Portland's slowdowns help advance science of transportation
PORTLAND, Ore. - It sounds like a Big Brother conspiracy theory, straight out of the X-Files.
Beginning in 1999, scientists at the infamous Los Alamos National Laboratory - birthplace of the atom bomb -- created clones of every man, woman and child in Portland on their super computers, recording their every move.
They weren't trying to control your mind - just change it, when it comes to getting around. These days, the scientists are still watching us - on the bus, on our bikes, at the stop light or speeding down the interstate. The information is meant to make it easier for governments to build the transportation system of the future. But what about navigating tomorrow's traffic jam? What really happens when you put data in the driver's seat?
As thousands of frustrated commuters can tell you driving around Portland during rush hour will leave you seeing red: brake lights everywhere you look. But what you may not know is that Portland's slowdowns have done more to advance the science of transportation than almost any other city in the country.
The Science of Traffic
Sitting behind the wheel the problem seems simple: too many cars, not enough road. But back in the 1990s, when government scientists put traffic under the microscope, so to speak, the laws of physics and human nature emerged as the real rules of the road, and Portland's driving habits helped prove it.
The idea that science could explain stop-and-go traffic made national headlines. But just showing that population patterns could be accurately simulated in the first place was a breakthrough, says Dr. Chris Barrett, who was in charge of the TRANSIMS (TRansportation ANalysis SIMulation System) project at Los Alamos.
"This was a big deal, and I think it was a nation-leading deal. I think Portland was really a test bed for the whole country," he said.
Barrett says they picked Portland for two reasons: the city was already known for having one of the best metropolitan transportation offices in the country. And besides being "a very well structured place, a very interesting place, [Portland] is big enough, but not so big like Chicago or New York, it was perfect," says Barrett.
Think of it like the Sim City computer game - but the Portland model was far more advanced - accurate right down to individual streets, buildings, homes and families. People's jobs, whether they had kids - thanks to census data and follow-up surveys - everything could be modeled.
Such a challenge had never before been attempted, and at the time, Los Alamos was one of the only places on the planet with enough computer-power to try it.
"Could we actually represent surrogates for every single person? And their motivations?" remembers Barrett. "You give them needs to go places and then you get them to try to execute their days and you measure what happens to the traffic." And traffic alternatives.
For example: the cost of downtown parking - simulations are run to find which price is just high enough to get more people into mass transit - but not hurt local business. The virtual rubber meets the real road all around us, every day. From the original TRANSIMS project, have sprung more advanced simulations that show whether using tolls around Portland is a good idea, to controlling how you get onto the highways.
The World of Traffic in a Grain of Rice
Stop to consider our metered on-ramps. This daily game of red light/green light may slow you down a bit, but in the end it's actually getting you to your destination faster. It's science.
"Metered on ramps are like pouring rice into a funnel," says Portland State University's Dr. Robert Bertini. "Metered at a reasonable rate, traffic flows through the bottle neck. But if you pour the rice really fast, it'll get clogged up."
Even this simple experiment (you can try it at home) shows how much of a difference controlled flow makes: when you dump all the rice in at once, it takes about 40 seconds to drain. Do it at a gradual pace, the time drops to around 25 seconds.
When it comes to teaching Driver's Ed, Robert Bertini is uniquely qualified. He's the brain behind PSU's Intelligent Transportation Systems Lab, and was tapped to help lead the US Dept. of Transportation's Research and Innovative Technology Administration. Currently, he's helping to oversee the school's new Portal project, a massive database to help shape the entire region's transportation agenda; who better to ask for data-driven solutions for everyday commutes?
Pro Tip: Follow a Trucker
"How as a driver can you make your commute more pleasant? You have different types of vehicles and you have different types of drivers," explains Bertini. "In some cases you have professional drivers - for example, commercial vehicle operators, other fleet drivers. Those are usually the ones I try to get behind if I'm stuck in traffic."
Like most traffic theory, it sounds counterintuitive (move slower for better results through snarls) until you look closer. Not only are fleet drivers kept up to speed by dispatchers about traffic tie-ups far ahead, the big rigs' slow but steady progress creates a bubble of calm in chaotic traffic conditions.
"If you are accelerating and decelerating gently, it's going to be a safer situation for you and it's going to be more pleasant," Bertini says.
The Lane Game
It's a question we've all asked, made famous in the opening scene from Office Space: if you see an opening next to you, should you change lanes? The 'grass-is-greener' choice was part of the earliest TRANSIMS modeling challenges. But the science is pretty well established: it's more dangerous, and it only looks like the other drivers are going faster.
"I avoid discretionary lane changing, I just stay in my lane," says Bertini. "That will actually reduce the level of congestion for everyone."
Humans are notoriously bad at maintaining consistent speed and spacing. You can see in this video that one driver slowing down just a bit can create a bottleneck out of nowhere. Scientists call it the shockwave effect.
Go Slow, Go Far
Don't fight the shockwave, absorb it. Next time you're in traffic, especially where there's a merge ahead, try this idea we experimented with from the Smart Motorist website:
- Instead of crowding, stay back a few car lengths from the driver ahead of you.
- You'll be able to spot a slowdown well in advance - but don't wait to brake.
- Use that space to reduce speed gradually - if you time it right, and if you've left enough room - your lane will start moving again before you reach the back of the line.
When we tried it, it worked under most conditions. It's the same idea behind the advent of variable speed limits - the speed limit changes depending on what's happening ahead of you. You'll pass through problem areas at a slower but more consistent speed reducing stop and go traffic.
Portland is now transitioning to variable speed limits with help from Bertini and the ITS Lab.
"We're working with ODOT to evaluate the system. That's really something. There's a system in Seattle, but this is really the second big implementation in the U.S. It's really exciting," Bertini says.
The variable speed limit has been used to great effect in Europe. But another European innovation - called the Zipper merge - has mostly flopped here in America. The idea is simple: a way to get drivers to use both lanes before approaching a construction zone or accident. It cuts congestion by as much as 40 percent. But drivers here won't do it, because we're too polite. It feels like cutting in line.