The city of Seoul, Korea just tore down a massive urban highway carrying 160,000 cars a day. And the result was an absolute catastrophe: the city’s economy soured, drivers were stranded in gridlock all day, and puppies and small children shed tears of remorse and longing for their lost highway.
Just kidding. Really, the results were beautiful, everything went just swimmingly, and the project has gotten rave reviews from residents (you can see photos here).
The whole article is worth a read, really, but here are the things that stuck out at me:
“As soon as we destroyed the road, the cars just disappeared and drivers changed their habits. A lot of people just gave up their cars. Others found a different way of driving. In some cases, they kept using their cars but changed their routes.”
The city had beefed up its bus service and given people options to avoid the motorway, and the effect on the environment was remarkable…
But wait, it gets better…
Last week, the verdict of ordinary Seoulians, asked at random what they thought of the development, was overwhelmingly positive. “The city centre is so much cleaner,” said Rhoda Chung, a young pharmaceutical worker. “The shopkeepers were arguing against the restoration. but now that they can see the difference they all like it.” Soo Chul Kwak, a retired driver, said: “Before, you only heard the traffic, but now you can hear the water.”
And they had this to say about traffic:
“The idea was sown in 1999…We had experienced a strange thing. We had three tunnels in the city and one needed to be shut down. Bizarrely, we found that that car volumes dropped. I thought this was odd. We discovered it was a case of ‘Braess paradox‘, which says that by taking away space in an urban area you can actually increase the flow of traffic, and, by implication, by adding extra capacity to a road network you can reduce overall performance.”
…The research team spent six months investigating what would happen to the traffic and developing a forecasting model which said it would slightly improve the traffic overall. It was put to the electorate that the motorway should be removed, and mayor Lee was elected partly on the environment ticket. “There were worries about the traffic,”says Hwang, “but we explained what would happen, and that there would be alternatives, and they began to understand.”
Now, just because it worked in Seoul, and San Francisco, and Portland, OR, and Milwaukie, WI etc., doesn’t mean it’ll work in other cities. But it certainly makes me think that Northwesterners who think that urban highways are indispensible should think again.
David Sucher
Do you truly think we are as smart as the Koreans?Honestly. We had an opportunity—the $$$ were committed—to build the first leg of a transit system. (Yes I mean the monorail) Our leadership turned it down, for reasons which still escape me.Yet all this dreamy talk of the “surface/transit option” requires vastly expanded transit.Do you truly believe we are capable of building transit to replace the Viaduct? I fail to see on what such a belief would be based.
David Sucher
Btw, can someone explain ‘Braess paradox’?Wikipedia was useless.(Using loaded and emotional words like “selfish” in the context of network behavior doesn’t do much for me.)
Norm Marshall
Braess paradox – adding a new road link or increasing the capacity of a road link can increase total travel time on the network, even if no additional trips are made.I am a transportation modeler, and I have seen this in a model. There also are real world cases where it appears to have occurred, but it is difficult to prove this because of other changes that are occurring simultaneouly. For example, in the Seoul case, auto trips declined. The Braess paradox strictly applies when the number of auto trips remains the same.Here’s an attempt to explain the paradox. In a road network, each driver seeks the fastest route for themselves. In doing so, they may avoid congested areas that result from other drivers seeking their fastest routes. The result is called user equilibrium.While optimal for each driver, it is far from optimal for the group of drivers as a whole. The entire system would be more efficient if some drivers could be routed around bottlenecks because their additional time in travel would be far offset by savings for the other drivers.In the Braess paradox, the capacity on one link in the network is increased with the unintended consequence that the combination of changed routes that results causes a worse set of bottlenecks than before. This will not result from every capacity expansion, but is something that should be considered in a congested urban network. The problem results from the interaction of thousands of decisions, so there is no way to work this out without a complex computer model. Even then, the model may not be accurate enough to model this issue reliability.
David Sucher
Thanks, Norm. I guess I still don’t get it even as a general proposition. I especially don’t get what it has to do with the Viaduct and how decreasing the capacity of a system (i.e. removing a link) will/may result in lower overall traffic.In addition you write “.. they may avoid congested areas that result from other drivers seeking their fastest routes. The result is called user equilibrium. While optimal for each driver, it is far from optimal for the group of drivers as a whole.”I simply don’t get that. It is indeeed optimal because drivers sort themselves out as they are able. The sorting isn’t perfect but that’s because people simply don’t have a choice (e.g. Evergreen Point Bridge) or they are operating with imperfect information and don’t know about accidents etc etc on specific roads. BTW, I am not saying that the Braess paradox doesn’t exist. I am not particularly mathematical. But I do have some commonsense and so far it looks to me that the Braess paradox is a convenence for people urging a certain public policy.
Arie v.
Like David when I hear of a proposal to not rebuild the viaduct the visceral reaction get is that there is a minority with an agenda and that there will be significant losers. I used to commute from Fremont to SeaTac on the viaduct, so I can emphathize with those who would have commute options taken away.However, hearing Norm explain some of the theories behind it and reading about the expressway in Seoul does put it in better context. I’m still concerned about winners, losers, and risks – but getting past the initial shock of hearing such a proposal is a first step to rational discussion.I think most folks aren’t even there yet. And passion for such a proposal can come across as indifference and smugness or as David points out putting an agenda first. For better or for worse this doomed it from the start and we never got to have a serous public debate.
