“Vision Zero” and Corrections to Intersections

A July 1, 2015 story in the local newspaper Metro told of the most dangerous intersections in Philadelphia:

The report was released to boost “Vision Zero,” an ambitious plan currently being spearheaded by the BCGP. Together with political and community leaders, including Councilwoman Cindy Bass, BCGP hopes to make policy changes to bring Philadelphia’s traffic fatalities and serious injuries down to — you got it — zero.

I took a look at the Vision Zero information and saw a well-intentioned program, but one that struck me as so broad that it was the equivalent of the cliche “boiling the ocean.” Yes, drivers need to slow down, but surely the problems of these intersections could have some more concrete explanations or solutions. So I examined three of the top 10 Philly intersections:

2. Broad and Race Streets (17 crashes).
8. Market and 11th Streets (14 crashes).
10. Market and 12th Streets (13 crashes).

I don’t have details on the types of crashes, or any of the logistics, but I did go to each intersection and take a careful look. Here’s what I saw:

1. All three are where a one-way street crosses a major 2-way street.
2. There was not really any substantial visibility problem… cars could see all of the pedestrians easily. Only the Race street intersection had one building that could reduce the advance visibility a bit, relative to the vehicle speed.
3. All three are flat in all directions, with no curves, so there is no sense of deception or confusing backgrounds causing reduced reactions.
4. All three are right-angles all around… no “jogs” or peculiar feeding patterns… so these are conventional intersections in every way.
5. The traffic lights seemed pretty modern. At Race I noticed they were suspended over the middle of the lanes. Probably the same at Market.

So where’s the danger? What circumstances are contributing to the accidents?

• All three are at high volume spots for cars and pedestrians… no surprise of course.
• All three have special buildings nearby. Race has a hospital; both Market Street locations surround the Reading Terminal train station. So you could say there’s a commonality in that pedestrians might enter these intersections with greater states of disorientation (from travel) and/or distraction (from health problems). The presence of a hospital could also suggest more infirmity… more pedestrians with less perfect faculties. (Note that there’s also a hospital at intersection number 5, Broad Street and Olney Avenue.)
• Race Street is a major long-range travel artery, providing access to the Ben Franklin bridge. At midday it wasn’t terribly hectic, but even then there were instances of cars stuck blocking the opposite traffic as the light changed. Eastbound drivers feel pressure to get through the light or risk waiting a very long time.
• Both Market Street intersections are probably major north/south arteries since so many of the nearby alternatives are not as good options. Again, a volume factor even if there’s no special urgency to crossing.

I’m left with only one conclusion that distinguishes the special dangers of these intersections, beyond the mere aggressiveness of drivers, which is understood from the outset: the pedestrians at these three intersections are too close to the “intersection,” by which I mean the precise square where the paths of the cars intersect. Take a look at the following diagrams:

First, here’s the car-to-car ‘intersection’:

And here’s where pedestrians walk, and where they might be hit by a car:

So the gap or margin between these two areas, shown in red, in the next diagram, is essentially zero, none. Both areas are immediately next to one another.

In fact, the only incident I saw when watching all three intersections was at Race Street, and demonstrated the problem: a pedestrian stepped immediately in front — 2 feet or less from the bumper — of a stopped vehicle THAT HAD A GREEN LIGHT and was properly waiting for room to clear in the intersection before proceeding. The pedestrian did this because there was a chance of crossing the street sooner, without even waiting for his own direction to turn green. As room in front of the vehicle cleared, the pedestrian saw this, and backed up onto the sidewalk without incident.

Why is this lack of margin a problem… why should the two types of “intersecting” spaces be separated? Because it requires drivers and pedestrians to make multiple decisions at once, and too many minor mistakes happen.

Drivers and pedestrians make mistakes every day… and it’s rarely a problem. For instance, a driver starts to change lanes before perfectly guaranteeing the clear lane; a pedestrian starts to cross on the wrong color light, or doesn’t notice an oncoming car before taking a first step. These mistakes are constant and, with millions of driving/walking decisions every day, occur at every intersection every day. But they don’t generally result in accidents because of two factors: 1) two parties have to make a mistake at the same time and place (otherwise the other party prevents the collision), and 2) most places have margin for error. Highway lanes have 4–6 feet between them. Most intersections have space to live… despite the occasional pedestrian mistake. The three intersections I examined have no margin for error.

Solution

There are many measures that can be considered that might mitigate the problem. If one could snap one’s fingers and institute whatever change one wanted — the insurance industry calls them ‘controls’ — I’d recommend several measures to literally intimidate drivers into approaching the intersection more slowly. The extreme example is to put up concrete parapets between lanes, such as at toll booths. But this could cause problems for trucks to turn if they were immovable. And reducing speeds isn’t entirely a solution. So let’s skip past the speed control measures and get right to an idea that will increase the margin for error.

Here’s what Broad and Race looks like. Notice, as shown by the dashed red line, that the white-striped crosswalks go to the very edge of the intersecting streets! In fact, one of the first observations I made was that pedestrians are incredibly trusting of the cars and even large trucks passing inches from them:

And here’s how to prevent the loss of countless limbs and lives at these and other similar intersections:

I’m suggesting that

1. The same parapets that are used for temporary construction projects be placed at the intersections, to force pedestrians away from the central square at which the vehicle traffic intersects. Yes, the ‘curb cuts’ for wheelchairs would need to be redone. (In the meantime, the parapets could have a gap with a movable bar or chain across the top.) But there’s really no other genuine objection, only the false logic of predictable naysayers. Pedestrians would still be visible long before the cars near the intersection. And cars would be quite capable of stopping before the crosswalk. Most importantly this solution is 99% “thing” and 1% “process,” meaning that the parapets simply need to be put in place to try it out. Maybe new “stop bars” could be painted for the cars, a trivial expense.
2. UPDATE NOVEMBER 16, 2019: for proven-deadly intersections, make the crosswalks raised about 2 inches as is commonly the case at airports now-a-days, and put diagonal yellow stripes on the ramp (upward angle) approach! It would take an outrageously egregious recklessness to hit a pedestrian with such forewarnings.

The first step, however, is simply to recognize that these are not merely regular intersections; they are places where pedestrians go to die. They call for special action. Are you going to help take action or help make excuses?

Originally written 4th July 2015 by Jack Bellis.