Boston Using IBM Engineers To Solve Traffic Problems

vu1986 writes "Boston won the opportunity to pick the brains of six IBM engineers — including one from Tokyo — who flew in to check out its traffic situation and figure out a way to consolidate, analyze and use existing traffic data feeds as well as new data sources including (of course) Twitter feeds, to ease the city's notorious traffic jams."

Comments

Re:Tokyo?

[ThatsMyNick]

IBM Tokyo is not responsible for managing Tokyo's traffic.

I coulda solved that..

[Cornwallis]

Easy peasy. Give me a billion dollars or so... let me build a really, really big tunnel... that'll solve all the problems... I'll call it the "Big Dig" so everyone can have really folksy stories about it. Problem solved!

Oh, wait...

Tweaks to the cultural problem

[A beautiful mind]

All the IBM engineers will do is decrease the issue of traffic by a couple of percent, maybe raise efficiency by 10-20% here and there, but the real issue is cultural. Cars suck for a dense urban environment, you need people on bikes, carpooling and the most important thing: good public transportation.

Good public transportation means though forcing cars out from city centers by creating bus lanes, creating tram lines on previously car-only roads, building enough parking space at the edge of the city where people could switch over to public transport, etc.

Re:Tweaks to the cultural problem

[garcia]

In my limited experience in Boston, the problem wasn't the people living within the dense urban environment, it was the people coming in from the suburbs that was the problem. In fact, it was only some of those people because many drove to train stations and rode those into the urban center.

Now, contrast that with Los Angeles where people can live less than five miles from work but still drive in knowing it will take 45+ minutes--at least twice as long as it would by bike and almost as much time as it would take to walk.

Boston has a great public transportation system which is easily accessible. Los Angeles, OTOH, has a public transportation system but very few were using it compared to the sheet number of idiots driving from all over fucking creation.

BTW, when I was in Boston I walked or took the train/trolley and when I was in LA, I walked.

Re:Tweaks to the cultural problem

[Hognoxious]

That must be why there are so many drive-thru shopping malls.

Re:Tweaks to the cultural problem

[PolygamousRanchKid]

The problem with public transportation, is that the rich folks think that's how the poor folks ride to work. Rich folks ride to work in a big 'ole SUV, because if they use public transportation, they won't feel rich anymore.

So all you need to do, is to introduce 1st and 2nd class compartments in public transportation. That way, rich folks can still feel rich by traveling 1st class, and the poor folks can feel better about themselves, because they ride in the same transportation as the rich folks.

Obviously, a win-win.

Oh, and maybe free in-transit lap dances in the 1st class would make it even more attractive.

Re:Tweaks to the cultural problem

[fuzzyfuzzyfungus]

Depends on the location.

In my time on the MBTA, my informal observations have been as follows:

In dense core areas, usually covered by the subway portion of the MBTA system, you see a pretty fair proportion of suits and techies and whatnot. Certainly times and places where the urine-scented/vaguely menacing/talking to voices only they can hear can be found; but the pleasure of crawling through traffic and paying exorbitant rates for parking is not a luxury good. Subways aren't all fun, of course; but dense core areas are the places where the economics of mass transit are best and the freedom of the open road and the right to travel according to your own schedule are most frequently, and most transparently, the freedom to pay more in order to endure traffic and fight for parking spaces.

The 'Commuter Rail' segments also do what they say on the tin, and you certainly do see all the suburbanites who left their cars at the lot at one of the train stations and then rode in to participate in the dense core areas above.

Where things can get a trifle...downmarket... is in places where bus service operates across the same territory as moderate to low density suburban areas that are designed with car-owning residents in mind. This is where public transit's advantages are least evident, its vices most evident, relative to cars(not counting rural areas, of course, where mass transit is largely nonexistent, save for certain bus and rail routes between urban centers). If you are riding the bus through the suburbs, this might well be because you don't have a choice.

Re:Tweaks to the cultural problem

[phantomfive]

No, because traveling by car is better than traveling by train. A lot of people feel so (including me), regardless of richness or poorness. You get there faster, you don't have to worry about transportation to the station, you can pick up a carload of groceries on the way home, etc. In short, a car gives you more flexibility, freedom, and speed.

Of course, in a really dense city, with bad traffic, the speed issue goes away, but there aren't many US cities like that.

