"Smart" Traffic Control?

Traffic control is a very complicated non-linear field of study. One of the major complications is deciding what will be the goal of assembling a linked system of traffic controls; maximizing major artery flow, minimizing accident occurrence, minimize accident severity, minimize likely fuel consumption, increase traffic capability with existing roadways. Very similar to the uncertainty principle of quantum mechanics, a traffic system can be designed to only try to achieve one of these goals. Compounding this difficulty in design is that the traffic flow itself is a random process condition that varies on the time of day and that changes with urban development. So with the exception of some simple conditions, doing a computer simulation analysis can quickly run into some serious, expensive analysis time.

This is all true and relevant. However, it is amazing to see some of the primitive (or absent) schemes that seem to show up (or not) for traffic lights....

This is surely not easy, if tryng to handle all possible scenarios; but virtually anyone must have noticed some cockamamie schemes, such as lights that don't turn green even though there is no cross traffic.

Downtown Portland, OR, USA, has a number of alternating one-way streets. This is conducive to coordinated staging of signals, which is quite effective. Kudos to their traffic department in that respect.

I realize the complexity, the government mandate for more efficient vehicles is an incentive for the auto manufacturers, requiring complex systems.

My point is, should there be yet another Federal Government mandate for traffic control or should the traffic flow patterns be a product of a crude system controlled by every local municipality along the main thoroughfare with a local political bias?

We are bombarded with "Green", and as I mentioned before, a lot of it seems to be "Lip Service".

You use the word "mandate" like that identifies what should be done. A government mandate just means that regardless of other people's opinion or correctness in that opinion a government rule has been established that supercedes any opinions. At each traffic controlled intersection there will always be different concerns that must be modified to suit the situation. Should every major road to a minor road left turn signal always be only 15 seconds in length? Should left turns be outright banned everywhere like New Jersey?

I say that there certainly should not be a Federal mandate for traffic flow design. Traffic flow is considerably more complicated than can be dictated by one distant agency establishing how it will be done everywhere. Besides, while Federal funding assists many municipalities highway departments they are rarely the majority funding of that budget.

Just because the traffic control changes in your area has made your commute more complicated doesn't mean that it's wrong or right. But because it is a local board that has say over your local lights you have the opportunity to influence that board directly. If the Federal traffic regulation board for the region that contained Chicago resided instead in Louisville Kentucky then I can guarantee you that your traffic complaints will be less heard than they are now. The effect may be exactly the same, but the critical mass of complainers will certainly be much smaller. (I do realize that Chicago has legendary local political corruption, but that's another matter all together.)

Your reply indicates you have misunderstood my question, in that the question was not about "complicating my commute" but about genuine efforts at conservation, which should not have to require mandates.

Well at least my stated assumption that this was about your personal commute finally got you to state what you think should be the goal of a central authority traffic control system.

The trouble with trying to improve vehicle efficiency in usage by traffic control is that it's actually a self defeating system. By making it more efficient for a vehicle to operate, more vehicles will operate.

The tried and true method to manipulate people into consuming less of any given product has always been to raise the price of that product. But when it comes to the ubiquitous fuel prices, this has always been political suicide to even attempt to deliberately raise the price of fuel. A valid part of this reluctance is that very few fuel efficient public transportation infrastructure still exists in the US. Those that do still exist are cutting back on their services instead of increasing them.

Thanks redfred,

I guess my main point was lost in the myriads of my own dichotomies.

Recently, here in Panama, Republic of Panama, the government contracted to have "smart traffic lights" installed in various parts of the city. The results have been more congestion, more traffic jams. The common comment by drivers now, when caught in a traffic jam: "Yo soy victima de semafores intelegente!" which, if I have the Spanish transcribed correctly, means, "I am a victim of the smart traffic lights!"

I have watched over the years as the government has made several efforts to modify traffic flow (everything from changing two-way roads to one way, changing flow direction by lanes on a single roadway, even changing office hours of certain ministries). The one thing is that is very obvious from all these efforts- those doing the planning haven't a clue as to how the traffic actually moves, or why it moves in the direction it does.

Panama also suffers from boom development, and the construction boom has radical impact on traffic circulation. No consideration is given to the potential impact until AFTER the problem becomes apparent.

Result: gridlock. Fortunately, Panama is still one of those places where one can comfortably walk to the grocery, the hardware store, the doctor's office, etc. And public transportation is cheap (although subject to the same issues with traffic jams and time delays as private transportation...)

Federal mandates for more advanced traffic control systems sounds reasonable. The government would have to make an assessment on the viability and cost involved.

Changing traffic control systems requires newer technology than what exists and the only real improvements (traffic control) made in any given community are usually 'driven' by fatalities. The tax payer will bear the burden of any changes no matter if

those changes actually improve the overall "traffic control" or not. Most will argue that unless significant improvements can be made, let's NOT try fixing it. What say U?

