Why is congestion bad?
Michael Weidler
| Where I have been is reading your paper and posts. You seem to be confused - or at least of several minds - on exactly what is the problem. However, now that you've declared the issue, let's look at Freeway Congestion. I have a very basic question: Why is congestion bad? Pollution is a non-starter. Emissions of CO2 and other combustion by-products can be ended simply by mandating that all ICE vehicles install start/stop technology. If the vehicle engine isn't running while stuck in congestion, then congestion can't be a source of pollution. I guess that brings us to economics. A large chunk of the economics is all that wasted fuel due to idling. Well, we just got rid of that problem in the last paragraph. Wasted time? 365 days per year minus weekends and holidays gets you about 250 work days per year. Los Angeles supposedly has the worst traffic in the Country with 72 hours per year per commuter. 72/250=0.288 hours. Let's go high and round up to 0.3 hours lost time per day. 0.3 hours is 18 minutes per day or just 9 minutes per direction. Sounds like a lot of complaining about very little time. So I ask again: Why is congestion bad? --- On Sat, 12/18/10, Dave Petrie <DaveP...@comcast.net> wrote: From: Dave Petrie <DaveP...@comcast.net> Subject: Re: [t-i] PRT presentation in New Jersey, by Ed Anderson To: transport-...@googlegroups.com Date: Saturday, December 18, 2010, 7:15 AM The problem is Freeway Congestion. Where have you
been?
----- Original Message -----
From:
Michael
Weidler
Sent: Saturday, December 18, 2010 1:50
AM
Subject: Re: [t-i] PRT presentation in
New Jersey, by Ed Anderson
--- On Sat, 12/11/10, Dave Petrie <DaveP...@comcast.net> wrote: From: Dave Petrie <DaveP...@comcast.net> Subject: Re: [t-i] PRT presentation in New Jersey, by Ed Anderson To: transport-...@googlegroups.com Date: Saturday, December 11, 2010, 12:05 AM In clearing my thoughts on ATRA leadership, ATRA
doesn't matter anymore, PRT mentioned less and less as time goes by,
and many of the contributors have faded away.
I am sure Ian Ford knows all this, but realizes
there are still a handful of people who want to exchange ideas. In
short, a natural course is taking effect; but the Chat Group is
still making a contribution, albeit blunted by a clear lack of
direction on how to solve one of our societies most In Your Face
problems. |
WALTER BREWER
--
You received this message because you are subscribed to the Google Groups "transport-innovators" group.
To post to this group, send email to transport-...@googlegroups.com.
To unsubscribe from this group, send email to transport-innova...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/transport-innovators?hl=en.
Brad Templeton
But the lost time is far more than you suspect, in particular due to
the lost predictability. Because you can't be sure how much
congestion there will be, you must leave enough spare time to allocate
for it even if it doesn't happen. Same for unpredictable transit.
On Dec 24, 11:25 am, "WALTER BREWER" <catca...@verizon.net> wrote:
> More congestion is the argument mass transit activists love to use to justify wildly optimistic shifts to expensive mass transit facility increases.
>
> San Diego's proposed $110 billion Regional Transportation Plan for 2050, is heavily biased toward LRT, and buses; even a few streetcars. It would not reduce auto commute times in 11 Key corridors of 40 minutes.Copying your calcs for SD yields 6.2 min.=15%
>
> Negligible? BTW 50% of delay is due to incidents, rather than demand>capacity congestion.
>
> Interesting approach; I'm going to dig out some more for total delays/mile, population, costs etc.
>
> Walt Brewer
>
>
>
>
>
>
>
> ----- Original Message -----
> From: Michael Weidler
> To: transit-group
> Sent: Friday, December 24, 2010 10:26 AM
> Subject: [t-i] Why is congestion bad?
>
> Where I have been is reading your paper and posts. You seem to be confused - or at least of several minds - on exactly what is the problem. However, now that you've declared the issue, let's look at Freeway Congestion.
>
> I have a very basic question: Why is congestion bad?
