Re: vehicles per hour - road and transit

Jonathan <jlippard@xxxxxxxxx> wrote:

John Lansford <jlnsf...@xxxxxxxxxxxxx> wrote:

j...@xxxxxxx (John F. Carr) wrote:

The theoretical maximum capacity of a freeway lane is about
2,000 vehicles per hour. The observed maximum, rarely reached,
is about 3,000 vehicles per hour.

Actually the theoretical maximum is 2200 vph, and observed maximums
are around 3200 vph.

Do you have a cite for this? I'd really like to read something more
detailed. Mostly so I can stick it to the local transit activists,
partly for my own edification. Any of the books I have just give the
theoretical max and then blandly talk about how you end up seeing more
cars than that in some lanes.

You also see fewer cars in some lanes.

Just because you have a lane available for traffic doesn't mean that it
will always handle 2200+ vehicles per hour. There are many examples of
peak throughput in the 1600 to 1800 vehicles per hour per lane. In most
studies I've seen where actual throughput was measured using loop
detectors on LA freeways, the maximums over 5 minute sampling periods
were typically in the 1800 to 2000 VPHPL range. The averages over an hour
were less than that. Even without the disruptive effect of an accident,
this can be a result of many things that influence flow, such as frequent
lane changing around on- and off-ramps, where drivers feel less secure
because of curves with restricted visibility, lane shifts, lane
narrowing, vehicles in breakdown lanes, and so on.

In cases where the throughput has been measured at rates like 3200 VPH,
it is typically over a short measurement period of something like five
minutes, and was not sustained over a full hour. Once the flow is
interrupted by only one stop and go, then the throughput can easily drop
to less than 1/2 of the high throughput, resulting in significantly lower
average throughput. Thus 3200 VPHPL is not a guaranteed throughput, but
more of a short-term abberation.

3200 VPH also implies a spacing of 1.13 seconds between vehicles.
Considering that typical brake reaction times are in the order of 1.5 to
2 seconds, such a spacing means that collisions are guaranteed if
anything happens that will result in heavy brake applications. Accidents
are a reality, and the major reason for throughput below design levels.

Also, I'm looking for a good source on the claims of "x lanes is most
efficient" since everyone I argue with around here has a different,
uncited number. Also so I can stick it to local transit activists who
claim that four-lane (two per side) freeways are most efficient and we
shouldn't add GP capacity.

While it is generally accepted that a maximum of four lanes provides the
best throughput, it just means that if you need more capacity, you add
separated express or collector lanes.
.

Relevant Pages

  • Re: Six lane underground rail lines
    ... of using a 34-foot diameter tunnel which could normally house ... tube to provide two levels with three lanes each level for vehicles of ... The Pennsylvania Turnpike, which runs 25,000 heavy trucks a day (out ...
    (sci.physics)
  • Re: Waiting in traffic lines.
    ... Ditto for bicycle lanes on the street. ... fatalities in Portland were standing in bike lanes at red lights. ... Large vehicles at their left crushed them doing right turns on red. ... In the suburbs filtering won't give you much time advantage at all ...
    (rec.bicycles.misc)
  • Re: Volumes where "breakdown" occurs
    ... where the capacity and level of service degrade to the point ... lanes in each direction, ... vehicles per day ... Only for signalized intersections, obviously. ...
    (misc.transport.road)
  • Re: hard shoulder
    ... I'd be more worried about emergency vehicles not getting ... if I was in an accident. ... There are regular gantries above all lanes. ... satisfactory contingency planning IMO. ...
    (uk.rec.driving)
  • Re: Variable speed limits dont cut congestion.
    ... So in theory the higher speed should result in more cars per hour throughput, but drivers don't keep to regular spacing and vehicles aren't of standard length. ... Ripple effects run through congested traffic which can even mean that traffic stops for no apparent reason. ... Slower speeds have been shown to reduce the ripple effect and lead to smoother flow which increases throughput. ...
    (uk.rec.driving)
Loading