At the CTS conference yesterday at the University of Minnesota, I was chatting with a traffic engineer who relayed an interesting anecdote. As a traffic engineer, he is used to addressing packed rooms of people, all filled with firmly held convictions on the way things should be done. He was chatting with a colleague, a civil engineer, about whether people ever offered any input at meetings concerning things like sewer systems. The answer was no.
It should be said that I’m of the opinion that, particularly in some local jurisdictions, community residents might actually have a better idea of how to control their streets than engineers working with standardized approaches; and that, too often, streets are merely viewed as sewers of a sort, channels for simply moving as much stuff — i.e., cars — as possible, with insufficient thought for other considerations.
But the engineer also had a quite valid point, which beleaguered traffic engineers face every day at town meetings across America when trying to, say, tout the benefits of a roundabout. Suddenly, there will be a volley of criticism: Those things are dangerous, they will make traffic worse, etc., despite all statistical evidence to the contrary. Of course, people offering these opinions typically never have actual evidence, nor have they studied the problem in depth, and yet they feel comfortable to make diagnoses on engineering problems which it seems they would not feel comfortable doing in any other arena.
I thought of this morning when reading a dispatch on how Kansas City is going to introduce ramp metering to its highways (thanks Bryan).
This, not surprisingly, prompted a letter in the local paper:
Metering entrance ramps to I-435 is a terrible idea (5/13, National/Local, “Engineers turn to ramp meters to ease gridlock”). I travel to Milwaukee several times a year, and they have metered ramps onto I-94. They slow traffic down, especially during rush hour.
The meters back cars up off the ramp onto the streets, which have intersections with stoplights, and no one can go anyplace. Half the time there is plenty of room for cars to merge in on the interstate, but because of the meter you have to stop and wait.
People who live in Milwaukee and drive on the interstates hate them. Each time I drive up there I can’t wait to get back to Kansas City, where we know how to let people get around.
Ramp meters, as I mention in Traffic, are a particular case where the individual windshield perspective of drivers cannot account for the larger flow of the traffic system, with its multitude of variables; user optimality trumps system optimality in the mind of the driver. As one engineer told me, people ask me, why are you stopping me, the highway’s moving? The highway’s moving because we’re stopping you.” But hold on, K.C. engineers, throw out those models, rip up those studies — we’ve got a driver who “travels several times a year to Milwaukee,” where “everybody” hates ramp meters (everybody, except, presumably the people who are moving more smoothly than they would be without them). But there is “room” on the highway for people to enter, this driver notes. “Room,” or “capacity” as engineers might more properly say, is, alas, not the only variable to consider in highway flow, and indeed, squeezing another driver into that “room” might disrupt the flow, pushing the stream past its critical density, plunging the system from a congested synchronous flow into stop-and-go congestion. Of course, some ramp metering schemes do send traffic back up the ramps — but one might also note that those ramps might typically be backed up already, and that in some cases this is actually made worse without ramp meters. And I need hardly point out that one cannot judge the success or failure of a ramp metering scheme simply by judging one on-ramp at any one time — rush-hour traffic on a congested urban system is an incredibly complex array of networks and flows that are well beyond the ability of any one driver to fully intuit what is going on.