>
> Pollution is a non-starter. Emissions of CO2 and other combustion by-products can be ended simply by mandating that all ICE vehicles install start/stop technology. If the vehicle engine isn't running while stuck in congestion, then congestion can't be a source of pollution.
>
> I guess that brings us to economics. A large chunk of the economics is all that wasted fuel due to idling. Well, we just got rid of that problem in the last paragraph.
>
> Wasted time? 365 days per year minus weekends and holidays gets you about 250 work days per year. Los Angeles supposedly has the worst traffic in the Country with 72 hours per year per commuter. 72/250=0.288 hours. Let's go high and round up to 0.3 hours lost time per day. 0.3 hours is 18 minutes per day or just 9 minutes per direction. Sounds like a lot of complaining about very little time.
>
> So I ask again: Why is congestion bad?
>
> From: Dave Petrie <DavePet...@comcast.net>
> Subject: Re: [t-i] PRT presentation in New Jersey, by Ed Anderson
> To: transport-...@googlegroups.com
> Date: Saturday, December 18, 2010, 7:15 AM
>
> The problem is Freeway Congestion. Where have you been?
> ----- Original Message -----
> From: Michael Weidler
> To: transport-...@googlegroups.com
> Sent: Saturday, December 18, 2010 1:50 AM
> Subject: Re: [t-i] PRT presentation in New Jersey, by Ed Anderson
>
> And precisely which problem would that be?
>
Kirston Henderson
On Dec 24, 2010, at 10:37 PM, Brad Templeton wrote:
> Even with hybrids there is still a lot of wasted fuel.
>
> But the lost time is far more than you suspect, in particular due to
> the lost predictability. Because you can't be sure how much
> congestion there will be, you must leave enough spare time to allocate
> for it even if it doesn't happen. Same for unpredictable transit.
>
I believe that elevated, weatherproof guideway based systems, be they
fully automated individual single-mode or dual-mode vehicles, or semi-
automated, small trains operating at frequent intervals offer an easy
and affordable answer to the congestion problems. And we don't have
to wait for some pie in the sky robocars that would add to the traffic
congestion.
Kirston Henderson
Jack Slade
What I find ridiculous is that such plans can be made, for 110 Billion dollars worth of trains, streetcars, etc, in just one area and sometime next week somebody will say " where are we going to get all the money needed to build guideways".
All of us who consider guideways to be necessary seem to agree on one point, namely that they should be profitable, without needing further subsidies.
Anybody who claims this for other modes are always wrong. This is just lying, as I see it.
|
|
|
|
Michael Weidler
| OK let's tack on 10 minutes each direction. It's probably starting to get painful here, but as you pointed out congestion is unpredictable. Since that 72 hours is an average,if you add 10 minutes, you are going to be early most of the time. This sounds like a good thing to me as you now usually have time for your latte. Why is transit unpredictable? And unpredictable by what margin? Is it off by a couple of minutes? Are customers constantly missing transfers? What exactly seems to be the problem? The biggest complaint I've heard - from you as well - is that transit is too damn slow. But let's not forget the original question: Why is congestion bad? |
--- On Fri, 12/24/10, Brad Templeton <bra...@gmail.com> wrote: |
|
|
|
|
Marsden Burger
Now to keep this correct, a little about transit.
First, no matter what system is created, there will be no alleviation of congestion. In fact, the very nature of our efforts, are to increase congestion, not to lessen congestion. Congestion is there because of a drive within us to congregate. People for the most part, want to be together. All of our efforts are not to decrease this drive, but rather to make the achieving of this drive easier - reduce the inconvenience to allow us to concentrate more and quicker.
Where-ever transit is successful, congestion increases. So it has been, and so it will always be.
"Joy to the world!" and lets look for a future of happily congested cities with "effective congestion management" through cutting edge transit!
Best wishes,
Marsden
Date: Sat, 25 Dec 2010 05:37:22 -0800
From: pstr...@yahoo.com
Subject: Re: [t-i] Re: Why is congestion bad?
To: transport-...@googlegroups.com
Jerry Schneider
>Why is transit unpredictable? And unpredictable by what margin? Is
>it off by a couple of minutes? Are customers constantly missing
>transfers? What exactly seems to be the problem? The biggest
>complaint I've heard - from you as well - is that transit is too damn slow.
