Human Transit

request for information: routine closures of transit streets

Posted on September 16, 2011 in Chicago, Downtowns, Information Requests, Minneapolis – St. Paul, Vancouver

Most cities that I know have one or more major downtown streets where parades and other major civic celebrations tend to occur. San Francisco's Market Street, Chicago's State Street (pictured), and New York's Fifth Avenue are obvious examples.

These same cities, if they value transit, often want this same street to be the core of their transit system, because they want transit to deliver customers to the "front door" of the city.

So it's normal to see huge, complex reroutings of transit service when one of the civic events is happening. The legibility problems of this shift are accepted because they happen only a few times a year.

Vancouver, however, does something different, and I want to verify how unusual it is. Vancouver's core downtown transit street, Granville Mall, is closed to buses (and all vehicles) every weekend evening. Specifically:

On Fridays, Saturdays, Sundays, Holidays and the day before a Holiday, buses will run along Granville each day until 9pm, then switch to the re-route, travelling northbound on Seymour and southbound on Howe. The re-routes will stay in place until the close of service.

So many of Vancouver's most important street-running transit lines (mostly trolley buses in this case) shift from one downtown street to another at 9pm on almost half of all days. So the process of explaining and remembering where to find the bus is complicated all the time, which is quite different from finding the street closed for a very rare parade.

What North American cities do this routinely with their main transit street? Nicollet Mall in Minneapolis is the only example that comes to my mind, but its role in the transit network is much less important than Granville's.

To be clear, I'm not interested in reopening the Granville Mall debate, where many advocated closing the street to buses entirely. I am interested in the precedent as it might apply to other important transit streets, in Vancouver or elsewhere.

school starting times can transform transit efficiency

Posted on September 15, 2011 in Oakland

Matt Conway writes:

AC Transit (Oakland, CA and surroundings) recently came to an agreement with schools to adjust the start times at those schools to reduce the cost of providing bus service. I've never heard of schools adjusting [start times] to save agencies money!
http://www.actransit.org/2011/08/26/end-of-summer-adjustments-in-bus-service/

Start and end times of schools can have a huge impact on transit agencies that are expected to serve them. If three large schools choose to start at the same time, rather than at staggered times, they can dramatically increase both the fleet needs and the operating cost of the transit system. To serve such a short, sharp peak of school demand, the transit agency needs more drivers to work very short shifts, and must own more buses for them to drive. Staggered school times can replace four buses with one, and replace two 1-2 hour driver shifts with one 3-hour shift. Note that drivers must often be paid more to work a short shift; often the rule is that they must be paid for four hours even if they work one. That's understandable; would you be willing to commute to work for just one or two hours' pay?

When I worked as a network design consultant for small agencies across California and the American West, I encountered this problem constantly. Schools were under intense budget pressure and were cutting their own school bus service, shifting their transportation needs on the transit agency. Schools would change start and end times for their own reasons, and falsely assume that the transit agency could deliver a fleet of buses at whatever time they choose. It's an example of a common problem wherever many government agencies overlap; your agency can appear to "solve" a problem by shifting it to some other agency.

Has this situation improved? Are many public transit authorities now co-operating with school districts to ensure that expensive transit resources are used efficiently?

good network maps: clear without color?

Posted on September 13, 2011 in Legibility, Maps, Portland

As I mentioned two days back, the new Spokane, Washington network map [Full PDF], designed by CHK America, is exceptionally clear in presenting the layers of the network so that you can see all of the following at a glance;

The network of frequent services, useful to you if you aren't willing to wait long.
The network of all all-day services (frequent or not), which is the total network if you're travelling midday.
Supplementary express services, mostly peak-only, that are likely to be useful only to the regular peak hour commuter. These always contain a high degree of complexity, so they must be presented in a way that visually recedes from the rest of the network, so that the all-day network is clearly visible.

The key point is that each layer is never allowed to distract from the ones above it.

But it has one other important feature that I should mention: If you look closely, you'll see that its content is still there if you copy it to black-and white. Line widths and styles distinguish all the service categories from each other. The only exception is the distinction between "Shuttles" and "Frequent Routes", both wide lines, and this matters less than it seems because the shuttles are frequent too.

All this is relevant not just because the world is still full of black-and-white photocopiers, but also because of color-blindness. Matt comments:

Recently I noticed playing around with Scribus, the Open Source Desktop Publisher (http://www.scribus.net) that "Scribus has a well developed tool, the Color Wheel plug-in, which helps to guide you selecting complementary colors, as well as visualize colors seen by folks with certain kinds of color-blindness.". So turning on an option for the three or four different types of colour-blindness lets you see what the colours look like for someone with each type.
Perhaps there is a chance that in the future a pdf viewer could incorporate something like that, such that bus maps (mostly they're rendered into pdf) and anything else in pdf could be seen as if you were colour-blind. Then the bus companies could experiment and release colourblind friendly maps.

