so in a sense, widening an urban road actually restricts supply for everything other than cars, for which you can never create enough supply to begin with.
As a rider of LA's metro, it's nice to see a train station that's actually wide. We have some station platforms that are so narrow two people can't pass each other in certain parts, such as by the stairs.
Thanks so much for the gritty details of traffic engineering. The United States population really needs this information if things are going to change in a positive direction.
Is induced demand actually a bad thing in itself? My point is: most movement (mobility in general) should provide some form of social and economic benefit that should be weighed against the negative external effects of the movement. We do not want to negative external effects (from car traffic) but we do want to positive impact of mobility. I think that is an very import distinction, because also improvements in public transit (higher throughput, higher speeds) induce demand. If we always argue against induced demand, we basically also argue against investment in any other form of mobility infrastructure. (Same goes for other areas, if you invest in healthcare, you ultimately only create more demand for health care)
Induced demand is a good thing in the abstract. It means people are doing things that they want to do. The problem is when there are significant negative externalities involved. People moving to the exurbs and commuting may be logical for them but it stinks for everyone else.
That was a really interesting question, and answer, but it was quite complicated to wrap my head around it. Had to watch the video 3 times and pause several times to make sure I heard everything correctly and make sense of it. Now I get it (think). Also your title is not accurate, we're not talking about "road widenings" here, we're talking about URBAN road widenings. Very different scenario to highways or arterial roads. I can certainly see how by widening a main shopping street that can't physically get any wider in terms of the buildings, but where you take away sidewalk space is going to cause problems for people accessing things along each side, and the reverse could make things better for them, and how if the intersections are already at capacity, you're not really increasing capacity, and thus not really inducing any more demand because people can't get anywhere faster. I do think it depends on road width and how busy the road or street is though. The effect could differ a lot if you tried it with a main shopping street versus a large residential road lined with just houses or apartments, where there isn't a ton of access required. Right?
It's true of urban roads where there are many possible routes, but widening a rural road (especially one with few alternative routes) doesn't seem to induce demand to the same extent. It also has the benefit of improved safety especially if a single carriageway is converted into a dual carriageway. For an urban area, while more expensive to build, a bypass is usually going to deliver a much better return than widening an existing thoroughfare.
Thank you for this video! So, I have a couple of questions: Does this bottleneck effect of intersections also apply to bicyles? And if so, what is the average of through-put? Is it fair to say that people will use the most convenient (=safest/quickest) form of transportation and so an widening of the road only increases the number of vehicles/bicyles/pedestrians if the road work made it a more convenient path to a destination? In some cities there are rings within the city that are just for cars and that have comparitively few intersections and no destinations. For these kinds of roads, would a widening of the road increase the number of vehicles/the capacity? And how much does the number of intersections affect the capacity? Thank you so much again :)
@@donnasmith6738 My thoughts: Bottlenecks absolutely apply to bikes. Forced stops at lights and stop signs are a major drag on biking, though less for e-bikes. I don't know the metrics. Mode choices are very personal. Cost, convenience, speed, vibe, and safety are all factors, but hard to model because everyone is different.
Bicycles have roughly the same capacity in vehicles per lane-hour as cars. Roughly one vehicle every two seconds. Cars can carry more than one person but the average in reality is 1.5ish. Bikes can fit about three lanes in place of one car lane. Also, speeding up to their optimum speed of roughly 15km/h is much quicker than cars accelerating to 50 (bikes are shorter, so you need cars moving quicker to fit a car every two seconds through), so you get better throughput in intersections, which is the main urban constraint.
Question about local access streets and through roads: How do you go about building a proper network of these in a city with a lot of quite narrow streets and complicated road/street geometry? Example: In Tartu (Estonia) you have huge intercity buses driving through streets too narrow to include a bike lane with current street geometry. You could make streets one way and have the other direction use another street but then that direction might take a long detour from an important stop. You could route major bus lines through the bigger roads that do exist, but then again you might miss an important stop. Maybe just slowing buses down with traffic calming would work but I imagine it'd be pretty uncomfortable to cycle. Any thoughts on that?