There are a multitude of reasons and operational problems in the
literature - it's not a mystery.
>But let's not forget the original question: Why is congestion bad?
It's a good opening question. But congestion is quite complicated in
terms of it causes and effects. There are sever different kinds and
many of them are only temporal in nature and affected day of the week
and by seasons, among other things. And there are several tags other
than "bad" that better represent these different types.
It would be helpful tto advocates of alternative solutions to devise
a taxonomy of "congestion problems" as a means to more clearly state
and prioritize possible solutions. This has been and is being done
on an ad hoc basis, here and there, and band-aids are being applied
as solutions, some beneficial, some not so good.
On the whole, a more organized taxonomy would be most helpful. I
suspect that with more and more autos and reduced spending on
highways, auto congestion of various types will get much worse in the
future. A good place to start would be comprehensive list of
"causes". I suspect more than one currently exists somewhere on the web.
> transport-innova...@googlegroups.com.
> > For more options, visit this group
> at<http://groups.google.com/group/transport-innovators?hl=en>http://groups.google.com/group/transport-innovators?hl=en.
>
>--
>You received this message because you are subscribed to the Google
>Groups "transport-innovators" group.
>To post to this group, send email to transport-...@googlegroups.com.
>To unsubscribe from this group, send email to
>transport-innova...@googlegroups.com.
>For more options, visit this group at
><http://groups.google.com/group/transport-innovators?hl=en>http://groups.google.com/group/transport-innovators?hl=en.
>
>--
>You received this message because you are subscribed to the Google
>Groups "transport-innovators" group.
>To post to this group, send email to transport-...@googlegroups.com.
>To unsubscribe from this group, send email to
>transport-innova...@googlegroups.com.
Brad Templeton
like Japan, where transit has very good on-time performance, there is
unpredictability when you don't know the schedule, because when you
arrive may be 1 minute before the next train, or 29 minutes. Of
course with trains that are every 30 minutes, you are strongly
incented to know the schedule, while with 10 minute and more frequent
trains many people don't bother paying attention to it.
Among transit agencies, the innovation of the last few years that has
most excited them has been the GPS-on-bus systems that allow you to
see when the next vehicle is on the web, your phone or a display at
the stop. They report surprising (for transit) increases in rider
satisfaction from just being able to predict when the vehicles are
coming so you don't sit waiting for long. When lived near a street
car stop in SF I found this a big improvement, even though the amount
of time it gives you is modest, and it doesn't make the trains any
faster or more frequent.
This of course is one of the things that is promoted as important with
PRT, the fact that you don't wait at the station normally. As mobile
networks expand, not waiting at stations will become common for most
forms of transit, and for other things too, and it will be interesting
to see how people compare waiting at home (ie. transit with good data)
with not waiting at all (vehicle waiting for you.)
Congestion unpredictability is also going to drop, as already we can
see maps of traffic and soon those maps will get predictive (what is
traffic likely to be like when I get there) and this predictability
will actually reduce congestion where there are alternate routes and
people naturally pick them to avoid predicted jams. There's a lot
that networks can do here, including temporarily turning sidestreets
into high-capacity routes with lane redirection, traffic light re-
timing etc. This of course makes it worse for those who wanted to use
or cross those sidestreets but the overall benefit of an instant
detour to clear congestion has value.
On Dec 25, 12:42 pm, Jerry Schneider <j...@peak.org> wrote:
> At 05:37 AM 12/25/2010, you wrote:
>
> >Why is transit unpredictable? And unpredictable by what margin? Is
> >it off by a couple of minutes? Are customers constantly missing
> >transfers? What exactly seems to be the problem? The biggest
> >complaint I've heard - from you as well - is that transit is too damn slow.
>
> There are a multitude of reasons and operational problems in the
> literature - it's not a mystery.
>
> >But let's not forget the original question: Why is congestion bad?
>
> It's a good opening question. But congestion is quite complicated in
> terms of it causes and effects. There are sever different kinds and
> many of them are only temporal in nature and affected day of the week
> and by seasons, among other things. And there are several tags other
> than "bad" that better represent these different types.