'Lip service' is an accepted part of goverment Bureaucrasy.....

I live in a widely spread out area of about 300,000 people. I can count the number of smart lights that actually work well on one hand. Either it is just too difficult for humans to program these devices and they only randomly get it right or the different people program them and most of them are incompetent. Overall, I would say there is only a low to moderate improvement in traffic flow as a result.

Check the Federal Government they have studys for everything including the safe and proper use of redlights and timers at intersections.

But the cities often do not follow the redlight timer reccommendation and speed up the timers which is the cause of many accidents around redlights and intersections.

I've often fumed at Lights. It would be so easy, on the busier junctions, to put a camera or speed device on the main route(s) or on complex junctions. This could then do three things:- 1. Change the lights against any vehicle exceeding the speed limit. 2. Allow all the vehicles in a stream pass through (reducing pollution and travel times). 3. Recognise gaps in the traffic and let minor road or cross over main road proceed. 4. Perhaps recognise Emergency Vehicles by the spectrum of their flashing lights. 5. Night time traffic flows would be much improved as lights turn green on approach (if you are below the speed limit.) 6. The existing timer limit would over-ride to allow side turning traffic to get clearance, particularly to allow non-recognised traffic through. A camera system would 'recognise' vehicles and bikes with additional facility to record registrations and alert on listed plates (and priority to listed buses and emergency vehicles). My home town has got a terrible system at the moment that seems designed to pass traffic from red light to red light. It has just identified hot spots of pollution at traffic congestion areas. How slowly the public sector minds work!! Yes there is a cost. Yes there are multiple advantages. Yes, DO IT.

Hi I have looked at this problem for years, one of the problems that you have at trafic lights is the drivers are distracted at traffic lights, and do not move the instance the light turns green, you have this delay, ie: First car off the lights takes 3 sec reaction, the second takes just as long, etc etc, so traffic does not get through, if when the lights changed all vehicles started moving at a slow pace and then gradually increasing in speed, this would allow more vehicles to move in one green light phase, but this wont work due to human reaction, we dont move until the vehicle in front has moved, this means that by the time that the 10th vehicle has entered the intersection 30 or so seconds may have passed.

We could put in smart cameras that monitior vehicle movment baised on vehicle movment trends at different times of the day, but this would be complicated and probally still not work.

I think that the best way to sort out traffic light problems is to review the way you travel at certain times of the day, I have done so. I live in Auckland NZ at about 2.30pm i have to travel to pickup my son from school, this trip usally took about 14 mins, due to traffic conditions (mums and dads picking up their kids) So I decided to change my route, now it only takes me about 7.5mins and the distance traveled is about the same, there is less than 400m between the two trips.

Also if there was an alarm in you car that went off to wake those dosing drivers we might be able to sycronise the drivers, but apart from that unless someone has a good idea, we are stuck in traffic.

The other problem is with the Red light jumpers, they also cause problems to traffic flow, penalise the red light jumpers, rid them out of the system and things will flow again more smoothly, there is no easy answer.

Cheers

Joe

The following has been in existence for a few years now and amazingly seems to work. Controlled Chaos - European Cities Do Away with Traffic Signs. http://www.spiegel.de/international/spiegel/0,1518,448747,00.html

Interesting article and concept. However, I didn't see any information on rate of traffic flow. The charm of cobblestones might be weighed against faster tire and suspension wear.

I would like to see a city organized in something like mile-square superblocks, with the streets in one direction elevated over the cross streets. By "square" I mean topologically; I would prefer curves in the roads. Each superblock could have access on four sides, and the superblocks could constitute zones for industrial, commercial, recreational, or residential usages.

I am partial to the concept of the "Central City" being closed to vehicular traffic, with feeder roads feeding into parking lots around the periphery. Service into the city areas would then be provided by electric trams or walking. Someplace like Los Angeles need not be limited to a singe "Central City". An interesting side effect of such an approach- shopkeepers talk of increased traffic and revenues, when sections of a city are designated "pedestrian only". It seems when people are forced to "slow down", they actually enjoy it...

As I recall, Curitiba, Brazil, is somewhat like that--and very livable.

Interesting idea....can you expand on it a bit more...maybe a graphic or two?

I will try, but I'm not sure how much I can promise. This computer has no scanner, and only some older AutoCAD Lite on it, and I haven't had much luck importing/exporting graphics. No doubt I am overlooking something really simple....

I understand. Thanks.

And at low traffic times it would be much more efficient to be in the flashing red and yellow mode, but around here mostly the only time you see that is when the power has been interrupted and the light came back up in the default mode.

Part of the problem with traffic control in my area is there are to many different agencies involved. You have State, County & Municipal Right of Ways.

We have over 90 Municipalities in our area.

