Traffic Lights
Were you ever under the impression that Massachusetts traffic lights are coordinated to force you to stop? I can’t speak for the entire state, but several years back I made a 6 week experiment, and made sure while driving (mostly around Newton, Waltham, and Watertown) to follow the posted speed limits, and to note the state of traffic lights when I got to them. I found that 62% of the time they were red. That is significantly worse than random!
Time in traffic burns gas, transferring money from MA to oil producing states and countries, many of which are our sworn enemies. Time in traffic isn’t spent with our families, neighbors, and friends, degrading our quality of life. The burnt gasoline ends up in our air, contributing to pollution and global warming.
It turns out that traffic lights in MA are managed by the cities and towns, a situation that right off the bat lends itself to mis-coordination at town boundaries. Worse than that, what city or town has the budget to pay for appropriate skills for (moderately complicated) light coordination? Worse yet, cities might choose to slow down through traffic, in comparison with neighboring towns, to avoid increased congestion. Worst of all, if traffic light management is contracted out, there is no accounting whatsoever for the motives and performance of the companies that so badly mis-coordinate these lights.
Suggestion: transfer traffic light coordination responsibilities to the state or to the MDC.
Public Transportation
There are two significant problems affecting the usability of public transportation in the Boston metropolitan area: lateral connections, and counter-rush connections.
Have you ever tried to go from Alewife to Newton Corner by public transportation? Although both are major hubs of public transportation, they are very poorly connected with each other. The (T) has many hubs outside Boston that are poorly connected to each other, practically forcing commuters who need to go from one to the other to either take their car, or suffer public transit via Boston.
If commuters opt, as I have when I worked in Woburn, to go by public transport through Boston, they run into the second obstacle, namely, that rush-hour buses more often go empty out of Boston in the morning, and empty into Boston in the evening. Not even taking passengers from terminus to terminus!
This means, for example, that it is hard to get by public transportation from Newton Corner to West Newton in the morning rush hours, or return in the evenings.
Simple low-cost changes, such as a few express lateral connections between transit hubs and two-way passenger traffic on express buses during rush-hours (at the very least terminus to terminus), can make public transit significantly more attractive to many commuters.
Such changes will go a long way to reduce congestion, pollution, and monetary support of Saudi Arabia.
I’m serious.
I was beginning to think I was a slow driver.
On evry type of road these freakin people are on my ass.
<
p>Yes, that means you Soccer moms in your SUVs.
And you, young independent minded groovy people with your gotees and tatoos and piercings.
<
p>I swear I may get a gun just to shoot tailgaters.
<
p>Whose with me?
There is actually a proposal to address this included in the 2003 service plan (which was basically a collection of possible expansion projects.) It would extend the 71 bus from Harvard to Newton Corner (this would involve expanding trolleybus wires.)
<
p>http://www.bostonmpo.org/bosto…
(scroll to page 5C-96)
<
p>It is rated low-priority.
<
p>You should let the MBTA know you support this extension. However, note that this is not exactly “low-cost” (1.5 million dollars, $1,400 operating cost per day). When you read through the PMT, you start to realize everything is quite expensive, and the state absolutely needs to be doing more (taking that debt over would be a start.)
<
p>You can also request a paper copy from the MBTA, I believe.
Great post.
<
p>The state did a congestion report in 2004 and it said that improving traffic signalization would result in reduced congestion at most of the buziest intersections around Boston and the surrounding suburbs.
<
p>Ahmen on improving the connection between T routes, but their is no money.
<
p>The only pot of existing cash is RMV fees that were reinstated to pay for the Big Dig cost increases. Those funds should be available in the next couple of years why not pledge them to the T and borrow today to upgrade the system and fix these things?
<
p>
… your experiment doesn’t make sense. You tested under conditions of following posted speed limits. By any reasonable standard you tested for non-real-world conditions. I don’t know what the answer is for how to correctly model traffic for such systems, but your anecdote isn’t actually helpful.
<
p>Sorry to point out a blemish on an otherwise fine post.
Using the system the way it is supposed to be used, legally, was not optimal.
… is misleading. Roads are designed with ‘design speeds’ in mind and they don’t necessarily correlate with posted limits. The designers are very cognicent of the fact that you’re example in your experiment is the very rare exception not the rule. This only makes sense, of course, because they are (and we want them to be) designing for the real world.
