Where do commuters hurt the most?
June 30, 2010 at 10:19 PM ET
IBM's latest "Commuter Pain"
index indicates that the traffic in Beijing and Mexico City is way worse than it is in Los Angeles or New York. Big-city traffic is never pleasant - but there's hope as well as hassle, thanks to some high-tech traffic strategies currently under construction. That's the reason why IBM started measuring the pain in the first place. This is the third year that IBM has come out with its CPI. "The first two years, we had focused intentionally on U.S. cities, but this year we decided to expand it a bit more and look at cities around the world," said Naveen Lamba, global industry lead for IBM's Intelligent Transportation practice. For this year's index, IBM worked with Survey Sampling International to poll 8,192 motorists between the ages of 18 and 65 in 20 international cities during the month of May. The drivers were asked 10 questions about metrics such as commuting time as well as their views on gas prices and road-related stress. Average scores for each city were converted to a 100-point scale. Moscow's drivers reported the longest-lasting traffic delays. When they were asked to report the length of the worst jam they've experienced during the past three years, the average time was two and a half hours. Across the board, the average longest wait time was an hour. Drivers in New Delhi and Beijing reported the highest incidence of traffic-related health problems (96 percent and 95 percent, respectively). Eighty-four percent of the Beijing commuters said traffic has had a negative effect on work or school. When all the scores were compiled and weighted on a 0-to-100 scale, Beijing and Mexico City were tied with a 99. Los Angeles, the most painful U.S. city on the list, was tied with Amsterdam way down at No. 13. The least painful city on the list of 20? That would be Stockholm, the even-tempered capital of Sweden. Here's the full list:
IBM This graph shows the scores of 20 cities on the Consumer Pain Index. Beijing and Mexico are tied with 99, the highest score. Other cities include Johannesburg (97), Moscow (84), New Delhi (81), Sao Paulo (75), Milan (52), Buenos Aires (50), Madrid (48), London (36), Paris (36), Toronto (32), Amsterdam (25), Los Angeles (25), Berlin (24), Montreal (23), New York (19), Houston (17), Melbourne (17) and Stockholm (15).
Roughly half of those surveyed said that traffic has gotten worse in the past three years, and 31 percent said the tie-ups have gotten so bad on occasion that they've had to turn around and go home. So why is IBM doing this, other than to make Angelenos feel better
about their commuting woes? Here's why: Using network science to ease traffic messes just might be a growth industry in the years ahead. The global market for intelligent transportation projects could reach $421 billion per year by 2015, according to Brian Cotton, vice president of consulting at Frost & Sullivan. IBM doesn't just make computers and the software that runs them. The company also sells fare-management systems for mass transit, flow prediction systems for traffic managers and lots of transportation-related software. Doing the Commuter Pain survey helps IBM figure out which products could make the biggest impact. "What types of solutions can we develop to mitigate pain points? ... The situation is so bad in certain instances that one or two solutions are not going to make much of a dent," Lamba said. Among the potential solutions: adjusting the traffic lights on streets and freeway on-ramps to reflect the ebb and flow of vehicles ... raising or lowering tolls based on congestion ... giving different routing guidance to different sets of commuters. "With all these different levers, we analyze what would be the impact of making those interventions," Lamba said. One of the biggest challenges is simply giving commuters and transit managers accurate information about the delays that are just ahead. "Real time is great, but real time is often too late. ... If you can get ahead of the curve, where you can predict very accurately a little bit into the future, then you can proactively manage these networks," Lamba said. IBM's traffic prediction software combines historical data and real-time traffic reports, road segment by road segment, to estimate what a commute will look like as the commuter continues down the road. The system was put to a successful test
a couple of years ago in Singapore. "It can accurately predict anywhere up to 60 minutes into the future and tell you what the speeds and volumes will be on each of the links that you have the information from," Lamba said. Those predictions can be distributed to commuters via electronic bus-stop signs or mobile devices to let them know when the bus is really
going to arrive, as opposed to when the schedule merely claims it will be there. The software can even fill in gaps in regional traffic data, although those gaps are likely to fade as GPS-based traffic monitoring takes hold. All this predictive power sounds great. But as quantum physicist Niels Bohr once said, prediction is very difficult, especially about the future (or was that Yogi Berra?
). "We can project this information, but the minute you disseminate that information, that prediction is no longer valid," Lamba acknowledged. If you tell commuters to take Highway 99 rather than Interstate 5, soon Highway 99 becomes as clogged as the interstate. That's where differential advice comes into play. "What is the desired end state you want [the road network] to look like, and given that end state, what is the information that you put out to different people?" Lamba said. Just as airport security agents direct people into different inspection lines to even out the load, commuters going from the same point A to point B may be pointed in different directions to even out the pain. If this sounds geeky, that's good: It's high time that we took game theory to the streets. To learn more about the linkage between network science and highway traffic, check out this report from Capital Ideas Online
, as well as the book on which the article is based, "Traffic: Why We Drive the Way We Do (and What It Says About Us)."More network analysis from Lamba:• Will embedded traffic sensors become obsolete?
Lamba said IBM is exploring ways to monitor traffic flow continuously by using GPS devices in cars, starting with taxis. "Now you've got a new wave of sensors. We can use cameras as sensors as well." Someday, the stoplight will know you're coming and turn green just in time. Or, on the flip side, snap a picture of your license plate if you happen to run the light. • Are there cultural differences in commuter pain?
Yes, Lamba said: "In certain European cities, the level of transit usage is much higher, and people walk more to work. That's the normal expectation, whereas in Los Angeles or some other cities, the expectation may be different." For example, a Swede might be more patient about waiting for pedestrians to pass, while U.S. drivers might prefer a route that avoids having to deal with pedestrians, even if the trip takes a little longer. • Why do traffic jams occur even when there's no blocking problem?
"There's this concept of hyperequilibrium, where there's so much demand on the network that any small disturbance has non-proportional impact. A small thing where a police car has its lights on, that causes so much delay." Lamba recalled a troublesome commute of his own, which was set off because the cars ahead of him slowed down as they entered a patch of the street where the sun was shining into drivers' eyes. "As soon as you cross the segment where you're not looking at the sun directly, everything is just fine," he said. • Can you escape congestion by moving over to an alternate route?
Lamba says that's usually a myth. In most cases, traffic flow quickly adjusts to even out the load. "People are thinking the other street is better, but in reality ... you don't get any advantage by taking the alternate route, assuming that no incidents have happened."More about traffic tech:Join the Cosmic Log corps by signing up as my Facebook friend or hooking up on Twitter. And if you really want to be friendly, ask me about "The Case for Pluto."