In my book Human Transit, I argued that the underlying geometry of transit requires communities to make a series of choices, each of which is a tradeoff between two things that are popular. I argued that these hard choices are appropriate assignments for elected boards, because there is no technical ground for making one choice or the other. What you choose should depend on what your community wants transit to do. Examples of these choices include the following:
- Ridership or Coverage? Should transit agencies be maximizing their ridership, which is easily done by abandoning service to low-ridership areas, or should they spread out service throughout their service area so that everyone has some service, despite the predictably low ridership that results. (HT Chapter 10)
- Connections or Complexity? If you try to design a transit network in which people won’t have to transfer, you’ll end up with a network with high complexity and low frequency. I first explained this here, and later, and more rigorously, in HT Chapter 12.
- Peak-first or all day? While it’s widely assumed that transit is most productive during the peak, this is not always the case, because service running only on the peak, and especially only in one direction, can be massively expensive to operate. Should the peak commuter be the primary focus of a transit network even given these costs, or should a transit agency seek to build an all-day network of services that encourages reliance on transit at all hours, while still meeting peak capacity needs? (HT Chapter 6).
- How far do you assume people will walk? Assuming higher walking distances allows transit lines and stops to be further apart, increasing both speed and frequency. Research has found repeatedly that people will walk further to better service, though when you first propose to increase walking distance they will complain. This is an especially tough tradeoff because frail seniors and disabled people are small in numbers but have a particular aversion to walking. (HT Chapter 5).
- Should transit be protected from congestion? Doing so, by giving transit a lane (or trackway) not shared with traffic will dramatically improve reliability and thus the degree to which people will find they can trust the service. However, this requires a city to display a profound commitment to transit in the allocation of its streetspace. (HT Chapter 8).
Note that while the framing of these choices tends to make them sound binary (“Ridership or coverage?”) the choice is not between two boxes; it’s of a position on the spectrum where the two terms define the extremes. You don’t have to choose between ridership or coverage, for example, but you have to choose a point on the spectrum between them, and you have to accept the mathematical fact that, as on any spectrum, moving toward one extreme implies moving away from the other.
In the Epilogue of HT I briefly introduced the idea that for each of these choices, one option seems to trigger a positive-feedback loop, while the other option does not. Here is a new and expanded diagram of that idea. The image below is fuzzy, so download here: Download Abundant access diagram streamlined ’15-05-08 (Tip: Print it on A3 or 11×17. There’s a lot here.)
What do I mean by positive feedback loop? On each of these choices, the option that points toward the center leads to a particular kind of network that supports all of the goals implied by all the centerward choices. For example, if you plan your network for maximum ridership rather than coverage, you also generate a network that encourages higher walking distances, that supports easy connections rather than complexity, and that tends to present the strongest case for facilities that protect transit from congestion.
On the other hand, the options that point outward from the center tend not to support each other. You can achieve the goal implied by these outer choices, but (arguably with some exceptions) each one will require a separate effort. For example, if you believe walking distances to transit should be low, as seniors tend to advocate, then you need to put parallel transit routes — and the stops on those routes — very close together. This does nothing to advance any of the outer “outward” goals, whether it be growing peak demand, or avoiding connections, or achieving a luxurious experience that will compete with a BMW.
Another way of saying this is that the choices that point inward all point to a single set of network design principles, and a single set of consistent priorities, while the choices pointing outward lead to many different kinds of network design and competing priorities.
What happens when you move toward the center? These choices converge on a single, consistent goal that I propose to call Abundant access, which means:
The greatest possible number of jobs and other destinations are located within 30 minutes one way travel time of the greatest possible number of residents.
[Why 30 minutes? The question is tangential to my point, and other figures could be used, but if one has to pick a figure the most solid basis is Marchetti’s constant [PDF], the idea that humans throughout history have tolerated about one hour of daily travel time. Obviously, shorter or longer periods may apply to trips other than the commute.]
For any individual, abundance of access can be visualized using a map of isochrones, like this one by Conveyal for downtown Portland, which shows the area you can get to within a fixed amount of time on some combination of transit and walking.