It seems to me that your answer is only relevant in a narrow set of cases, when other modes are being shifted away due to the widening. No one is walking, biking, or taking a non-existent bus along an urban highway no matter how many lanes the highway has, for example. Induced demand is usually about making a desirable destination viable in a way it was not before, getting people to use the new transportation option to reach it.
It's not that uncommon. Widening a road from 2 lanes (1 per direction) to 3 or 4 is something that happens all the time. Widening a road from 4 lanes to 6 is less common, but it does happen, and it does decrease the number of people using non-car transport along that route. There's a lot of wiggle room in between a quiet street and a classic American stroad.
@@davidpolansky5848 It only happens if the numbers say it happens. A widening happened recently in my city. Normally I would say that it was simply a bad investment due to the opportunity cost and lack of value. However, the city got caught in a really stupid deal where it would have cost them more to cancel the widening than to go through with it so they just did it. It made the area worse, but it had no measurable effect on demand for either car or non-car modes.
Well in the video where he said that the capacity of urban roads is limited by intersections, he also said that lane capacity still applies to rural roads (as in any roads that don't have access points), so in turn the "expected" way of induced demand might still apply to them.
@@jyutzler so it didn't happen with one particular road? how precise was the demand measurement? if it really didn't affect demand, why not? what was that road like, and is it generalizable to all other kinds of roads?
@@davidpolansky5848 It didn't affect driving capacity because the road wasn't anywhere near capacity anyway. It didn't affect biking because it was already a mediocre place to bike. The changes didn't really make a difference. I think the expansion was originally sought due to a planned business park that never materialized. If it were up to me, they'd abandon any notion of it being an employment center and build a mixed use development neighborhood with all the modern accoutrements. But city leadership has zero vision.
i experienced the same thing myself - here in berlin was a bottleneck where the motorway was not finished for over 1000 meters and of course there was frequently traffic jam - then the gap was closed and for a while there was no more traffic jam - the traffic jam was in other places at the entrance to the motorway and at the exit - a year later i saw the first traffic jam exacte at the place where the gap had been closed - which is literaly what all the experts say
I don't think that it is true that widening doesn't increase throughput. Maybe in some cases it doesn't, and it would be often be cheaper and just as effective to just optimize the intersection, but if a road is congested, adding a lane will certainly add throughput. It will likely not improve travel times or congestion though, and will make driving easier and other modes worse, therefore inducing demand
I saw it and it's a good video, I am just (perhaps pedantically) saying that it's not true that adding a lane inherently doesn't increase capacity, since as you say yourself a lane can handle 1800 vehicles/hour. There are plenty of highways that move more than that because they have multiple lanes. Obviously there are a lot of negative externalities to doing this and I am not advocating for it, I'm just saying that adding a lane to a hypothetical one lane road with no traffic signals that becomes congested when you try to move more than 1800 vehicles per hour would have more vehicle capacity with an added lane
@@nathang4682 We're talking about the urban context, which applies to suburban contexts as well: intersections limit your throughput. Always. It doesn't make sense to widen a road through a city or suburb when the intersections themselves can't handle the traffic. When we move to actual highways this limitation basically ceases to exist because highways don't have intersections and not many onramps and offramps. What we do see however, in the specific case of highways is that it would be more effective to keep lanes to the minimum required to fit the actually needed capacity so you can splurge where it's actually helpful. This happens to be why Dutch freeways in around 90% of all cases don't go beyond 2x2 lanes. It's actually really hard to exceed road capacity of a 2x2. But like Steffen said in his last long form video, it's counterintuitive. I would like to add the fact that Americans don't know the first thing about driving on highways, so they'll get clogged no matter what you do.
so in a sense, widening an urban road actually restricts supply for everything other than cars, for which you can never create enough supply to begin with.
Great explanation, and one that I hadn't thought much about before!
As a rider of LA's metro, it's nice to see a train station that's actually wide. We have some station platforms that are so narrow two people can't pass each other in certain parts, such as by the stairs.
I love the short videos on a single topic.
