Featured Post

Part 1: Mark Baumer Reflection: Impounding Vehicles & Immigrant Rights

This is part of a multi-part reflection I've been doing following the death of my friend, Mark Baumer . There's nothing graphic i...

Does RIDOT Know that Rhode Island is Steadily Losing Population?

See full report here. It seems hard to imagine that a state that is losing people is going to have an exponential explosion in traffic congestion, especially because most theorists of traffic congestion recognize that people will reach a particular critical mass where congestion delay pushes them to change the time of their travel, their route, or other factors to prevent being stuck in a traffic jam.

Adding more lanes, fancy flyovers, etc., does not help traffic congestion. 

The only way to address traffic congestion in a meaningful way is to accept its existence, and work to establish alternatives to driving. The ironic thing is that if the models predict that there's a huge, ballooning demand for travel in a centrally-located urban space, that's exactly the sort of space that would be ideally suited to transfer some of its travel demand to transit. If Colombia can figure it out, why can't we?

We're all in this moment where we're so disgusted by Donald Trump that the alternatives seem quite palatable (believe me, I feel that way). But I'm disgusted that our Congressional delegation and Mayor Elorza signed onto a letter that expresses such nonsense about ever-expanding demand to drive. How are we going to solve the crises facing us if this is the best we can do?

And let's imagine a scenario where suddenly climate change is a hoax by the Chinese. In that situation, what financial sense would any of this make?

Constructive Thoughts

So last night I took a glance at the press release for this and flitted through the plan for a few minutes, gathered my frustrations at the world, and tweeted this:
Which has its place! 

Today, let's see if we can come up with something constructive to say about how to make this plan less expensive and better for Rhode Island and Providence.

First off, does anyone else see how crazy it is to invest more money into this location? Rhode Island is not going to to tear the Viaduct down today (unfortunately) but it should be thinking towards that as its future goal. To the extent that this bridge needs to be made safe, fine. But the map above shows that the state already has a great alternative in place (295) which already connects to I-95 in Cranston and Attleboro, and could easily be re-designated as 95 in the future, no construction needed.

Added Construction-- How About Not?

From the plan:

To my mind, it seems like the cheaper way to avoid the weaving conflicts at this spaghetti bowl would be to eliminate the ramps. Work on the Viaduct as a bridge that connects two sections of I-95, but don't rebuild any ramps that connect off of it to the street. The plan above says that it's eliminating the existing ramp in-bound, which means this:

But then it calls for constructing a new ramp in its place off of W. Exchange & Dean. This was part of the "parkway" discussion for the 6/10 Connector, and so maybe I shouldn't be surprised. But it's a pretty lackluster idea. Why not just get rid of the Dean Street ramp and call it a day? It's not far for drivers to just get on I-95 or 6/10 directly.

8 minutes by walking. Do we really need an exit/entrance ramp set here? (It
takes even less time to drive to the other ramps.)
In an earlier post, I was asking "Are More Ramps Better?" about a different section of highway, and my point in that post was that a highway is an express route, and so none of the logic of a street or boulevard, with its fine-grained connections, makes sense. If you're going to have a highway, the point is for it to be a route for people who are going a long distance, and to discourage others who are not going a long distance from getting in the long-distance-travelers' way. If you don't believe my personal mutterings on that subject (which, you know, is fair. . . ) then you could keep in mind that this is also what an actual P.E. said at the public forum for the 6/10 Connector. Ian Lockwood noted that in a project he worked on in Connecticut, the initial goal was to remove inner exits and entrances from the city, so that the highway would become a more useful through-route for cars making long trips.

This proposal says that it's multimodal, but the multimodal parts of the project are marginal. The Dean Street Bridge is supposed to become a great connection between neighborhoods. The best way to do that would be to keep the highway traffic away from it. Allow drivers to cross it like an ordinary bridge without highway ramps, build protected bike lanes and wide sidewalks. But don't complicate the situation and make it less welcoming to the local connections by mixing highway ramps into the mix.

There's been a lot of talk about how building new or different dazzling connections on highways is going to take traffic off of local streets (for instance, recently with "the Missing Move" on the 6/10 Connector). But I don't really buy the notion that people in actual cars follow our prescribed taxonomies about where they should be, unless there's something about the way the road is designed that constrains or nudges them in a certain direction. All the Rhode Island drivers I know jump on the highway system in Providence for local trips because it's been built in such a way to work for that (the contrast for me is my childhood experience in the Philadelphia area of rarely using full-blown highways unless we were headed somewhere more than 15 miles away). That means that the 6/10 Connector, I-95, and 146 are often clogged at 3 in the afternoon. Creating a multi-layered ramp system will remove the immediate queueing that RIDOT is describing in their report, but it will also send out a welcoming signal that there's no traffic, and no reason to avoid this area, until it's back to the same conditions.

A Proof
Let's imagine you "know a guy" in Rhode Island, and so you have an entrance/exit ramp pair right next to your house (love don't mind the noise, and love driving). You want to go to the grocery store, which is one mile away. You think to yourself in a kind of Crimetown way, "I ain't no stooge. I'm taking the highway. Not gonna' get lawwwst in local traffic." If the highway is designed so that the next nearest exit is right next to your grocery store choice, this makes sense.

But if the exit ramps were even spaced out to two mile increments, it no longer makes sense. Now, no matter how uncongested the highway is, you arrive on the other side and have to backtrack the exact same amount (1 mile) in local traffic that you would have if you hadn't take the highway at all.

This is a taste of what's going on: the highway's usefulness in Rhode Island is being diluted by an insistence on too many individuated connections. The more that you increase the distance between the ramps, the more that that road acts as a proper express route, and it naturally sifts out the people who don't belong there.