I respectfully disagree with you.
<
p>Traffic lights and posted speed limits should be coordinated. Lack of coordination encourages reckless driving.
<
p>Real world conditions are created by our actions and choices. These include bad traffic congestions, pollution, speeding, and a cumulative impatience that gives our drivers and state a bad name. Lets strive to correct what we can.
I’m no Ph D in civil engineering or traffic engineering — but I’ve just about finished mine on degree constrained spanning trees, which have a lot to do with roadway intersections.
<
p>You can coordinate traffic lights in one direction for many roads (north — not north/south) on a perfect grid. Think Manhattan. You can coordinate traffic lights in one direction for a single road in a non-grid system, like just about every road system in New England. If there are a few major roadways, and they have unbalanced load, you can try to time a few of them together and get mostly green.
<
p>But…
<
p>it turns out that it’s very sensitive to the number of cars in a patch of road. If that number is exceeded, even perfectly timed lights will create stops at reds quite often. Why? Because the lights aren’t the same distance apart, any variation on speed from the precisely designed target, any variation on the number of cars due to folks pulling on to the road, any instant that someone touches their brakes — and induces each of the 30 cars behind him to touch his brakes one at a time — you throw off the timing.
<
p>Additionally…
<
p>Making arterial roads faster often creates tremendous problems elsewhere. Maybe it’s worth the tradeoff… but think about highway exits. Ever seen the exit queue back up to the roadway? That’s not a good situation either, and it happens because the mean arrival rate exceeds the mean service rate of the intersection. In fact, even if the mean service rate of the intersection is larger, you can still get nasty queues because the variances don’t play well together.
<
p>Furthermore…
<
p>It’s not at all obvious that reducing the travel time of drivers (by reducing their stop time at lights) will reduce their trip time at all, or save gasoline. Why? Well, because as the road becomes more attractive, it will gain more users. People will take that road instead of their current road alternative, and people who weigh the cost/benefit of driving by themselves vs. carpooling/rail/bicycle/walking/tele-commuting/time-shifting their drive will reassess, and some will now take the newly attractive road.
<
p>In conclusion…
<
p>Some traffic engineering does result in net fuel and time savings, but generally the system is about as efficient as it can get for autos without massive unintended consequences (and lots of expenditures), all of which are generally not good for the neighborhood which they are in, due to more traffic, more pollution, and faster moving vehicles.
<
p>If you want to reduce traffic, improve the infrastructure that allows people to choose to not drive. Better commuter rail and T, better Silver Line, better buses, more widespread high bandwidth availability to encourage telecommuting, converting more highway miles to HOV lanes, improved bicycle facilities, revisions in zoning to encourage more mixed use and allow more people to live within a mile of work. This won’t allow every single person to stop driving to work, and it doesn’t have to. Traffic is nonlinear, so a 5% reduction in autos can often have a far greater than 5% reduction in time of travel, fuel consumption, and/or stress levels.
Reading your note I was struck by three things:
<
p>1. Your explanation of single-direction feasibility. During rush hours very many roads can indeed be considered as having one significantly more relevant direction.
<
p>2. You completely ignored my point that in a large metropolitan area, it doesn’t make sense to leave traffic coordination (a complex problem that could probably use PhD level education) to hundreds of little cities and towns.
<
p>3. Complexities aside, the worst case scenario should approximate random (45% to 55%). My experimentation results were significantly worse, and that fact, in and of itself, warrants some investigation.
One of the things that results in frequent traffic jams, even at off-hours (non-rush-hour periods), but at rush-hour especially is that in many places, including parts of Somerville, the traffic lights are very poorly synchronized, causing much gridlock at various intersections. I know this by experience; I’ve ended up sitting through three, four or more lights because of this, at rush-hour. Therefure, part of the solution would be to synchronize the traffic lights properly so that traffic will move more smoothly and not cause such gridlocks and jam-ups.
<
p>Regarding public transportation problems, one big problem is the fact that the trains do not run frequently enough. This is especially true of the Green Line. If the trains ran more frequently, there wouldn’t be the need to pack people into the MBTA cars like sardines. I remember being up in Montreal, where the subway trains (or the Metro, as it’s called up there) ran every two minutes, as opposed to every 10, 15, or 20 minutes. They weren’t so crowded, either. I believe that if our MBTA trains ran more frequently, many problems could/would be solved.