Re:Tweaks to the cultural problem

[bosef1]

I agree with the sentiment that people feel like public transportation is for the "poor", especially the bus. I have often felt that you could actually encourage more "rich" people to take the bus for communiting if you changed nothing about the experience but raised the price to, say $5-7 for a one-way trip instead of $1-2. That would serve to exclude the creepy homeless guys, dangerous teenagers, etc. You would still have to run $1-2 busses for the "poor", and all the busses would run on the same routes.

Re:Tweaks to the cultural problem

[illogict]

Wait, what? There's been no 1st class in Paris metro since 1991.

Achtung schweinhund!

[Hognoxious]

Have you ever been to Berlin? I was there several years ago and watched the traffic from the old East German TV tower that was nothing to do with spying at all, not even a little bit. It was amazing how smoothly the traffic ran. It was like clockwork.

According to a local colleague a) they adjust the lights to favor traffic moving away from busy areas and restrict it entering the jams and b) anyone blocking an intersection is taken out und geschossen.

Contrast that with Brussels or Paris where you can sit through three green lights because some imbecile on the cross street is stopped in the middle of the intersection.

Re:Boston Innovation District

[PopeRatzo]

Over the past several years, many IT and biotech startups as well as mature companies have moved to the rapidly developing South Boston waterfront, which is accessible via subway but not too friendly for people driving cars who have to contend with lots of traffic and parking hassles.

And that makes it our problem how? Because these companies decide to move somewhere that doesn't have sufficient services, they expect subsidies, tax abatements, and other taxpayer-funded giveaways.

Then, they'll be the first ones to lobby against tax increases or regulations because...teh free market!1!.

Like that Ricketts guy who is screaming about big government this and big government that, but wants the taxpayers to buy him a nice new stadium for the Cubs that he owns. And this is going on in practically every big city with a pro sports franchise. "Give us money for a new stadium or we'll move away."

I hope Boston decides to send the bill for these "IBM engineers" to the companies that are going to benefit from any improvements that make things easier for them, but somehow, considering the climate where states and municipalities have to provide juice payments for any companies that want to move there, they'll probably just take money from the schools or cut teacher salaries or firefighter health care to pay them.

what if you dug a gigantic tunnell...

[decora]

right under the city? it would probably solve those traffic problems for good! also, it wouldn't cost that much, and it wouldn't take that long.

Re:Why IBM?

[K. S. Kyosuke]

Exactly. Microsoft, on the other hand, has decades of expertise in this area. I'd recommend deploying Microsoft Traf-O-Data 2011, the newest version featuring seamless interoperability with MS Office and other popular software. That IBM stuff will still be using punch cards for sure anyway.

Re:Why IBM?

[LesFerg]

You think punch cards are bad? IBM is till pushing Lotus Notes as an email application.
Think I would prefer punch cards.

Free the market

[Ichijo]

Given that traffic congestion is a shortage of available road space for the number of motorists who want to use it at a particular time, the solution is obvious to anyone with an ounce of economic sense: stop setting the price below the going rate determined by supply and demand. Get rid of the government-imposed price ceilings on freeway travel, and suddenly the traffic jams will start to clear up.

Ideally, the price should rise and fall throughout the day to keep demand constant and prevent overcharging anyone.

Re:Free the market

[fuzzyfuzzyfungus]

The trick with market-based pricing of 'public' infrastructure(whether or not this is an argument against it is a matter of taste) is that it requires you to take a sharp, and not uncontroversial, stand on the purpose and meaning of 'public'...

There are really three-ish basic possible positions(though it is certainly possible to mix and match and hedge and squirm a bit at the cost of some complexity, and certain sorts of 'public' things fit more naturally into one category or another).

1. 'Public' in the sense that ownership is vested in some body that represents 'the people', but exploited under the usual conditions of profit maximization. This one crops up with mineral and other natural resources most frequently. The nominal owner is 'the people'; but the obvious expectation is that 'the people' will sell/lease/etc. the asset for the best possible price to some other entity and then rake in the cash.

2. 'Public' in the sense that ownership is vested in some body that represents 'the people', and that the property in question is, in some sense, 'for' the people as well as owned by them. National Parks are the most obvious example. They are 'owned' in approximately the same sense as above; but public opinion would likely be hostile if we simply sold them off and cut everybody a check. There is a sense, often poorly articulated; but reflected in generally low ticket prices, that 'the people' should have enjoyment of them, as well as ownership.

