Metrics Matter: The Road Congestion Edition

From the Commonwealth of Virginia, we read the following (boldface mine):

But the Virginia General Assembly requires NVTA to prioritize projects that reduce road congestion. Before NVTA can fund any projects, the Virginia Department of Transportation (VDOT) has to run each proposal through a computer model that rates its ability to reduce congestion.

VDOT’s rating system for NVTA projects rewards expansions of the busiest highways, on the assumption that more road capacity will reduce congestion. It’s a flawed 20th century metric that ignores decades of real world experience that bigger roads actually make congestion worse.

The VDOT system does not measure things like how a project might benefit safety, or increase accessibility, and doesn’t take into consideration how land use changes are driven by infrastructure.

The biggest problem is simply that VDOT’s model doesn’t know what to do with short distance trips, which are the exact type of trip that transit-oriented development produces more of. So when a transit or pedestrian project makes it possible for thousands of people to walk two blocks instead of drive five miles, the VDOT model doesn’t always show that as reducing congestion.

Thus, road expansion projects end up looking good, and other things have trouble competing. Transit does OK if it relieves traffic on a major road, but pedestrian or bike projects are almost impossible.

Many other regions are using broader metrics for measuring transportation performance and congestion mitigation, but Northern Virginia can’t because the General Assembly won’t let it.

How you measure things–which is often discounted as a mere ‘technical’ detail–is really crucial. Those technical details reflect our priorities. It’s really easy to ‘define away’ policy options, especially when a really boring mechanism is used to do so.

And the congregation responds: This is yet another reason why we can’t have nice things.

This entry was posted in Transportation. Bookmark the permalink.

1 Response to Metrics Matter: The Road Congestion Edition

  1. dr2chase says:

    Accurate traffic simulation is hard. About ten years ago I attended a conference where favorable results of a then-new computationally intensive method were presented — this simulated traffic flow on a car-by-car basis, and produced more realistic (and notably different) results from older methods.

Comments are closed.