>
> It would be helpful tto advocates of alternative solutions to devise
> a taxonomy of "congestion problems" as a means to more clearly state
> and prioritize possible solutions. This has been and is being done
> on an ad hoc basis, here and there, and band-aids are being applied
> as solutions, some beneficial, some not so good.
>
> On the whole, a more organized taxonomy would be most helpful. I
> suspect that with more and more autos and reduced spending on
> highways, auto congestion of various types will get much worse in the
> future. A good place to start would be comprehensive list of
> "causes". I suspect more than one currently exists somewhere on the web.
>
>
>
>
>
>
>
Jerry Schneider
>snip -------------------
>Congestion unpredictability is also going to drop, as already we can
>see maps of traffic and soon those maps will get predictive (what is
>traffic likely to be like when I get there) and this predictability
>will actually reduce congestion where there are alternate routes and
>people naturally pick them to avoid predicted jams. There's a lot
>that networks can do here, including temporarily turning sidestreets
>into high-capacity routes with lane redirection, traffic light re-
>timing etc. This of course makes it worse for those who wanted to use
>or cross those sidestreets but the overall benefit of an instant
>detour to clear congestion has value.
Does it not make it worse for the people who live on these sidestreets and who
highly value their relative lack of traffic? Wasn't the main idea behind the
proliferation of cul-de-sacs a response to the desire for very low traffic
residential locations? I would expect quite a bit of outrage from such
people who would feel that their residential environment was being destroyed
by traffic volumes that are far higher than they can tolerate. While
cul-de-sacs are
mostly in suburban areas, I would also expect a similar reaction to residents
in some grid areas where the residents also highly value low traffic
volumes on their streets.
Jack Slade
--- On Sat, 12/25/10, Jerry Schneider <j...@peak.org> wrote: |
|
|
|
|
|
|
|
|
|
|
|
|
Jerry Roane
Kirston Henderson
First I am responding to this to say Merry Christmas and the best to all of you for the Holiday Season!
Now to keep this correct, a little about transit.
First, no matter what system is created, there will be no alleviation of congestion. In fact, the very nature of our efforts, are to increase congestion, not to lessen congestion. Congestion is there because of a drive within us to congregate. People for the most part, want to be together. All of our efforts are not to decrease this drive, but rather to make the achieving of this drive easier - reduce the inconvenience to allow us to concentrate more and quicker.
Where-ever transit is successful, congestion increases. So it has been, and so it will always be.
"Joy to the world!" and lets look for a future of happily congested cities with "effective congestion management" through cutting edge transit!
Best wishes,
Marsden
oded_roth
year.
One of the most interesting papers about congestion is:
The Congestion Evil: Perceptions of Traffic Congestion in Boston in
the 1890s and 1920s. / Asha Elizabeth Weinstein / UCB , 2002
http://www.uctc.net/scripts/countdown.pl?diss074.pdf
I would say we need better balance between congested and vacant
places.
Oded Roth,
http://www.transportationet.com
Michael Weidler
| I guess I've been lucky then. I have rarely found transit to be any more unpredictable than driving - after all they are using the same streets. I've lived in and used transit in Pittsburgh,PA, Wash DC, Atlanta,GA, Seattle,WA, and San Diego. I have found it to be very inconvenient from time to time - especially in San Diego. Mostly, the biggest disadvantage to transit is it tends to be slower than driving. |
--- On Sat, 12/25/10, Jerry Schneider <j...@peak.org> wrote: |
Michael Weidler
| Here is a working link to the paper. http://www.uctc.net/research/diss074.pdf --- On Sat, 12/25/10, oded_roth <rot...@zahav.net.il> wrote: |
|
|
Michael Weidler
| This paper is a long one! So far, I have found it rather amusing. Not the content itself but the general observation that the more things change the more they stay the same. The author is currently an Associate Professor at San Jose State University. http://www.sjsu.edu/faculty/weinstein.agrawal/ I wonder if she views this list at all? --- On Sun, 12/26/10, Michael Weidler <pstr...@yahoo.com> wrote: |
|
WALTER BREWER
Jerry Roane
Brad Templeton
streets if sometimes traffic is redirected along them. (Obviously in
the case of freeway congestion I would be talking about redirecting
the traffic mainly only the arterial streets along the highway which
are often present, but slow because of the traffic lights and other
factors.)