(In these maps, blue is a 15 minutes travel time, green is 30 minutes, and pink is 45 minutes. Unlike some imitators, these calculations include waiting time and therefore accurately convey the impact of frequency.)
Of course, any quantification of abundant access must consider not the area but the amount of stuff in it: jobs, retail, recreational opportunities, hospitals, all the things that form the universe of destinations. As a matter of public policy, it must also maximize over the largest possible share of the population. In the case of transit, the population over which abundant access is most cost-effectively maximized tends to be a more urban, high-density population, because among readily available measures, density is one that best predicts the intensity of ridership that will arise from a service investment. (Fortunately, that’s fairer to than it sounds, because people living at high densities use much less road infrastructure per capita than people living at lower densities.)
The goal of abundant access has several kinds of appeal.
First, it can be measured objectively without recourse to psychology or culture. Ridership estimates are based heavily on travel times that approximate the notion of abundant access, but they also add psychological factors that are less stable, such as observed preferences for particular technologies. These factors may be emotionally vivid, but like many emotional factors they are likely to change with time and especially with generations — just as emotional attitudes toward cars are changing now. Abundance access measures a fact that is entirely objective — travel times. Unlike emotional reactions to technologies, the value of access has been constant across millennia of human experience.
Second, abundance of access is literally a quantification of freedom, in the sense that matters to us in transportation. Isochrone maps like Mapnificent’s, in particular, show us our freedom in a very immediate way: here is where you are free to go, now. Abundant access measures the transportation element of opportunity of all kinds, which is one of the main reasons people have moved to cities since their invention.
The concept of freedom is sadly undervalued in much urbanist discourse, and I am always looking for ways to reintroduce it. Much urbanist writing, for example, is blatantly prescriptive (“you should want this kind of community”), which feeds conservative stereotypes of urbanism as manipulative or coercive. We need to be able to talk not just about ideal communities but about freedom and personal responsibility, a frame in which all the great urbanist ideas — and all the urgent environmental imperatives — can be stated equally well. In that frame, the key idea is not “the good” but “choice,” where freedom of choices also implies responsiblity for your choices.
So again, all of this work is descriptive, not prescriptive. I’m not saying that you should like the goal of abundant access; that would be value judgment, and you get to make those for yourself.
I am saying, though, that abundant access as an idea has certain features and consequences, including a tendency to be self-reinforcing. As mapped along the various axes of choice, abundant access is a single consistent vision whereas the opposite choices lead to many unrelated visions.
Still, you don’t have to like it. My role, as always, is to describe consequences clearly, and help people choose.
[Updated 8 May 2015 with streamlined diagram.]
I’d like to say again on the “how far people will walk” that for older and disabled people this is hugely dependent on what is there when they get there. If what they reach is a pole with a potential ten minute standing wait, they need to be pretty close together. If on the other hand it’s a shelter with seating for that wait, they can be further apart.
What I mean is, walking time and standing time need to be considered together — they’re not different things for many of us. And if this were properly understood, providing seats at the wider-spaced stops could be a relatively easy fix that would relieve a lot of those initial complaints.
Jarrett,
This is interesting, but I am wondering if it is one step too simple. I now often ask if the models used to explain how transit works can explain the Zurich transit system’s success. This model seems to work for explaining Curitiba and Bogota, but not so much Zurich.
Does this not direct us towards a few radial rapid transit lines, rather than a frequent transit network (including rapid transit) providing decent access to and from most places in a region? I know your whole book does not, but diagrams get taken out of context.
Eric. I think the degree of voluntary transit dependents achieved in a place like Zurich does in fact depend on a more comprehensive network than just a Radial one. So I would contend that the ultimate ridership achievements in Zürich, which depend on voluntary transit dependence, depends also on the comprehensiveness of network. This does not of course necessarily mean that every suburban or rural bus Is essential to Zurich’s outcomes.
This all seems plausible and makes sense, but there’s something I don’t quite understand. I like the focus on geometry, but I normally think of time and space as dual in an important sense when talking about geometry, so that they should show up symmetrically in all considerations, unless there’s some explanation for why not. But it seems to me that there are two natural parallels between time and space considerations that seem to be reversed on this picture.
