How Big Data Can Help Prevent Deaths By Speeding

Photo: Anthony Quintano/Flickr
Photo: Anthony Quintano/Flickr

What if you could reliably pinpoint the street segments all over the country where speeding traffic poses a constant threat to life and limb? That could be a very powerful tool to identify and redesign dangerous streets.

At a time when traffic fatalities are rising rapidly and the National Transportation Safety Board acknowledges that excessive vehicle speeds contribute to America’s sky-high traffic death rate, the time is ripe for that kind of analysis.

Two transportation researchers say a national map of dangerous traffic speeds could be possible. Eric Sundquist and Michael Brenneis at the State Smart Transportation Initiative report that U.S. DOT has purchased a vast set of traffic speed data collected by the firm INRIX, which it will share with state DOTs.

The feds aren’t thinking about analyzing speed as a safety problem, however. “The US DOT is buying the data from INRIX so that DOTs can comply with performance measure requirements around delay and reliability,” Sundquist and Brenneis write. In other words, U.S. DOT is thinking about congestion and traffic delay.

But with the data now in the hands of state DOTs, Sundquist and Brenneis say there’s no reason it can’t be used to save lives:

One application we haven’t seen — and we’d like to hear from anyone doing it or interested in doing so — is for speed management.

The notion is simple: The data provide average speeds by time and place, which can pinpoint congestion and reliability issues. They also indicate areas with excessive traffic speed. And while federal performance measures focus on the former, it’s possible to imagine speed management measures as well.

And in fact some exist. This is a measure established by the city of Los Angeles: “Ensure that 80% of street segments do not exceed targeted operating speeds by 2035.”

The data isn’t as fine-grained as detailed speed studies, which capture the distribution of driver speeds, not just the average. But Sundquist and Brenneis believe that average speed can still be a useful metric to identify dangerous conditions. They’re asking researchers who want to refine these ideas to contact them.

More recommended reading today: Plan Philly reports on SEPTA’s plans to improve service and increase ridership by redesigning its network of 126 bus routes. And Better Burque shares photos of “Suddenly Ending Bike Lanes” around Albuquerque.

25 thoughts on How Big Data Can Help Prevent Deaths By Speeding

  1. This is an excellent proposal. Near-term, it may be — politically — too heavy a lift.

    Contrasting / comparing: for E-ZPass to be widely-accepted, a sub-set of users demanded the assurance the data wouldn’t be used for speed enforcement.

    Instead of looking at speed (which I agree is a very likely cause of accidents) how about merely looking a locations where there’s a statistically-significant occurrence of accidents? It seems to me that with that data in hand, asking “why?” would be an unavoidable follow-up question to “where?”.

  2. Another great example of @INRIX data at work to help humans navigate our roadways, as well as save lives. Blessed to be part of it.

  3. To me that makes more sense. If there are no pedestrians or cyclists, like on limited access highways, speed really doesn’t matter. Ditto for country roads with very few pedestrians or cyclists. We’ll get a lot less push-back from drivers if we target speeding only in locations where it causes the most harm. That’s mostly urban streets.

  4. This is good, and I would like to see it in connection with some action against reckless/negligent drivers. We can reduce speed limits/redesign streets, and that will help, but some people really just shouldn’t be behind the wheel. Targeting those people will *hopefully* make the politics easier to manage. After all, people who drive as well as they claim, have nothing to fear.

  5. It’s simpler than this. Much simpler. All we have to do is look at the other countries around the world that have seen the largest drops in their traffic mortality rates and do what they do. In the meantime, why does Streetsblog continue to lead with death and carnage? How does this crap help anyone who isn’t already on a bicycle choose to start riding one?

  6. I cycle on country roads and highways, so thanks for telling motorists I’m fair game, Joe. Why is it that so many cyclists can’t see beyond their own narrow self interest. The obvious solution is to educate motorist and cyclist alike and enforce existing traffic laws. That’s what they do everywhere else.

  7. As much as it makes sense to reduce speeds and redesign roads, it’s always politically a mess. What should have been a non-issue is to implement better enforcement… automated speed enforcement. There is no reason why anyone should fear speed or transit/bike lane enforcement if we are all obeying the rules of the road. No, you’re not getting a ticket for going 1 mph over, or maybe not even 5mph, but it simply makes you THINK about what you are doing and how fast you are going.

    All i really want is for people to respect one another. Give people some space, pull off the road to drop off or pick up passengers… Dont follow so close, you’re not going to get there any faster.

  8. It’s always amusing to hear from the peanut gallery (those that are uninformed and make assumptions). Speed isn’t an issue on rural roads? Really?! Perhaps you know something that all of us that work in transportation planning and engineering don’t. Please enlighten us as to our erroneous conclusions and the faulty data that shows otherwise.

