Anatomy of a Crash


Sam Adams apparently has some concept of the limitations of his power as Transportation Commissioner. Last week he convened what he’s calling an “operator’s steering group” of all the agencies that have something to with operating transportation in the City of Portland. I scored an invitation, I’m not sure whether it was for my Streetcar connections, or to provide some leavening as a citizen activist (there was one other neighborhood activist invited as well).

It was interesting just how many different folks have their hand in transportation in the city, as illustrated by one of the items on the agenda.

We spent so much time on one agenda item that we had to push out several items to a future meeting. The featured agenda item was the “anatomy of a crash”. We got a detailed analysis of a fender bender on I-5 (northbound) during a recent morning rush hour.

The non-injury accident (75% of accidents in the city are property-damage-only) closed 2 lanes on the freeway, and took 1 hour and 8 minutes to clear. The estimate of the dollar value of the delay for drivers and freight is $150K.

So how can we do this better? Clearing the accident 10 minutes faster would save $25,000 in productivity, shaving 20 minutes off would be worth $50,000. Given that this happens many times per year, it seemed to Commissioner Adams that we might do something about this, and this was the major topic of conversation at the meeting.

The first reaction was to say that ODOT should get tow vehicles out sooner. But surprise, towing on the Portland stretches of the freeway system is governed by a contract managed by the Portland Bureau of Revenue (which regulates tow operators). Did the Bureau of Revenue include performance metrics for clearing freeway accidents in its contract? Why would they if they don’t see themselves having a transportation mission. Now they do!

Some of the other barriers identified:

  • Tow is not requested until an assessment is done on whether there are injuries – rolling a tow truck on a contingency basis my cost money in some cases, but save a lot of economic value in other cases.
  • Evaluating tow contracts for the dispatch location might provide better geographic coverage.
  • The vehicle owner can request a specific tow company (some states don’t allow this) potentially slowing the process.

Another idea was to equip ODOT’s Comet quick response vehicles with limited tow capability to at least get vehicles off the freeway.

Hats off to Commissioner Adams for getting all the involved parties to think together. I hope future meetings of this group are equally productive.


13 responses to “Anatomy of a Crash”

  1. http://www.wsdot.wa.gov/incidentresponse/

    The problem in Oregon is there’s too much planning and methology and graphing…and not enough “what do we need to get it done”, and “let’s make it happen”.

    I’ve sat in far too many business meetings where I finally had to throw down the hammer and say “enough is enough, what are the steps” instead of discussing ad nauseum the individual steps and processes, the pros and cons, etc.

    We know there’s a problem – traffic collisions cause backups.

    We know that reducing the length of the problem will fix congestion.

    We know that in order to move a car off the highway, you need a tow truck.

    What is preventing ODOT from buying tow trucks, or having tow companies on call with a guaranteed 10 minute response time, or what? We can do it with ambulances, we can do it with tow trucks.

    If necessary, ODOT needs to simply tell the City of Portland that state trumps city, if the city doesn’t like it the city can take over the state highway system within city limits (including all freeways).

  2. Kudos to Sam for starting this process…we have known for a long time that incidents account for 50% of the congestion on I-5, but while millions have been spent on building, or at least studying, new capacity, reducing incidents has been ignored…at least til now.
    Other steps should include adjustable speed limits, more enforcement and higher fines to prevent incidents from occurring in the first place.

  3. I agree that this was a nice first step by Sam Adams to identify the separate silo’s of control and responsibility that effective hurt the City more than PDOT and this identification of the real cost of not getting these lanes of travel back open, it just ripples.

    When we effectively have NO alternatives to a critical corridor like I-5 through Portland for our north/south transportation needs any disruption of any significant time can bring catastrophic costs and implication to all stakeholders.

    When a corridor’s fixed capacity is stressed with demand (the number of lanes to the number of vehicles) it creates an environment where accidents happen at greater rates and the I-5 corridor is at that point.

    What are the solutions:

    Do what ever it takes to clear out accidents out of a stressed corridor as fast as we can by what ever it takes.

    Take steps to reduce conditions that cause accidents like turbulence; that can come from lane changes, the number of on and off ramps (needs to be reduced), create longer off ramps/starage area’s away from the travel lanes, reduce the number of vehicles to the to approximate the capacity of the number of travel lanes, and create new capacity that redirects vehicles to alternate corridors thereby not hurting our economy.

    Things that we cannot do is to think that this disruption cost will go away by just getting the vehicles out of the corridor faster. Any disruption ripples long and fast and has unacceptable consequences to all stakeholders.

    Congestion is a disruption of acceptable levels of service and we now know that the I-5 corridor has greater then 7+ hours of LOS “F” conditions with NO fix in sight through 2030 with the realization that the problems and costs are going to just get worse.

  4. Entering northbound on I-5 from SE Morrison St has always been for me a challenging experience. First it merges with another onbound ramp and then, together, these have a second, rather short ramp to join the freeway. This situation is repeated similarly at the NE Broadway entrance to I-5. To not have a reasonable distance to merge into what usually is a heavy crush of traffic is asking for trouble.