My recommendations:

*Fix the Viaduct so it's safe. Don't add any exits or entrances. If there is weaving at a particular section, it might make sense to close an exit/entrance pair, rather than try to create flyovers to separate the traffic. 

*Build the Dean Street bridge for mixed traffic (cars, bikes, pedestrians) but don't give it ramp access to the highway. That has no worked out well.

*Keep open the idea that someday, we're going to want to remove I-95. I wrote a post some time ago talking about how we could even begin temporarily imagining that reality on weekends, or special events, by having full-blown CĂ­clovias on I-95 in Downtown Providence or Pawtucket. In the long-run, it makes far more sense for 295 to be the major trucking route, so let's invest in a way that acknowledges that.



  1. A few thoughts, though I don't think about this as often as you, so I may be missing something.
    1 - In one of the meetings with RIDOT, it was apparent they don't buy the proven logic of induced demand, so it really does appear they should explain what traffic model they are using.
    2 - Adding yet another lane to get rid of weaving is, in my mind, a way to help people leave the city more quickly. People who hate the city want to get out as quickly as possible at the end of the day, or after the event they attended at the Dunk or PPAC.
    3 - You misspelled "Colombia"
    4 - I'm usually a driver and I use the freeways for local trips. My brain automatically computes the quickest way to any particular place, taking into account likely traffic levels. Sometimes city streets are better, but not usually, especially where I am going only one exit up so there's no need to actually merge onto the freeway. Also, businesses locate to the exit ramps, not the other way around. If you spread out the exits, the grocery stores will not remain between exits, they'll migrate to them, when possible.
    5 - I do wonder what it was like before they built the interchange next to the mall. As I understand, it was only done when the river relocation project was done. How bad was traffic on Dean that caused this? Was traffic backing up all the way to Johnston or did people just use I-95 instead of Routes 6 and 10 to get to downtown?
    6 - good post. I agree with most of what you said.

  2. Also, please don't quote these things in your future writings or tweets, if possible. These are my thoughts, not meant to represent PPS, which is what people will assume. Thanks.

  3. Thanks for the correction on Colombia. I don't know why I didn't know that. And also!--- apparently the Ivy League university does spell the word Columbia, so maybe that's it.

  4. Yeah, for whatever reason, RIDOT doesn't seem to buy induced demand. I don't know what to do to get them to recognize the reality of it. I think you're right that the idea is to come into and leave the city quickly without having to percolate through the neighborhoods at all, and to the extent that that works, I think that's to the city's detriment. Of course, to a great extent I think because of induced demand the "quick" part won't happen, but it will still mean that people only go from one exit directly to another.

    As for businesses locating directly to exits-- I think there's a degree of truth to that. On the other hand, are grocery stores in Providence relying on people from the highways to stay in business? I have a hard time understanding that model. Don't people tend to buy groceries within a few miles of their house, no matter where they live? I can't imagine people from Cumberland are coming to Providence to buy their stuff for dinner. I sort of think the Rhode Island habit of using the highway for local trips is just a dysfunction of the way things are designed here, and not something that emanates directly from business interests (but who knows?). I tested out my theory on my mom this morning, who lives in Delaware County outside Philadelphia, and drives most places. She seemed to think the idea of using a highway to get from neighborhood to neighborhood was a really weird concept, which is what I thought she'd think. Maybe this is a really common thing in many metros? I imagine in your home metro of Atlanta, it's probably even more like this, no? It's not like that in Philadelphia (and not because of transit use, because people are used to using the normal street grid to get places when they drive).

  5. If I read the tea leaves correctly, a collector-distributor road simply means that drivers getting on at Atwells and from the 6-10 connector can no longer get through to route 95 north. They all have to go onto 146 north or onto Orms Street. Well, it does solve the northbound weave. The southbound weave? Sit in it!

    If the RIDOT actually wanted to solve the weave then they'd do what New York City did at the George Washington Bridge: have both left and right entrance and exit ramps. Not that the GWB and the Cross-Bronx still isn't strangled all the time. Turning the left lane into a designated thru lane with double yellow lines would also help because no one would stop cold from Branch Avenue down to Thurbers Ave southbound and the same northbound.

    The Thurbers Avenue on-ramp is a big highway choker.

    The good news is that urban roadway cars can be reinvented as elevators that travel on above-grade cables. Automation is easier in the air where there are no pesky bicyclists to hit. Also, intersections are easy as pie when one cable simply goes above another cable, and bridges across rivers are also pretty easy. The system can become intensely disability-friendly so that just about anyone can go anywhere. Afterwards we'll wonder who on earth paid for all those worthless freeways?

    Until then, I want traffic lights that sense when GPS-equipped buses are approaching and that let the buses fly through.

    --Paul Klinkman
    Anonymous as usual

    1. You could add a left and right exit ramp, but again, the issue is cost and space. How much money do we feel like throwing into this issue, and do we have the space to build such ramps? The thing that makes RIDOT proposals so ridiculous is that they tend to ignore or downplay these downsides, as if there's nothing but good coming from everything we build.

      The real question we have to ask is if we should have so many exit/entrance options in the first place. I'm talking with an engineer who has done work on removing exit/entrance ramps from cities (while keeping the underlying highways) and I'll be posting more on this when I have everything together from what I've learned.

      Ultimately, I-95 needs to go from Providence and Pawtucket. So we should also ask to what extent we want to double down on infrastructure that should eventually be taken over in function by I-295. I believe we've learned from the 6/10 experience that we need to intensify our push for changes in the way infrastructure is looked at, not back down from those demands.