However, the city pays for pavement and it pays a bit more for heavy
use roads than cul du sacs, but by and large the roads are put in with
tax money for public use. I think if thousands of people are waiting
in congestion on one road and can readily be redirected to another
public road, that their need trumps the rights of the people on that
road to have a quiet street for their own enjoyment. If it's a
private road, and they paid for the road construction and maintenance
then they should have the quiet enjoyment of it.
Of late, suburban design has gone to having a few arterials and mostly
semi-private networks leading to cul-du-sacs, because people want
their quiet streets, and it's not hard to see why. But of course, if
they built or bought on such a street network, it is because they
wanted to design it so that people who could make use of it to get
through are blocked (by the fact the roads don't go through) and I
think that's OK if they pay for that.
In grid cities, traffic is distributed over a wide array of streets,
though today many people seek "traffic calming" like speed bumps to
get people off of "their" grid street and onto other streets. This
can make sense during off-peak hours, to put all the through traffic
onto non-residential streets, though it's silly at rush hour to have a
jam packed arterial and scores of light traffic speed-bump (or don't-
go-through) streets -- especially as we move to a time where the cars
can't hit pedestrians.
Jerry Schneider
>It would indeed be a "burden" on the people who live on "quiet"
>streets if sometimes traffic is redirected along them. (Obviously in
>the case of freeway congestion I would be talking about redirecting
>the traffic mainly only the arterial streets along the highway which
>are often present, but slow because of the traffic lights and other
>factors.)
OK - I agree that diversions should be limited to arterials. And that
people should somehow not be allowed to deviate from them to
other paths that they find (or expect) will work better for them. They
should delegate all route selection decisions to the "system" and refrain from
thinking about what's going on.
How would you prevent this kind of ("override-type) behavior?
Or enforce limitations to arterials, which can be a pain because of
multiple access points and intersections and traffic controls?
What factors would the "system" consider in making route selection
decisions? Is any "customization" likely to be possible? And are there
enough of them in the right locations with unused capacity to be
workable? Or would a program of "constructing" new arterials be needed?
When I see some of the routes suggested by Yahoo Maps, it's not hard to
find alternatives that are at least as good or better by making some
minor customized adjustments. The more familiar you are with the locale
the easier it is to find what is, to you, a more suitable route.
Of course, that is a function of the speeds assigned
by Yahoo Maps to the various segments of the network - so far as I know
they are the posted speed limits, and are not adjusted by time of day or
influenced by real time congestion or other relevant factors.
>However, the city pays for pavement and it pays a bit more for heavy
>use roads than cul du sacs, but by and large the roads are put in with
>tax money for public use. I think if thousands of people are waiting
>in congestion on one road and can readily be redirected to another
>public road, that their need trumps the rights of the people on that
>road to have a quiet street for their own enjoyment. If it's a
>private road, and they paid for the road construction and maintenance
>then they should have the quiet enjoyment of it.
Obviously, there would be a "suitable diversion route" ready and waiting
in some cases - the questions are how many, at what times, under what
conditions
relative to the demand? Of course, it the robocars select the routes
and can't be
overridden, then driver route selection via an override becomes an
irrelevant factor.
>Of late, suburban design has gone to having a few arterials and mostly
>semi-private networks leading to cul-du-sacs, because people want
>their quiet streets, and it's not hard to see why. But of course, if
>they built or bought on such a street network, it is because they
>wanted to design it so that people who could make use of it to get
>through are blocked (by the fact the roads don't go through) and I
>think that's OK if they pay for that.
I think that if people in some areas experience significant volumes of traffic
that outrage them, they will find ways to block parts of the network
to eliminate
such traffic and probably be willing to pay whatever it takes to do it.
