answers to these questions.
Finally, in Levinson's view, it is unfair to think that ramp meters can carry the entire load of traffic growth, given the fact that driver demand is outpacing investment in highway capacity. While ramp meters can help at the margins, delaying the onset of freeway breakdowns and making freeways flow more smoothly, they cannot eliminate congestion entirely.