    (Another hazardous merger, in my opinion, is off SW 6th Ave to I-405 northbound. One must quickly get across two other, busy lanes.)

    Hey, I know nothing’s perfect. In the case of the northbound entrances on to I-5 there is room to lengthen the ramp, thereby allowing more time to merge. Even if an accident occurred a mile from this busy stretch of merging lanes it could have been due to this problem, since other drivers might make quick, unplanned manuevers to let merging vehicles in.

    So what is my opinion?

    “On busy urban expressways design longer entrance ramps, adjacent to the main traffic lanes, wherever logistically possible”

    Or reduce the speeed limit on urban expressways…

  5. Speaking of tow truck gaurantees. The police have a gauranteed 30 min arrival time. I unfortunately know this first hand! :)

    Progressive insurance on the other hand has a horrid 2 hr gaurantee, which for some dumb reason they paid for.

  6. The freeway loop around inner Portland was designed with too many on and off ramps. At least one was removed a few years ago…you used to be able to make a right off the east end of the Steel Bridge and get on I-84.
    More ramps need to be removed and lower speeds set and enforced at the approach to incident prone areas.
    Imagine a process…like a pulp mill…where you are shut down half the time due to one mix up or another. Would you build another mill to get more product out? or go to work reducing downtime? Its about time the incident situation was addressed on I-5…they account for 50% of its downtime.

  7. Again – too much studying, not enough action.

    Here’s how a planner works:

    Let’s draw a list of all the on-ramps and off-ramps, study traffic volumes, hold several open houses, discuss mass transit options, decide that more mass transit is needed, wait around 10-15 years for mass transit federal dollars. When the decision to remove/move/update the interchange comes, completely remove the interchange (at a cost of several million dollars, including substantial sums for landscaping alone.)

    Here’s how the solution is:

    Take a map of a highway, showing traffic counts for each on and off ramp. Superimpose crash volume history on the map, showing number and location of crashes.

    Where there are on-ramps and off-ramps with high crash volumes, and on-ramps and off-ramps with low utilization, close the ramp with a jersey barrier.

    Construction cost: the cost of buying 20 feet of jersey barrier, installing it, and removing signage and traffic signals related to the now closed ramps. In California, they don’t even remove the signs, they just place green panels over the signs to “blank” them out.

    Result: poor on and off ramps are eliminated, thus improving traffic flow and reducing potential collision points. In the event of an emergency or other situation, the jersey barriers can be quickly lifted and the ramp re-opened. Sure, it doesn’t look pretty, but what freeway does?

  8. So how does closing ramps help? People who want to get on the freeeway will crowd onto the available ramps, making them far more congested. That would be a nightmare situation.

  9. The point is to close ramps that are deemed “dangerous” – i.e. very short or non-existant acceleration lanes, sharp curves on the ramps, located in areas where users have to make sudden and sharp lane changes to get to or from the ramp (i.e. 6th Ave to NB I-405, crossing over the exit lanes to 10th Ave/US 26 WB). True, the traffic won’t go away, but it’ll be routed to a more suitable (and less dangerous) freeway on/off-ramp.

    Closing the ramp is much cheaper than waiting five years for a permanent fix, and having dozens of wrecks in the meantime.

  10. Closing ramps might also reduce use of the freeway somewhat, depending on what percentage of freeway users are making “local” trips. If their preferred on-ramp no longer is available, they might simply stick to the street grid for their trip. Lower traffic should yield fewer accidents.

    I’ve wondered for some time what would happen if we shut down most of the on-ramps and off-ramps in the Metro area, leaving access points every five miles or so, thus effectively restricting the freeways to longer trips. What would happen to freeway congestion? How would traffic be dispersed across the rest of the grid?

    (No, I have no expectation ODOT or PDOT or anyone else would actually try this…)

  11. A agressive reductions in the number of on and off ramps that reduce the use of our freeways from being local arterials is very smart use of limited capital.

    The point has been made that we sould look at all history, of the locations and reasons for most of the accidents and determine if their are reasonable and affordable changes that can be made without major and costly re-works that delay most everything from happening. A KISS approach is better then nothing.

    The tuning of our flow rates to reduce congestion on our streets and particutarly our freeways to ensure that get the maximum number of vehicles and people through a stretch of road is smart with the environment and with the use of our money.

    If we have 2 or 3-lanes of through traffic that could acheive 2,000 vehicles per hour in each lane and we find that it is only acheiving 500 or 800 vehicles per hour in peak periods in two of three of the lanes we should be trying to find out what the problems are and what some of the fixes could be that are affordable and acheivable without prolonged studies after studies.

    The example is the highly congested I-5 corridor through Portland with the up-front knowledge in the hands of ODOT, PDOT and Metro that this corridor is experiencing in sections the 3rd worse in the nation air quality problems, that is killing people (most of us read in the Oregonian about the Benzene problem) and nothing is in the STIP process to address fixing and/or correcting conditions that create this problem. We cannot wait 30-years!

Leave a Reply to Erik Halstead Cancel reply

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