3. 'Public' in the sense of being a necessary response to market failure. Utilities are the most obvious example. Unlike #1, 'the people' are both the owners and the customers, so profit is generally seen as a bad thing; but unlike #2, where appeals to intangibles like 'national heritage' are common, public opinion generally just wants the system to run not-for-profit; but as efficiently as reasonably possible.

If you adopt market-based pricing for roads, you are (though it is not polite to say so), adopting the theory that, if enough people cannot afford access to this 'public' feature, it will be more efficient, and more pleasant for the remainder who can. This isn't necessarily wrong; but it implies that you are essentially rejecting the notion that 'the people' have any right, beyond that of 'customer', to the enjoyment of a 'public' facility. This is pretty uncontroversial in something like a mineral deposit(Show of hands: would you rather have the right to grab your shovel and go get your share of the bauxite, or just sell the mineral rights to FooCorp and get your share of the proceeds?); but becomes a bit thornier when the 'public' asset is something more like a utility. Is a 'public' road a thing that 'the people' have the right to use, or is it something that 'the people' sell, by means of their representatives, to the subset of them that can afford the equilibrium price of access?

Re:Free the market

[Ichijo]

If the revenue from congestion tolling a road is invested back into the road, it lowers the amount of money that must be collected from the gas tax in order to maintain the road. Therefore, congestion pricing transfers wealth from people who can afford the market rate for travel during peak periods to those who can only afford the off-peak rates.

And because the gas tax and other user fees only cover 65% of the cost of the roads, then congestion pricing also reduces the road's maintenance burden on people who cannot afford to drive at all.

My results

[slashmydots]

My lab of engineers have came up with this. Take away the 1% of drivers who have no business driving and hold up hundreds of people behind them and get in multiple accidents that cause a 10 mile backup and traffic will move a hell of a lot better than 1% better. There have been numerous studies saying 1 person can affect hundreds of people in any traffic system. So get grandma, the 20 year old semis, and borderline psychological problems people off the road and that'll do better than any AI routing.

CSMA/CD

[PPH]

Everyone goes. When a collision is detected, everyone backs up and tries again.

Re:Tokyo?

[colinrichardday]

Boston traffic worse? That would be an achievement!

Here's an idea:

[Voogru]

Get rid of the damn traffic lights.
https://www.youtube.com/watch?v=ZS_wjo378h4
The only problem is they can't put red light cameras for free money, oh no.

TRAFFIC "EXPERIMENTS" AND A CURE FOR WAVES & J

[xluap]

TRAFFIC "EXPERIMENTS" AND A CURE FOR WAVES & JAMS
1998 William Beaty Electrical Engineer

My first 'experiment': accidentally erasing waves!
Once upon a time, years ago, I was driving through a number of stop/go traffic waves on I-520 at rush hour in Seattle. I decided to try something. On a day when I immediately started hitting the usual "waves" of stopped cars, I decided to drive smoothly. Rather than repeatedly rushing ahead with everyone else, only to come to a halt, I decided to try to move at the average speed of the traffic. I let a huge gap open up ahead of me, and timed things so I was arriving at the next "stop-wave" just as the last red brakelights were turning off ahead of me. It certainly felt weird to have that huge empty space ahead of me, but I knew I was driving no slower than anyone else. Sometimes I hit it just right and never had to touch the brakes at all. Other times I was too fast or slow. There were many "waves" that evening, and this gave me many opportunities to improve my skill as I drove along.

    I kept this up for maybe half an hour while approaching the city. Finally I happened to glance at my rearview mirror. There was an interesting sight.
It was dusk, the headlights were on, and I was going down a long hill to the bridges. I had a view of miles of highway behind me. In the neighboring lane I could see maybe five of the traffic stop-waves. But in the lane behind me, for miles, TOTALLY UNIFORM DISTRIBUTION. I hadn't realized it, but by driving at the average speed of the traffic around me, my car had been "eating" the traffic waves. Everyone ahead of me was caught in the stop/go cycle, while everyone behind me was forced to go at a nice smooth 35MPH or so. My single tiny car had erased miles and miles of stop-and-go traffic. Just one single "lubricant atom" had a profound effect on the turbulent particle flow within the entire miles of "tube."

http://amasci.com/amateur/traffic/trafexp.html

http://www.google.com/search?q=traffic+site:amasci.com

Are they nuts?