Each has their own system and ways of doing things. None of witch are actually connected to each other. In a couple of miles of roadway you may have all 3 types of systems and they are somewhat timed but since they don't really communicate with each other they don't adjust according to current traffic flows.

Then there is the issue of cost to update the older systems to meet today's standards. Most of the municipalities around here are struggling to pay their bills so there is no extra money for updating their systems either. We've just recently started to update to LED lights around here. And it seems that the only intersections being updated are the ones with traffic cameras that send you a ticket. Guess that is the only way they can justify the cost of updating them is if they can pay for themselves in a few years by earning & collecting traffic fines.

In PA we use a lot of induction loops to run our traffic lights, along with timers.

Most intersections are set-up so that the induction loop notifies the light that a car(s) is/are present, which in turn starts a timer to change the lights. Primary roads have a hierarchy over secondary, tertiary, entrances/exists, etc. So the loops are set-up on the lower class roads.

Yo, there is a simple system that has been utilized for a long time that is self regulating and (I believe) unambiguously improves traffic flow. It is called the "traffic circle". Traffic circles also reduce accidents and pollution. They are being installed in Colorado as a pollution mitigation device (for those municipalities that violate the clean air laws). I've lived in a couple of neighborhoods where they have been introduced and people always gripe at first about how hard they are to use and then after a few months don't even notice. As far as I can tell the only significant disadvantages are cost to install and the space they occupy.

Re: traffic circles. Obviously, you have never driven in Boston during rush hour. Or Panama, where they have tried a few traffic circles- which have actually INCREASED the number of accidents, because people don't slow down and don't understand about yielding the right of way. Elsewhere, where I have used traffic circles (i.e., many cities in Europe), one has the same issue- getting into them and out of them at the right point is not for the weak of heart. If everyone were traveling at horse-and-buggy speeds, they would probably be fine...

Over here in New Zealand we call them Roundabouts.

Actually, I have driven through Boston roundabouts. It was way back in late 70's/early 80's. Back then Boston (maybe all of Massachusetts) had an odd rule that traffic entering the roundabout had right of way. This results in the obvious situation that traffic stops when the roundabout gets filled. Most of the country has right of way for the traffic in the roundabout.

Why use any form of electric/electronic? NO TRAFFIC LIGHTS!!!

Studies have shown that roundabouts cost a simialr amount to design and build, but the running costs are really tiny (grass cutting occasionally?)

No CO2 footprint

Works day and night even if the power fails.

Looks much nicer.

ERADICATES HEAD TO HEAD CRASHES COMPLETELY.

When accidents occurr, the vehicles are probably going in a similar direction and damage will be dramatically reduced.

There are more advantages but I cannot remember them at this time.....

I think they could do better traffic flow designs if they gave an bunch of 8 year old boys a box of Dinky toys.
I think Harlow must be the world capital of unecessary traffic lights, we have lights 5yards in front of another set...I don't know anyone who knows WTF they are there for.
We have bus lanes which are traffic light controlled...they sense the bus approaching and turn the main flow red and then..
.
..
...
eventually after ages the bus, which has now stopped and lost momentum gets a green light, which stays green while the bus gets 1/4 of a mile away. The bus lane light then goes red...
So we are sitting at two reds for another 10 seconds... oh heck...you get the idea.
A correctly painted set of white lines across the road would serve the purpose far more efficiently. It's funny when the traffic lighs are out of comission we don't get the traffic build up so bad!

Traffic system designers should have to drive through theirs systems every day at rush hour for 6 months before sign off of the scheme (as unpaid overtime)! I bet they would soon improve the fine tuning.
Del

It seems that you have brought up a subject of interest to many, witness the verbose responses.

Yes, I believe that in general a "smart" system would help, and so to my response:

-Mimic the traffic officer.

-Determine approaching traffic with a sensor at some distance from the intersection, not at the intersection.

-Use an adaptive system that would automatically change as the traffic pattern changes throughout the day.

Among the incentives are decreased dependence on oil, improved traffic flow, possible fewer accidents and happier motorists.

Our government is interested in improving the life style of the public, aren't they?

I quote:

"Our government is interested in improving the life style of the public, aren't they?"

You have hit on why I started this.

It is a very big question.

One of the most promising systems for traffic control I have ever encountered was years ago in Tokyo, where there were real-time monitors installed at key points along the major freeway that showed the level of congestion in the roadway ahead, and offered suggestions for alternative routes- all graphical. This allowed the driver the opportunity to make spot decisions and some control over congestion.

Also, years ago, many radio stations used to broadcast traffic conditions in the US- a lot of the systems I experienced were less than effective, having significant delays in getting real-time information out to the public, but some were quite effective.

It would seem that modern technology would be able to come up with some sort of GPS-based real-time traffic report that one could receive in one's automobile, based on a strategically distributed sensor system and offering suggestions for alternate routes. This would allow the participants in the traffic circus at least some modicum of control over their commute time...