>In grid cities, traffic is distributed over a wide array of streets,
>though today many people seek "traffic calming" like speed bumps to
>get people off of "their" grid street and onto other streets. This
>can make sense during off-peak hours, to put all the through traffic
>onto non-residential streets, though it's silly at rush hour to have a
>jam packed arterial and scores of light traffic speed-bump (or don't-
>go-through) streets -- especially as we move to a time where the cars
>can't hit pedestrians.
Traffic calming is a big enterprise in many cities in the US and is
quite highly
developed in the EU . The impacts of heavy traffic on people are many and
complex - and not likely to be regarded as "silly" or "must be tolerated"
just to serve the serve the desires of through-traffic commuters who wish
to save a little time and stress.
>--
>You received this message because you are subscribed to the Google
>Groups "transport-innovators" group.
>To post to this group, send email to transport-...@googlegroups.com.
>To unsubscribe from this group, send email to
>transport-innova...@googlegroups.com.
>For more options, visit this group at
>http://groups.google.com/group/transport-innovators?hl=en.
- Jerry Schneider -
Innovative Transportation Technologies
http://faculty.washington.edu/jbs/itrans
Brad Templeton
There are many ways it could be done. Already most highways have real
time traffic measurement. A simple thing that could be done would
be, in a period of congestion, to take an arterial and time its lights
to move a large flow of cars along it past the traffic jam ( a
temporary detour ) and to inform drivers this route is available.
This alone would do a lot. To allocate drivers to one route or
another is more complex and requires a congestion charging system be
in place on the appropriate roads that is able to charge people for
using a road. In that case you can tell a driver to take one specific
route, and charge them in the congestion charging system if they elect
to take the other. Stay on the road you are allocated and pay less
(or nothing.)
This would not necessarily affect those who were not on the highway to
begin with. They need not be charged.
The light timing -- indeed making the arterial solid green for 10
minutes to take traffic during the congestion -- would make it hard
for people to cross the arterial of course.
In a city designed for this it would be less of an issue as one could
have occasional bridges over the arterial, or it could simply be that
you don't need to cross it. Or you could just expect delays or not
leave the lights solid green for long periods. In a fully designed
system the arterial could even be temporarily made one-way with
appropriate signals to deal with the direction with the congestion, if
there is just one.
In a designed city, the road beside the highway would be a business
road which, during period of congestion, eventually temporarily
becomes a bit of expressway. It doesn't have to be for very long. In
some cases you just have to clear a logjam caused by some congestion
causing event like an accident, then with the accident cleared to the
side of the road, and ideally covered over with an opaque portable
fence, the capacity is restored to the highway, and you make sure you
don't put too many cars on it.
One could also just have a lottery or round robin of who is supposed
to leave the crowded freeway and who gets to stay. In this case, it's
like reverse on-ramp metering -- you tell cars they have to leave or
pay a large price, and this happens to you once a month or something
but the highway keeps flowing. In this case the arterial roads are
not altered. This is an idea that needs more development.
Dennis Manning
Brad you wrote:
> It is a new way of thinking, not possible in the pre-network era.
> There are many ways it could be done. Already most highways have real
> time traffic measurement. A simple thing that could be done would
> be, in a period of congestion, to take an arterial and time its lights
> to move a large flow of cars along it past the traffic jam ( a
> temporary detour ) and to inform drivers this route is available.
> This alone would do a lot. To allocate drivers to one route or
> another is more complex and requires a congestion charging system be
> in place on the appropriate roads that is able to charge people for
> using a road. In that case you can tell a driver to take one specific
> route, and charge them in the congestion charging system if they elect
> to take the other. Stay on the road you are allocated and pay less
> (or nothing.)
I agree this kind of coordination is technically quite doable and would
alleviate some congestion. Light synchronization has been around for
decades. Yet it has only been spotingly adopted. It's also one of the best
measures on a cost benefit measure. So why hasn't it been done? I believe is
it a systemic problem with the way projects are prioritized and funds are
allocated. In a word the politics. We need more politically savvy people.
Dennis
Jerry Schneider
>>It is a new way of thinking, not possible in the pre-network era.