[DoofusOfDeath]

The IBM guys are going to insist that every car's firmware gets a license for Lotus Notes.

Re:Tokyo?

[Taco Cowboy]

IBM has several research facilities in the US

Umm ...

Sorry I had to break this for you

If you go to any research facility inside the United States of America these days, you would likely meet with researchers who were imported from elsewhere in the world - from places like India, Israel, Korea, Russia, China

Re:Layout and mass transit

[dr2chase]

I think there are several reasons why a good subway system could be correlated with traffic jams.

1) good subways are installed to alleviate pre-existing traffic jams, so there are already traffic jams

2) good subways are correlated with old cities with weird layouts, so there are traffic jams

3) good subways allow you to sustain more economic activity than could possibly be accommodated with cars alone. Some fraction of those people on the subway are "marginal" subway users, meaning, if the traffic were not so bad, or parking so expensive, or (fill in the blank, in some way, less sucky), they would drive. This means that if you add freeway capacity or remove bottlenecks, some of those people leave the subway and consume all that shiny capacity that you just added, jamming it up all over again.

Don't need an IBM engineer

[xaoslaad]

I had to drive into Boston for a few days last week. 1:45 minutes to get into the city the two days I drove myself. Over two hours when I took the train, because first I had to get to the subway. Then I had to wait for the first train, which kept stopping, so it was a long and delayed ride. Then I got to the the Green line and had to wait for another train. Eventually I got where I was going. When sitting in my car is more comfortable and faster, there is little incentive to take the train. Make public transportation faster and more reliable and maybe I'll be more inclined to take...

Furthermore, on both days that I drove 15 minutes of my ride was getting through a short section of MA Ave, where the lights were perhaps 10's of yards apart. First light turns green. But the light ahead is red, so no one moves. Green light turns red, red light turns green. Next time the light turns green I'm able to move up just enough to get through the intersection and wait at the next red light... I don't know, maybe like get the lights back in sync now and again so traffic can actually flow smoothly?

Re:2 major ways to ease traffic jam

[mjwx]

Essentially there are only 2 ways to ease traffic jams

1. Widen existing roadways and build more new thoroughfares to accommodate the vehicles

or

2. Cut down on the number of vehicles that travel on the road

Don't need IBM engineers to figure that out

Yes, but IBM engineers can make driving so user unfriendly and convoluted that fewer people will want to drive, thus achieving solution number 2.

Re:2 major ways to ease traffic jam

[ceoyoyo]

In some places those are the only two ways to ELIMINATE traffic jams. You can ease them considerably, and possibly eliminate them in some situations, by making some fairly small changes to the way traffic flows. Timing lights, replacing lights with overpasses (or just blocking access from some streets), reversing lanes at certain times of the day, etc.

There's one place on the freeway near me that is almost always bumper to bumper. The road before and after this spot is usually fine. What's the problem? Some idiot highway planner designed an on ramp that comes up to the (elevated) highway level blind, then the merge lane is nonexistent. So anybody coming up that on ramp finds themselves suddenly in a highway lane, and anyone in that highway lane instantly tries to move over to the left, etc. The problem could be solved by either making a reasonable acceleration lane at highway level, getting the on ramp to highway level faster, or even blocking off the rightmost lane of the highway (generally the highway isn't at capacity anyway).

Re:2 major ways to ease traffic jam

[ShanghaiBill]

Essentially there are only 2 ways to ease traffic jams
1. Widen existing roadways and build more new thoroughfares to accommodate the vehicles
2. Cut down on the number of vehicles that travel on the road

3. Fix traffic light timing
4. Identify and fix critical bottlenecks
5. Convert more lanes to HOV to encourage car-pooling
6. Eliminate underused HOV lanes, so everyone can drive in them
7. Convert traffic circles to traffic lights or stop signs
8. Convert traffic lights to traffic circles
9. Build more off street parking, so people pulling in and out of on-street parking don't block traffic
10. Handout hefty fines to people that stop in intersections, causing gridlock
11. Encourage the purchase of automated cruise control systems (these reduce the accordion effect in traffic jams)
12. Ticket slow drivers in the fast lane
13. etc, etc, etc