All good, but the simplest solution for color-blindless issues is to design maps so well that the information is all there in line weights and styles, so that the color is supplementary — very, very helpful for those who can see it but not essential. The Spokane map does this.

The current Portland map also tries to do this, with a different line-weight for each of its four layers. The four layers are:

Light rail: colored line with black outline.
Frequent Bus: heavy solid colored line (and, if you look really closely, a yellow-shaded line number)
Basic Bus: slightly thinner colored line.
Peak-only: dashed line.

Portland's TriMet uses different colors for different lines, but if you copy it to black-and-white you should still be able to make out these four line weights (though not, of course, the feeble yellow shading of frequent line numbers).

I agree with many observers that the distinction between frequent bus and basic bus is insufficiently strong on the Portland map, whereas the Spokane map shows this distinction dramatically. In the past, when I've tried to use the Portland map as an example of clear delineation of network layers, I've been told that the distinction just isn't clear enough, so now I'll use the Spokane map instead. Still, Portland's intention is clear enough.

stop spacing: risks of multiple patterns

Posted on September 12, 2011 in Stop Spacing

In a recent post on stop spacing, I quoted an eloquent defense of very closely-spaced stops based on the needs of mobility-limited persons. This view is unfortunately in tension with the need to move stops as far apart as possible to increase the speed and reliability of operations, and thus attract more passengers.

I was surprised at how many comments suggested that the answer is to provide a mixture of local-stop and limited-stop or "Rapid" services. This is absolutely the right thing to do on the extremely major streets where you can afford very high frequency (say, every 10 minutes or better) on both patterns. Most New World cities have just a handful of these streets. Examples include Mission and Geary in San Francisco, Broadway and 41st Avenue in Vancouver, Western Avenue in Chicago and Wilshire in Los Angeles. Key features of these streets are (a) very high demand supporting two frequent services and (b) relatively long trips, so that speed advantage of a rapid stopping pattern outweighs the longer walking time it may require.

But if you can't afford high frequencies, overlaying local-stop (every 200m or less) with Rapid or limited-stop service (every 800m or more) can be really unsatisfying. Should you wait 11 minutes for a local at your stop, or walk to a Rapid stop 400m away where the next bus comes in 14 minutes but might be faster? Those are the uninspiring choices presented to a customer when the frequencies are only, say, every 15 minutes but two patterns are being offered.

When you consider the major streets that support frequent locals plus frequent rapid services, Seattle's long and busy Aurora Avenue might come to mind, but in fact, King County Metro abolished that pattern a few years ago, creating instead a single stopping pattern so that they could run the highest possible frequency. That's the key. Especially for trips of under 10 km or so, waiting time easily overwhelms in-vehicle time in determining door-to-door travel time. So in those cases a reasonable "compromise" stop spacing — not as close as senior/disabled advocates want, nor as far apart as speed advocates want — is actually the fastest at getting everyone where they're going.

Another approach, which I advocate looking at, is to accept that the constituency for very closely-spaced stops may also accept poorer frequency. If you look at the part of a route that is halfway between two Rapid stops, and thus most dependent on the local stops, and you then subtract all the people there who are willing to walk 400m to the rapid stop, you end up with a fairly small number of people. So perhaps locals should be less frequent than rapids. Transit agencies sometimes try to be neutral about this, carefully calibrating local vs rapid service based solely on current ridership. But in fact, transit agencies have a strong reason to prefer rapids: faster service is cheaper service to operate, because transit vehicles complete their cycles is less time, and we pay drivers by time, not distance.

But it's definitely not adequate to say that we can resolve the conflict between close and wide stop spacing simply by running two separate lines on the same street. We can in a few places, and if public transit had a lot more money we'd do it in a few more. But transit agencies need a stop spacing policy that works for the more ordinary street, where you can afford maybe 10-15 minute frequency on just one line. That means just one stopping pattern, so we have to pick one.

spokane: a very clear network map

Posted on September 11, 2011 in Frequent Networks, Loops, Maps

Spokane Transit (Washington USA) has a new network map out that is one of the clearest I've ever seen. It carefully delineates not just frequent services from infrequent ones, but also presents cases where basic infrequent lines combine to form frequent segments, and ensures that peak commuter express services are visible but can't distract from the clear all-day pattern. The whole thing in its full glory is here: Download Spokane 2011 map. The legend, too, is both clear and wonkish at the same time.

Congrats to Spokane Transit for designing this map for the public.