Peak vs. all day seems to be the time equivalent of the question ridership vs. coverage. The first asks whether we should run transit at all times or just the times when we can get the most people, and the second asks whether we should run transit at all places or just the places where we can get the most people.
Similarly, “how far will people walk?” seems like the space equivalent of the issue of frequency, which you don’t mention here, but which I appreciate your focus on. Should there be a boarding time every few minutes or should they be less frequent? Should there be a boarding place every few hundred yards or should they be less frequent?
Perhaps there’s a sense in which you’re giving a similar answer to all these questions – there should be stops that are approximately evenly separated in time all the time except for the middle of the night, and approximately evenly separated in space over the entire city except for extremely low density or geographically complicated areas. But maybe there’s a bit more that can be said about the apparent asymmetry in your answers to these questions? Is it just that you’re giving symmetric answers, but that the proposed alternatives in time and in space pull in different directions?
Or, to paraphrase P. T. Barnum (in followup to Kenny’s comment):
“You can serve all of the people some of the time, or you can serve some of the people all of the time, but you can’t serve all of the people all of the time”.
The suggestion seems to be to, given this limitation, elect to serve SOME of the people (those that live in transit-friendly places) ALL of the time; to make transit a premium product in those places that it serves well, and a limited one in those places that it doesn’t. And of course, transit agencies do ultimately make this choice in their service boundaries–generally not providing any service outside of cities. You can’t take the bus to the middle of nowhere, at least not unless nowhere is on the way to somewhere.
The political difficulty is that many people live in hard-to-serve places, yet demand transit service (or seek to defund it otherwise). And even ignoring that political issue, many useful destinations may be in otherwise hard-to-serve locales; the problem of serving industrial parks, for example.
Kenny: I think the concept that answers your question is complexity. Adding more trips on an existing route does not increase network complexity, while adding a new route does. Complexity is bad, because it makes the network less legible and can present the rider with too many choices and make it too difficult to pick the optimal one.
Also, you’re right in that service span is analogous to coverage, and the analogy can be streched a bit further: most trips are, ultimately, round trips, so the service span needs to cover both halves in order for the trip to be possible by transit. This is analogous to the way, in the space dimension, a service (or combination thereof) has to serve both the origin and the destination of the trip in order for the trip to be possible by transit.
Kenny. I’m happy to have triggered such a philosophical response, but I don’t see much of a basis for proposing analogies between time and space. Space is symmetrical in ways that time is not. Time is also emotive in different ways than space, because space appears to be stable while time is not. Space generates feelings of power, while time generates feelings of fear. Every life story ends with bad news.
So I wouldn’t follow you in conjecturing that temporal and spatial dimensions should be analogous. They are as different as banquets and landslides.
Scotty’s analogy points to interesting questions about how you might begin to differentiate a land use framework based on the local characteristics the transit picture. Correspondent land use trade-offs are implied in Jarrett’s diagram and an interesting deliberation can be had with each of those petals. If a city can’t readily alter the transit picture for abundant access, because of geographic and service limitations, then maybe the question can turn to what we can do with the land use ramifications.
Ha Jarrett, I just saw your response to Kenny’s questions. I was musing that perhaps it is better not to tease out too much “symmetry” from your flower diagram. I think the parallelisms evinced by paired binaries are indeed useful and interesting, but I would use the classical meaning of “symmetry” here to understand the important point. I’d suggest we understand “symmetry” more correctly in this case as the proportional adjustment of subpart dimensions, each with distinct function and comparative importance, to create a harmonic composition. When you say that binaries are sometimes in tension when they slant outward but harmonize in a predictive form when they slide to the center, that suggests classical “symmetry”. This indeed is Vitruvius’ usage of the term when he described the “symmetry” of the human body and Greek temples. He used the term in description of dimensional proportions. Symmetry meant to him a whole system that harmonizes. (He didn’t limit the term to bilateral relationships.)