Clark Williams-Derry
By the way, I doubt that the Braess paradox applies in the case of the Viaduct. There may be Braess-like effects in other parts of Cascadia, though. I think about the Gateway project in the lower mainland of BC, where they’re talking about adding road capacity in ways that seem to me like they could concentrate more traffic onto various chokepoints and low-capacity segments (in the Wikipedia example, the short green lines).The Braess paradox is just one a number of weird network-related effects. I recall that, a couple of years back, transporation planners were looking at doubling the number of lanes on the SR-520 bridge in Seattle. But modellers concluded that it would be a disaster: it would bring way more cars into the downtown corridor, which is already a terrible choke point. Adding capacity in one segment would have worsened the overall performance of the system, including for people who never travelled on 520—but not, if I understand it correctly, because of Braess effects.David—this article helped me understand it a bit better. http://msdn.microsoft.com/msdnmag/issues/05/12/TestRun/default.aspx#S2Or try this. Imagine that a bunch of drivers are trying to get from point A and point B. There are 2 basic routes, and each route has 2 segments—a long, fast one, and a short slow one. One route goes has short, slow segment first, then a long fast one; the other route has the long fast segment first, then a short, slow segment last. (Think “drive on back streets and then hop on an that goes straight to my destination”, or “hop right on a nearby arterial and then drive on back streets to my destination.”) Call the 2 routes slow-fast, and fast-slow. They’re in rough equilibrium most of the time.Then, “improve” the network by connecting the middle points with one superfast segment (say, a superhighway). Given just the right network conditions, drivers will start to take the route slow-superfast-slow, because it’s faster than any other route (fast-slow, slow-fast, or even fast-superfast-fast, since the fast sections are also pretty long). But as more and more people take that route, they *overload* the slow sections. (In our example, the city streets get way more traffic as they can handle.) So the slow sections become superduperslow.Again, under certain circumstances, superduperslow-superfast-superduperslow still winds up being preferable to superduperslow-fast. So everyone keeps taking the fastest route available to them, given the decisions of all the other drivers. But that “fastest” route actually takes longer to traverse than than the old slow-fast or fast-slow routes.
David Sucher
Thanks, Clark. But I still don’t get it. :)At any rate, the way the Legislature set it up this issue will be determined by what benefits the Governor. And I don’t t6hink she is going to go on TV and tell people that we can tear down the Viaduct because of the Braess effect.As I have said many times, she has been given a terrible hand—two very bad choices, neither one of which is politically feasible. It’s the weirdest think I have ever seen—she has allowed the Legislature to paint her into a corner.Does she have the commonsense to chooose the obvious choice i.e the Retrofit? She certainly hasn’t created the political space to do so—her ‘Expert Panel’ was a joke and the wisespread expectaion is that she will chose between one of two loser. But she could of course come out and speak the truth that there is no realistic option but Repair. That would be a welcome surprise but she hasn’t laid the ground-work to do that.I am puzzled.
mijahi
Agreed, the Governor sure does have a difficult choice in front of her – a choice between a bad and worse option. Of course if there was political support (from people not politicians) she may be convinced to consider the surface+transit option. It may be late in the game, but I think if the governor were to here from a wider group of people supporting the surface+transit option she could be nudged away from the WSDOT-peddled alternatives. What to do?Convince those people who are on the fence, those people who haven’t made up there minds, those people who haven’t even thought about it, those people that are able to take themselves out of the equation long enough to consider something beyond the status quo, and lastly, those people who don’t want to see billions spent on a Seattle-centric transportation megaproject, i.e. all those people east of the Cascades (that’s an easy one). And urge them to contact the Governor in support of the S+T alternative. The believers of this surface+transit alternative need to get even more creative and begin to actively convince the public that the benefits far outweigh the negatives that have been forcasted by WSDOT and Co. The PWC has done a fine job, but it needs to be more than just a handful of people, and it is time for new tactics. Any ideas?
M.R.
I just wanted to add some clarification to the article and conclusions that are being made about the roadway being discussed. First of all, this highway was torn down at least three years ago as part of a downtown beautification project whose focus was to reclaim a stream that had been buried under the concrete decades ago. As such, the great popularity that this project has enjoyed is more to do with the revitalization of the area around City Hall (which is two blocks from this former elevated highway) and the esthetic/recreational value of this reclaimed stream and pathways that have been added alongside it. As noted, shopkeepers are now pleased as many citizens come to walk along the stream, stop for coffee in their restaurants, buy items in their shops, and so on. Secondly, the noted improvement in traffic in Seoul has nothing to do with removing this highway. Instead, bus lanes have been added throughout the city in the past three years resulting in less gridlock for the buses and increased ridership/slightly decreased car use. It should be also noted that even with an “improvement, gridlock at any time of day on any of the major routes through Seoul is the still the norm.In short, as there is so little green space in Seoul and the air pollution is so bad, any additional green space is receiving widespread support from citizens who have a growing awareness of the environment and appreciation for public/green space. I certainly thought this project made a great improvement in the central core and am aware that many other beautification projects are in the works based on this success.I hope this helps balance the perspective a little.