Nos Países Baixos é impossível gravar um vídeo mais longo enquanto se espera pelo comboio. Obrigado!
Thanks so much for the gritty details of traffic engineering. The United States population really needs this information if things are going to change in a positive direction.
Is induced demand actually a bad thing in itself?
My point is: most movement (mobility in general) should provide some form of social and economic benefit that should be weighed against the negative external effects of the movement.
We do not want to negative external effects (from car traffic) but we do want to positive impact of mobility.
I think that is an very import distinction, because also improvements in public transit (higher throughput, higher speeds) induce demand.
If we always argue against induced demand, we basically also argue against investment in any other form of mobility infrastructure.
(Same goes for other areas, if you invest in healthcare, you ultimately only create more demand for health care)
Induced demand is a good thing in the abstract. It means people are doing things that they want to do. The problem is when there are significant negative externalities involved. People moving to the exurbs and commuting may be logical for them but it stinks for everyone else.
Thanks for making these videos. I enjoy your long and short form content. Take care to not burn yourself out.
That was a really interesting question, and answer, but it was quite complicated to wrap my head around it. Had to watch the video 3 times and pause several times to make sure I heard everything correctly and make sense of it. Now I get it (think). Also your title is not accurate, we're not talking about "road widenings" here, we're talking about URBAN road widenings. Very different scenario to highways or arterial roads. I can certainly see how by widening a main shopping street that can't physically get any wider in terms of the buildings, but where you take away sidewalk space is going to cause problems for people accessing things along each side, and the reverse could make things better for them, and how if the intersections are already at capacity, you're not really increasing capacity, and thus not really inducing any more demand because people can't get anywhere faster. I do think it depends on road width and how busy the road or street is though. The effect could differ a lot if you tried it with a main shopping street versus a large residential road lined with just houses or apartments, where there isn't a ton of access required. Right?
It's true of urban roads where there are many possible routes, but widening a rural road (especially one with few alternative routes) doesn't seem to induce demand to the same extent. It also has the benefit of improved safety especially if a single carriageway is converted into a dual carriageway.
For an urban area, while more expensive to build, a bypass is usually going to deliver a much better return than widening an existing thoroughfare.
Thank you for this video! So, I have a couple of questions:
Does this bottleneck effect of intersections also apply to bicyles? And if so, what is the average of through-put?
Is it fair to say that people will use the most convenient (=safest/quickest) form of transportation and so an widening of the road only increases the number of vehicles/bicyles/pedestrians if the road work made it a more convenient path to a destination?
In some cities there are rings within the city that are just for cars and that have comparitively few intersections and no destinations. For these kinds of roads, would a widening of the road increase the number of vehicles/the capacity? And how much does the number of intersections affect the capacity?
Thank you so much again :)
@@donnasmith6738 My thoughts:
Bottlenecks absolutely apply to bikes. Forced stops at lights and stop signs are a major drag on biking, though less for e-bikes. I don't know the metrics.
Mode choices are very personal. Cost, convenience, speed, vibe, and safety are all factors, but hard to model because everyone is different.
Bicycles have roughly the same capacity in vehicles per lane-hour as cars. Roughly one vehicle every two seconds.
Cars can carry more than one person but the average in reality is 1.5ish.
Bikes can fit about three lanes in place of one car lane. Also, speeding up to their optimum speed of roughly 15km/h is much quicker than cars accelerating to 50 (bikes are shorter, so you need cars moving quicker to fit a car every two seconds through), so you get better throughput in intersections, which is the main urban constraint.
@@matejlieskovsky9625 thank you!
Question about local access streets and through roads: How do you go about building a proper network of these in a city with a lot of quite narrow streets and complicated road/street geometry?
Example: In Tartu (Estonia) you have huge intercity buses driving through streets too narrow to include a bike lane with current street geometry.
You could make streets one way and have the other direction use another street but then that direction might take a long detour from an important stop.
You could route major bus lines through the bigger roads that do exist, but then again you might miss an important stop.
Maybe just slowing buses down with traffic calming would work but I imagine it'd be pretty uncomfortable to cycle.
Any thoughts on that?