Jarrett – That sounds right from the emotional perspective, but for all the reasons you’ve mentioned before I would still like to see how much can be explained from the purely geometric perspective. For all the reasons you mention about potential cultural changes relating to technology preferences and the like, I can imagine cultural changes that change people’s relations of power and fear to space and time. (Admittedly, as an academic philosopher with a background in mathematics, I may also be more interested in the strengths and limitations of geometric thinking than someone who actually has to work with real transit planners at the end of the day!)
One way that stop spacing and frequency are geometrically similar is that they affect the time required to access transit in the same way. Increasing the distance I have to walk to a stop and increasing the time I have to wait when I get there are basically the same, in terms of discouraging me from using transit. (I’m not sure if time and distance trade off at exactly walking speed, or something slightly different.)
One way they are quite different from the geometric perspective is that frequency affects everyone approximately the same, while stop spacing affects people differently. That is, everyone lives at every minute of the day and could want to access transit at any minute, but different people have different spatial distributions – I mostly access transit that comes to my neighborhood and to a few other neighborhoods, and especially transit on my block, whereas someone who lives a few blocks down accesses it at different places. Decreased frequency affects everyone equally, while increased spacing affects people who live (or work or shop) between stops more than people who live (or work or shop) near stops.
Another geometric difference is that although stop spacing and transit frequency play the same role from the perspective of the passenger trying to catch the transit, once you’re on it, frequency is irrelevant, and tight spacing still slows you down. That is, spacing and frequency affect costs and access in the same way, but frequency has no effect on speed, while spacing slows down speed, so from a purely geometric point of view there is already an asymmetry here.
I was already in my previous post starting to realize one of the geometric asymmetries between spatial and temporal coverage. There are only 24 hours in a day, so temporal coverage really has a finite extent, and you can promise to cover every time. However, space is effectively unbounded (from the point of view of a city or metro area), and there will always be places that one must leave uncovered, so there are reasons to prioritize the densest trip-generating areas that don’t apply to prioritizing the densest trip-generating times.
Anonymouse seems right about complexity too.
Like the new map of choices alot and I think your penultimate paragraph here is an important discussion point for all of us.
Also Mapnificent can become a social equity tool for understanding transit travel times. I’d be interesting in your take on Title VI and environmental justice and how we operationalize “equity.”
I am sure you remember the framework you created years ago for the discussion for the Washington State Transit systems regarding costs of service – wish you would also post that. It is a good starting point for discussions about educating policy makers and the public about the anatomy of transit costs. Best Wishes, Stefan
I have trouble with the notion that it’s only a small group of seniors and disabled people who don’t like longer stop spacing. There are a lot of people out there who aren’t disabled, but aren’t really in condition to do 1/2 mile plus walks to a bus stop (and won’t do it).
Often it seems to me that when we talk about stop spacing, we act like all the potential passengers live on the bus corridor itself. But of course they don’t, many of them have to walk to the bus corridor, all the longer if we’re thinning out the network of routes to have fewer but better lines. In an ideal city, the density would be concentrated on or very close to the corridor, but that’s not necessarily the case. In California cities, the bus corridor is often primarily commercial, and most of the residents live on the streets behind it.
I wonder what our target spacing is and therefore how long we’re asking people to walk, and how long people will walk before 1) they feel like they’re unduly increasing their travel time; 2) they feel like they’re physically unable to do it; 3)they feel like it’s just too much hassle.
Let’s look at cities that have higher than average ridership and see what characteristics they have in common. The three large cities that I am familiar with are Toronto, Chicago and New York in that order.
Toronto has a grid system of surface lines that cover over 240 square miles of the city. There is also a subway system of two main lines which fit into the grid system. Toronto has a large number of subway stations where the surface lines enter the fare control area so that connections (transfers) can be made between surface lines or surface and subway lines without the need to show a paper transfer. I believe that this is unique among North American transit system. Toronto also allows free transfers between all surface lines. This September is also the fiftieth anniversary of the introduction of a grid based suburban bus system around the old city of Toronto which comprised 34 square miles. The majority of lines run a service interval of 15 minutes or less for most of the day, from 6:00 a.m. to 11:00 p.m.
