Gone in a Flash
By: Mr. Wilson on
January 23, 2013
Anybody who has driven through Lincoln at night knows how infuriating it is to be stopped over. And over. And over again at intersections when there are no other vehicles on the road for blocks in every direction. One proposal, which I frequently and enthusiastically support, is to convert most signaled intersections to flashing yellow / flashing red lights so that one direction of traffic doesn't need to stop at all, while the other direction can proceed after stopping. The approach makes logical sense and it is used effectively in many communities around the country.
Lincoln won't be so lucky any time soon, according to a piece by Nancy Hicks in this morning's Journal Star. Omaha has used the flashing light system after 10pm for three or four decades and it "has worked well for the city", according to their traffic manager. However, apparently research by the Federal Highway Administration says flashing lights are unsafe. I'm not sure what research was being referenced in the article, but I did find this case study from Winston-Salem, North Carolina.
The scant data available online for that study do suggest that nighttime flashing lights increase right-angle collisions. That's not too surprising; it's reasonable to hypothesize that drivers might assume that intersections with flashing lights are safer to drive through without stopping. After all, traffic engineers wouldn't put flashing lights at a busy intersection, right? And of course the drivers would be wrong. So +1 for the "no flashing lights" argument.
The data also suggest that crash numbers went down regardless of overall traffic volume. Traffic volumes at some intersections went down; some stayed the same; and others increased. In every case, crash counts decreased. Notably, injury crashes seem to have decreased even faster than total crashes, indicating that the accidents that did occur after the switch to normal light operation were less severe. It's easy to understand why: drivers are more likely to drive slower as they approach a static red light than a flashing one. +1 for no flashing lights.
And yet I'm going to continue to advocate for Lincoln to move to flashing lights at night. Why? This case study clearly shows that flashing lights are more dangerous. Am I nuts?
No. First, this is merely a case study. It's a small, deliberately-selected set of data from a single community. The intersections chosen for intervention were chosen because they featured high crash counts. The fact that changing the signal pattern decreased crashes does not suggest that all use of flashing signals is bad. Instead, it demonstrates that these specific intersections never should have used flashing signals in the first place. The data available here are far too limited for us to draw any conclusions about the effectiveness or safety of flashing signals in Lincoln. I assume there are more studies available somewhere out there. A broader analysis of the data will give us a better understanding of how these signals might work locally. Show me the data and, if they are compelling, I will change my mind.
Furthermore, I'm not worried about intersections like 27th and O. They should stay as they are. I'm concerned with ridiculous situations like those at Lucille and Pioneers, and Pioneer Woods and Pioneers; or 84th Street at Hazelwood, Sandalwood, Rockledge, or Firethorn; or 40th and Wildbriar; or many more. (These just happen to be situations I bump into the most frequently.) There is no reason for drivers on the major streets to have to stop for nobody (as I so often do), and there's no reason for drivers on the minor streets to have to sit around twiddling their thumbs and wasting gas waiting for a green light when there's nary a car in sight. The status quo does more to promote global warming than traffic safety.
Besides, it's not like Lincoln's traffic engineers are afraid to let efficiency trump safety. Take a look at all of the intersections around town with curbs arced to promote right turn speed rather than pedestrian safety. The smoother turns lead to fewer rear-end collisions, yes, but at the expense of forcing pedestrians to deal with swifter (and therefore more dangerous) traffic.
Traffic management, like all things, is a matter of balancing priorities. We cannot have perfect traffic efficiency, perfect safety, and low costs. But we can do our best to explore the available options and make adjustments as necessary. In this case I think adjustments are prudent. But heck, why not let data drive this decision? First we need to gather up all of the studies that talk about this sort of thing. Based on our analysis, Lincoln should then run its own experiment. Start by choosing five to ten "very low risk" intersections and give the flashing lights a try. Observe the data. If the data look good, expand the trial. Continue that pattern for a couple years and report back. I suspect we'll see no substantial sacrifice of safety, with healthy improvements in local traffic efficiency. Perhaps not. At least we'll know.