Full disclosure: I was the lead planner on a restructuring study for Spokane Transit back around the turn of the century, and if I remember right, our project invented the continuous two-way frequent loop of Lines 33 and 44. (The loop is closed on the west side as Line 20, as shown on the full PDF.) Despite many excellent improvements (and some sad service cuts) since then, it's great to see it still operating.

This kind of two-way loop is often useful as a way to combine radial lines and grid elements into a single service. Line 33, for example, intersects Line s 24, 25, and 90 in a grid manner, one line north-south and the other east-west, allowing for a range of L-shaped trips via a connection at this point. However, Line 33 also flow through so that the same segments can also be experienced as radial; if you stay on the bus, you'll get downtown eventually, and to a lot of other useful destinations.

I sometimes caution against excessive attachment to loops. In some contexts, with far more financial resources, I might applaud the breaking up of this loop, as I did of London's Circle Line. Given the extreme financial pressure on US transit agencies, though, I would contend that Spokane's frequent loop was an efficient solution, maybe even an elegant one.

UPDATE: The next post on the Spokane map, looking at colorblindness issues and comparing the map to Portland's, is here.

dissent of the week: impact of wider stop spacing

Posted on September 4, 2011 in Dissents, Stop Spacing

Moving bus stops further apart achieves a range of benefits in speed and potentially frequency, as I've argued here and here. Zef Wagner at Portland Transport recently laid out a similar case in the Portland context. The post and most comments are worth reading, but I wanted to quote this dissent from commenter Cora Potter.

Please keep in mind that you have to weigh any time/cost savings of stop consolidation with providing adequate mitigation for people with barriers to accessing transit stops (disabilities, cognitive barriers that require clear landmarks, etc).
Shuffling these riders off to LIFT [demand responsive paratransit for disabled persons] is not a cost effective solution. The average LIFT ride costs nearly $30 (one way). Even with a conditional eligibility system, you run the risk of increasing costs to the point that you might actually end up with more operating funding deficits than gains. In addition, while LIFT is a complimentary service, meant to be comparable in timing and experience for the consumer, there are tolerances to the system built in and a LIFT trip is usually longer, less direct and less convenient in that it requires an advance reservation. The ability to access the fixed route system is usually the least restrictive and most convenient means of travel for people with disabilities who can reasonably access a transit stop.
When you raise the bar for reasonable access by spacing stops at .25 mile, or 4 blocks or greater, you start significantly limiting what a person with barriers to walking or cognitive barriers can accomplish and make it far more of a challenge for them to reach a stop in a reasonable amount of time, or with a reasonable amount of challenge. The difference between walking one block in the cold and rain vs. walking three blocks in the cold and rain can present a huge challenge to a person who can not walk at a 3 mile per hour pace, and is more sensitive to cold temperatures due to normal aging.
So – just keep in mind that wider stop placement, particularly for bus service, will exclude a segment of the population from accessing fixed route. You need to weigh the gains in convenience for people who have few or no barriers to accessing transit with the social and financial costs of excluding the people who do experience barriers to access. My personal opinion is that everyone loses when older adults and people with disabilities are excluded from fixed route by design. You might gain 3 minutes and the sense that you're getting somewhere faster – is that really worth service cuts on other bus lines to offset the increased costs for ADA paratransit? Is it really worth socially isolating people with disabilities? And, with the aging of the population, we really need to start taking the needs of older adults seriously and not constantly tune our system to serve just the needs of commuters.

A quibble: I disagree with dividing the population into "older adults" with "commuters," as Cora does in the last sentence. The consituency for larger stop spacing is basically anyone who wants to get where they're going as soon as possible, and who is open to walking a bit further in order to achieve that. This is not everyone, but it's a larger group than "commuters," who are people making rigidly scheduled trips to or from work or school.

But I agree with both the practical and ethical dimensions of the comment. The mobility issues faced by older and disabled people are real and the costs of serving them with paratransit are high. I also believe that allowing older people to become socially isolated is both economically and ethically unacceptable.

On the other hand, it's very, very hard to organize the mobility limitations of small, scattered numbers of people into facts. In the early 1990s, when US transit agencies were struggling to implement the Americans with Disabilities Act, no issue was more vexed and emotional than eligibility for expensive paratransit services. For example, ADA is clear that age, by itself, is not a disability — an important insight because people are not only living longer but remaining able-bodied longer. The process by which a local government assesses people's disabilities is obviously highly emotional for the individual in question and challenging for all concerned.

A similar issue applies to bus stops. When we move bus stops further apart, lots of people complain, and people with mobility limitations complain bitterly. But given the high cost of the close stop spacing, how should those claims be assessed? In some transit agencies, the default policy has been: "If anybody complains, cave in." I'm not ashamed of suggesting that transit agencies be more forceful in articulating the tradeoffs, which requires educating their elected officials about what's at stake. Cora's done an excellent job of articulating one side of that tradeoff.