On top of this the Province of Ontario introduced a radial system of commuter rail lines that carry over 175 000 passenger per day, mainly into the urban core. The greater Toronto area runs both types of systems; a grid system with 18 plus hours a day of frequent service with a mainly rush hour radial system to carry commuters into the city centre from the outer suburbs. The province is spending a lot on infrastructure upgrades to run the radial system in the off peak hours. It seems that there is a need for the two types of systems.
Chicago has a grid bus system on top of which there is a radial rapid transit system, the “L”. On top of this there is the commuter rail systems which is basically radial and mainly rush hour with some off peak service. The Metra Rail commuter system is mainly radial to get passengers to the downtown area. Again the two types of system compliment each other.
New York has a grid Subway system in Manhattan with a grid bus system. The subway system into the surrounding Burroughs seems to be a radial based system with local, mainly grid based buses. The extensive commuter rail system is mainly radial in nature.
Other cities that I have visited seem to have mainly a radial oriented system without free connections between lines. This seems to limit the usefulness of the transit systems.
From the fall of 2009 to the spring if 2010 my wife and I toured the US using the inland river systems to the Gulf of Mexico and the Atlantic Intra coastal Waterway and the Hudson River back to Canada. Having grown up in Toronto I was appalled to see how poor public transit was in much of the US. There are cities in Ontario with a population of 40,000 to 50,000 that have better service than cities of 400,000 to 500,000 in the US. Many times I would ask if there was a bus service to an area I wanted to get to and none of the people, mainly middle class whites over 40 had any idea if there was one. They would direct me to a car rental agency. Most middle class middle aged or older citizens had never been on public transit and would not consider using it. This is the main problem with expanding transit in much of the US.
I could also not get over the amount of land given over to roads and parking in the “Downtown Area” of most cities. This is a tremendous waste of money and resources, not to mention a blight on the urban fabric of the cities,
So to conclude my wishy washy answer, both systems are needed to serve a large urban area effectively. There is a need for a frequent grid based system to serve the basic transportation needs of the main urban area with a mainly rush hour radial system on top of it to provide commuter service from the outlying area to the CBD. In my un-humble opinion the priority is frequent grid then limited radial.
You know, I’ve been thinking about acceptable commuting times recently. Don’t have the time to find the sources, but apparently the average car commute is about 25 minutes, while transit is about 45 minutes.
As we move into a less car dependent future for cities, we really need to get transit travel times down to the 25-30 minute mark, door to door. Some of this will come from improving the urban form, so that you can reach more destinations within such a time frame, but it will also have to come from improving transit speed.
I understand the rationale of Jarret’s argument on accessibility being different than mobility. Yet, if we are talking about transit being an option that substitutes individual car mobility, isn’t geographic accessibility an important feature still?
From a strictly utilitarian viewpoint, people could still find their “resources” close to home given the appropriate land planning, but isn’t providing quick access to other places people don’t necessarily “need” to go an important feature of any transportation system that aims to take market share from cars for purposes other than commuting?
Using a smaller scale comparison: people could theoretically take vacations only at the closest beach, ski resort or wooden mountain. Yet, many travel across the globe just for sake of seeing something different, “clinching” another country/park to their list etc.
On a metropolitan scale, people could theoretically live only limited-transportation lives where they patronize the pub that is 3 blocks away all the time, take their children to the playground 10min walking from home all the time, attend a church (if that is the case) that is necessarily the closest to home and even try to participate only on charities that actuate on their walking radius. But is that necessarily a goal? After all, history of transportation is one about widening horizons and slashing time distances.
Andre Lot …
Again, “coverage” means “coverage to areas whose low density or obstructed street patterns yields a very low ridership per unit of investment.” In a city with abundant access, you do have access to competitive choice as a customer, because there are many options within the area that transit can serve abundantly. Your argument, which I’ve made many times, is the whole argument for cities with abundant access in general, and is the argument against “slow transit” which traps you in your neighborhood or adjacent ones.
However, in an abundant-access city, it will still be difficult to get to houses in culdesac golfcourse developments, and to go into the country for the weekend you’ll probably need a carshare. Ultimately, it’s the suite of modes that delivers.