It seems to me that your answer is only relevant in a narrow set of cases, when other modes are being shifted away due to the widening. No one is walking, biking, or taking a non-existent bus along an urban highway no matter how many lanes the highway has, for example.
Induced demand is usually about making a desirable destination viable in a way it was not before, getting people to use the new transportation option to reach it.
It's not that uncommon. Widening a road from 2 lanes (1 per direction) to 3 or 4 is something that happens all the time. Widening a road from 4 lanes to 6 is less common, but it does happen, and it does decrease the number of people using non-car transport along that route. There's a lot of wiggle room in between a quiet street and a classic American stroad.
@@davidpolansky5848 It only happens if the numbers say it happens. A widening happened recently in my city. Normally I would say that it was simply a bad investment due to the opportunity cost and lack of value. However, the city got caught in a really stupid deal where it would have cost them more to cancel the widening than to go through with it so they just did it. It made the area worse, but it had no measurable effect on demand for either car or non-car modes.
Well in the video where he said that the capacity of urban roads is limited by intersections, he also said that lane capacity still applies to rural roads (as in any roads that don't have access points), so in turn the "expected" way of induced demand might still apply to them.
@@jyutzler so it didn't happen with one particular road? how precise was the demand measurement? if it really didn't affect demand, why not? what was that road like, and is it generalizable to all other kinds of roads?
@@davidpolansky5848 It didn't affect driving capacity because the road wasn't anywhere near capacity anyway. It didn't affect biking because it was already a mediocre place to bike. The changes didn't really make a difference. I think the expansion was originally sought due to a planned business park that never materialized.
If it were up to me, they'd abandon any notion of it being an employment center and build a mixed use development neighborhood with all the modern accoutrements. But city leadership has zero vision.
Great detail! By “Long through movement” you mean transient traffic or people not from the area?
i experienced the same thing myself - here in berlin was a bottleneck where the motorway was not finished for over 1000 meters and of course there was frequently traffic jam - then the gap was closed and for a while there was no more traffic jam - the traffic jam was in other places at the entrance to the motorway and at the exit - a year later i saw the first traffic jam exacte at the place where the gap had been closed - which is literaly what all the experts say
This has nothing to do with induced demand, but rather fluid dynamics.
OT: Was that a sprinter? Looked like a sprinter.
PAVE THE WORLD!
One World!
One People!
One Giant Slab of Asphalt!
I don't think that it is true that widening doesn't increase throughput. Maybe in some cases it doesn't, and it would be often be cheaper and just as effective to just optimize the intersection, but if a road is congested, adding a lane will certainly add throughput. It will likely not improve travel times or congestion though, and will make driving easier and other modes worse, therefore inducing demand
Watch the latest long video to understand why
Are you my city councilman? "I have been concerned when we change the number of travel lanes on our arterial roads..." 😂
I saw it and it's a good video, I am just (perhaps pedantically) saying that it's not true that adding a lane inherently doesn't increase capacity, since as you say yourself a lane can handle 1800 vehicles/hour. There are plenty of highways that move more than that because they have multiple lanes. Obviously there are a lot of negative externalities to doing this and I am not advocating for it, I'm just saying that adding a lane to a hypothetical one lane road with no traffic signals that becomes congested when you try to move more than 1800 vehicles per hour would have more vehicle capacity with an added lane
@@nathang4682 We're talking about the urban context, which applies to suburban contexts as well: intersections limit your throughput. Always. It doesn't make sense to widen a road through a city or suburb when the intersections themselves can't handle the traffic. When we move to actual highways this limitation basically ceases to exist because highways don't have intersections and not many onramps and offramps. What we do see however, in the specific case of highways is that it would be more effective to keep lanes to the minimum required to fit the actually needed capacity so you can splurge where it's actually helpful. This happens to be why Dutch freeways in around 90% of all cases don't go beyond 2x2 lanes. It's actually really hard to exceed road capacity of a 2x2. But like Steffen said in his last long form video, it's counterintuitive.
I would like to add the fact that Americans don't know the first thing about driving on highways, so they'll get clogged no matter what you do.