>>There are many ways it could be done. Already most highways have real
>>time traffic measurement. A simple thing that could be done would
>>be, in a period of congestion, to take an arterial and time its lights
>>to move a large flow of cars along it past the traffic jam ( a
>>temporary detour ) and to inform drivers this route is available.
>>This alone would do a lot. To allocate drivers to one route or
>>another is more complex and requires a congestion charging system be
>>in place on the appropriate roads that is able to charge people for
>>using a road. In that case you can tell a driver to take one specific
>>route, and charge them in the congestion charging system if they elect
>>to take the other. Stay on the road you are allocated and pay less
>>(or nothing.)
Maybe I'm confused about the driver-vehicle relationship regarding
route selection.
Is the computer telling the vehicle which route to take or is it
telling the driver which
route to take? If the vehicle is fully autonomous, I would guess that
the driver is
not in the loop and better not try to get involved.
>I agree this kind of coordination is technically quite doable and
>would alleviate some congestion. Light synchronization has been
>around for decades. Yet it has only been spotingly adopted. It's
>also one of the best measures on a cost benefit measure. So why
>hasn't it been done? I believe is it a systemic problem with the way
>projects are prioritized and funds are allocated. In a word the
>politics. We need more politically savvy people.
Very good question. It could be that the "monument building" goal is
not being served well by a project that may not get a lot of
visibility and recognition. Also, I think that such projects usually
generate both winners and losers - sometimes the losers notice and
make a lot of noise.
WALTER BREWER
Somewhere in the '70's TRW took a fee for performance contract to reduce
congestion in a medium size area S E of Downtown LA; 15 or so miles square.
Not sure how congestion was measured, probably average speed in main
corridors.
Considerable had to do with synchronizing traffic signal, and they did a
good job technically. Including real time observations at key areas to
adjust light timing, etc. Manually that is from a remote control center.
It worked well for awhile, but faded away after 3 to 4 years. TRW lost out
financially. Two reasons:
Money disappered to man the control center, and maintain the equipment.
Influential politicians who were inconvenienced on their route by light
timeing, had the timing changed.
I was driving to work then along the main artery along the south cost. Rush
hour flow was noticeable better most of the way, but there was one signal
completely out of time. Turned out a city councilman lived on that street.
Also one favorable step was to use the streetside parking lane through
commercial districts for rush hout traffic.
OK in the morning befor stores opened, but not at night when they were open.
Walt Brewer
----- Original Message -----
From: "Jerry Schneider" <j...@peak.org>
To: <transport-...@googlegroups.com>
Brad Templeton
There are very successful timed light sections in many cities, though
it works best with one way streets being used. Of course it gets more
complex the more streets you have to involve, and when there is cross
traffic it gets even more complex. There is not a simple answer (at
least not for human drivers, there are more answers for robots) but it
is able to do things.
Certainly for the proposal in question -- tell cars to leave a
congested freeway to detour on an arterial with timed signals -- it's
easy to make workable as the hassles for people crossing are
temporary. To do it right you would actually start with a red that
collects a large cohort of cars coming off the freeway, and then start
a green pulse that moves them all non-stop past the traffic jam.
Typically only one direction is needed to be coordinated even on a 2-
way street.
There are several ways to do this. You could just do it organically
-- people get advised to do it by signs or by their traffic monitoring
GPS system (which many of us carry and can be found in almost all
smart phones.) Or you could get more advanced, advising or forcing
some people into a detour when a detour makes sense. The former is
obviously easier. You must get enough cars off to clear the jam.
Sadly though, special signs cause jams of their own, and many jams
cause a jam in the other direction as people look to see the jam on
the other side. Crazy but real. It's one thing robots will not be
guilty of. Their passengers might ask them to slow but even so I
think we can program the robots to do the slow-and-accelerate without
reducing capacity too much. A reduction in the number of lanes is
another major cause of congestion which robots should do much better
than people at, both by doing faster and better zipper merges when
needed, or doing better stop-and-go merges (which should be done by
filling both lanes to the merge and then taking turns, according to
studies, and not by having everybody try to get in the left lane in
advance while some people zoom up the right, angering those on the
left.)