So does gentrification and the inversion of cities help or hurt the cost-effective goal of choosing quality network over route coverage?
On the one hand, the inversion near the CBD forces cities to move away from the radial spoke system to a network of crosstown routes.
On the other hand, the economic reset that has sped up the filtering of suburban housing stock farther from the CBD now creates even more demand, where the built environment is least efficient to run quality network.
Either way, the cost-efficiencies are at risk of ever materializing. Basically, it takes quality built environment to reap the rewards of quality network. And many US metros lack quality places to walk, much less transfer.
Given the limited urban environment of many US cities, the gentry will continue to demand luxury circulators in their small, recently invaded core, while the captives increasingly need to overcome longer travel distances across the sprawling metroplex.
The pedestrian wilderness is a unique American challenge to creating more quality network. The place to start may be linking key nodes with limited-stop services. If developers would actually build places with human scale in mind, then transit could focus more on the mobility between quality places, as opposed to the quantity of mitigating transport to overcome such places.
Nobody thinks that to visit Paris (assuming you don’t live in Europe), you need to own your own airplane.
It might be more useful to think in terms of a range of concentric isochrones – 15 minutes, 30 minutes, 45, etc. – rather than concentrate on just one such isochrone. The reason is that for specialized things, usually there are only a few in a metro area and unless they can afford CBD rents they’re not in the CBD, but it’s unrealistic to expect everyone to have 30-minute trip times. The more specialized the trip, the longer a trip time people accept. The Marchetti constant is not that much of a constant; people in transit cities where most non-work trips are short and done on foot readily accept much longer commutes (the Tokyo average is an hour, but that may be a special case coming from diseconomies of scale). Conversely, for people with many errands to run, the 15-minute isochrone may be more important than the 30-minute isochrone.
Concretely, my commute is about 25 minutes in each direction door-to-door. My errand trips range from 5 to 10 minutes, and are often bundled into the return trip from work. And I only need to be on campus 4 days a week. So counting only those trips, I spend ~45 minutes a day traveling, well below the average range of 60-90 minutes. But then once a week I go to an event that rotates between Burnaby and Langley, and this drives up the average a lot. Suddenly I’m spending 60 minutes a day traveling on average. So in my situation, I chose a neighborhood based on a combination of 30-minute trips to UBC, a wide availability of local errand destinations within 10 minutes on foot, and acceptable rents. (I didn’t know I’d ever find myself in the suburbs when I signed the lease, and even now it’s only once a week so there’s no point moving.)
I wrote a long, by no means aggressive, comment about the difference between access and competitiveness – which has apparently been deleted. I’ve re-read the commenting policy and see no way in which it would have run afoul. Jarrett, I’d appreciate a note as to why.
M1EK. DId it never post? If so, that’s a fault of TypePad, which loses track of you if you’ve been writing for too long. Open a new comment window and paste the comment again. I did not delete your comment. Jarrett
Sorry, it’s long gone now – but I did see it post – saw it reflected in the static page, even.
I find it very odd that teasing apart accessibility and mobility as distinct concepts is apparently such a novel insight! It’s been in the landuse / regional science / transportation literature for years, particularly in the more geographically alert parts of that literature. Indeed developing various measures of the accessibility yielded by different transport system configurations is a whole cottage industry of its own in certain parts of academia. Having said that measuring accessibility via PT is perhaps less commonly done than for road networks…
This is a decent picture.
I would point out, FWIW, that “Technology as Tool” leads you, inevitably, to rail on trunk routes. “Technology as Goal” is what gets you lots of buses.
Worth noting: the removal of streetcars, railroads, and trolleybuses in favor of gasoline and diesel buses was, historically, driven *very clearly* by a “technology as goal” thinking.
The GM-Firestone Tire-Standard Oil conspiracy is only the most obvious evidence of this. For them, the technology was quite explicitly the goal, because gasoline buses made money for all three of them, while streetcars didn’t.
Alon writes:
“It might be more useful to think in terms of a range of concentric isochrones – 15 minutes, 30 minutes, 45, etc. – rather than concentrate on just one such isochrone. ”
I do find such maps far more useful than maps with only a single isochrone.