  9. The point is there are a lot fewer pedestrians and cyclists on rural roads, and none on limited access highways (it’s illegal to ride or walk on highways, at least where I live). The solution is to have separate infrastructure parallel to high-speed roads, which is what they do in Europe. With the infernal distances people drive in this country, you’ll get a lot of pushback if you start trying to slow people down on roads where they need to drive fast to make reasonable time for an occasional cyclist or pedestrian. I think we should choose our battles wisely and start with urban streets where pedestrians/cyclists are often the majority users.

  10. As I said below, the best solution is separate, parallel infrastructure along high-speed rural roads.

  11. This doesn’t require “big data”, just big cojones for politicians. We know where speed is a problem (usually everywhere) and we know what needs to be done (redesign for lower speeds). Fiddling around with data doesn’t accomplish what we really need to do: change the standards. Doing that ensures that we don’t perpetuate the problem with any new development that may be occurring while also making it easier to face backlash against retrofitting existing infrastructure (“just following the standards”). The only role for big data is pinpointing crash locations for prioritizing retrofits. As for standards, I strongly suggest the Dutch “Sustainable Safety” model that bases road design on its classification, not traffic studies. Change the units and copy it.

  12. I suspect most states will use this data to raise posted speed limits to where 85% of motorists aren’t exceeding the current ones. Oddly enough they don’t include pedestrian and cyclist speeds in their calculations.

  13. Long range tactics and strategy are usually overlooked by bicycle advocates in what is essentially a long war, perhaps the longest and most deadly undeclared war in our history. Instead we get the latest “solutions” that ignore what was learned before. One important strategy right now is combining forces with pedestrians and other lower speed vehicle users and neighborhood residents to target specific roads and curb lanes for 20 mph through design and enforcement.

  14. That’s sort of what I said. We’ll get the most bang for the buck targeting urban streets and suburban residential streets for lower speeds as those are the places most likely to have people walking or cycling. It’s a war, so let’s start with the relatively easy victories first.

  15. Data on accidents is terrible. It’s 2018. All this stuff should be tracked, publicly available and heat mapped based on severity of injuries by Zillow, trulia, etc. so that prospective buyers/renters can make informed decisions on where to live similar to their crime heat mapping. It’s should be sortable both based on location of accidents and residences of those involved, mode of transportation, hour of day, etc. We should be able to see trend lines over time, run functions based on our traffic study data.

    I’m sure insurance companies are sitting on all this information, but our gvmt is clearly flying blind and/or operating without any second order thinking skills and just applying tired mid-century pseudoscience. When our rate of traffic deaths is an order of magnitude higher than that of other 1st world countries, but where their roads were laid before the traffic engineering profession existed, something is clearly broken.

  16. Peanut gallery?? Our rate of traffic fatalities is the worst in the first world, and lags significantly behind nations where their cities & road networks were laid well before the transportation planning and engineering profession existed.

    I’ll take the “peanut gallery” all day.

  17. In the meantime, why does Streetsblog continue to lead with death and carnage?

    How would you suggest building the political will to make the necessary changes without highlighting the death and carnage?

    In an ideal world, yes, it would be as simple as politicians looking at the data, comparing to other countries, and making changes. In practice, pretty much any time anyone actually tries that in the US a bunch of drivers start yelling about a war on cars, and not much happens. In a political environment where loud outrage seems to be the only effective political currency, being loudly outraged about carnage is what it takes to get safety improvements.

  18. It’s obvious to me what’s broken. There’s no incentive to fix things because lots of people profit off the status quo—hospitals, insurance companies, emergency responders, body shops, car companies, oil companies, even military contractors if you want to include the wars we fight to keep gas prices low. Most other first world countries have national health insurance paid by taxes. They have a huge incentive right there to reduce unnecessary deaths and injuries.

    I sometimes think we should hold car manufacturers liable for the carnage their machines cause, including the health effects of air pollution. If automotive carnage caused lots of people to get poorer instead of richer there might be some incentive to fix it.

  19. If you need to post a speed limit sign to get drivers to drive at a safe speed for the road, you designed the road badly.

  20. This data should be used to raise speed limits to the speeds drivers clearly want to drive, not to slow down their freedom.
    James C. Walker, National Motorists Association

Leave a Reply

Your email address will not be published. Required fields are marked *


The American Highway Safety Establishment Warms Up Some Leftovers

A group of heavy hitters in the road building and traffic safety establishment recently came out with a plan called “Toward Zero Deaths” [PDF], presented as an ambitious strategy to cut traffic fatalities in America. But don’t get too excited by the branding — the ideas inside don’t present much of a challenge to practices that have made the U.S. […]