I'd also note that in Europe and Australia, stop spacing is wider than in North America. In Europe, many stops also have major infrastructure that signals their permanence. Everyone, whatever their mobility limitation, works with that. The result is that service runs faster and more reliably. So from that perspective, this is partly a question about how to transition to that outcome. It could be done most peacefully if it happened over decades, but ours is an impatient age, and transit agencies are under too much pressure to wait that long.

Part of why this is less of a problem in Europe, I'm guessing, is that the permanence and infrastructure of many stops signals clearly that you should take them into account when deciding where to live. If an older adult can anticipate becoming dependent on public transit in the future, the location of transit stops should be part of their location decision. This, to me, fits into a much larger agenda of insisting that everyone who makes a location choice — especially about where to live — should be required to acknowledge the transit impacts of that choice. Today it's still common to encounter the other sequence, in which someone (a) signs a lease or deed of sale or development agreement and then (b) yells at the transit agency because the service isn't what they desire.

So as usual, I don't have an answer, but I do think the question needs to be seen as geometrically inevitable, which means that those who disagree with us have a rational and ethical basis for doing so.

dueling academics on cost-benefit of rail

Posted on August 30, 2011 in Advocacy, Language, Rail Transit

Two new academic abstracts set up a striking debate, on the question of the cost-benefit analysis of rail projects. From Peter Gordon and Paige Elise Kolesar:

Rail transit systems in modern American cities typically underperform. In light of high costs and low ridership, the cost-benefit results have been poor. But advocates often suggest that external (non-rider) benefits could soften these conclusions. In this paper we include recently published estimates of such non-rider benefits in the cost-benefit analysis. Adding these to recently published data for costs and ridership, we examine 34 post-World War II U.S. rail transit systems (8 commuter rail, 6 heavy rail and 20 light rail). The inclusion of the non-rider benefits does not change the negative assessment. In fact, sensitivity analyses that double the estimated non-rider benefits and/or double transit ridership also leave us with poor performance readings. Advocates who suggest that there are still other benefits that we have not included (always a possibility) have a high hurdle to clear.

Meanwhile, from Robert Cervero and Erick Guerra, both of UC Berkeley:

The debate over the costs and benefits of rail passenger transit is lively, deep, and often ideological. As with most polemical debates, the truth probably lies somewhere in the middle of extreme views. Some rail systems have benefits that outweigh their costs, while others do not. Applying a commonly used transit-fare price elasticity to 24 of the largest light and heavy rail systems in the United States and Puerto Rico, assuming a linear demand curve, and accounting for a counterfactual scenario, we find that just over half of the systems have net social benefits. Although Los Angeles’ rail system does not “pass” our back-of-the-envelope cost—benefit analysis, as the network expands, it will begin to mimic the regional spatial coverage and connectivity of its chief competitor—the auto-freeway system—and approach the fare recovery rates of other large, dense American cities.

I don't have time to dig into the papers, but I think the difference in tone of the abstracts is interesting in itself. The issue of network completeness identified by the second abstract is of course critical. Discuss.

(Hat tip: Murray Henman. Post composed via wifi on the Amtrak Cascades, which is excellent!)

not a hiatus but …

Posted on August 30, 2011 in Meta

… but there are moments in the life of a consultant when project needs spill over and consume all of your time, even threatening sleep. I may not post again until the weekend, but I am still here. High-quality guest posts, or recommended cross-posts, are welcome!

faq

Posted on August 26, 2011 in General

Questions about the book Human Transit will be discussed here.

chicago: a new bus rapid transit plan

Posted on August 24, 2011 in Bus Rapid Transit, Chicago, Connections, Transfers

A Chicago-area planning nonprofit, the Metropolitan Planning Council, has released a plan for 10 Bus Rapid Transit within the City of Chicago. The final report is here: Download BRT TRB Report Final

The work, led by consultant Joshua K. Anderson, is admirably wonky. It analyzes a huge range of arterial segments to identify those that appear best from a standpoint of both constructibility, demand, and nexus with livability values. The report is a "screening" study, which means it seeks to narrow the range of possibility and encourage more detailed study of those that remain.

Bus Rapid Transit is defined quite vigorously:

BRT is 4 defined by four main components: 1) dedicated bus lanes, 2) at-grade boarding, 3) pay-before-you-board stations, and 4) signal-prioritized intersections.

This definition is met by almost none of the things now called BRT in North America, or at least not continuously from one end to the other. But screening is a time to be ambitious about such things.

The 10 corridors that survive the screening are shown on this map. They're a mixed bag: portions of segments, some of them maybe too short to be effective as BRT, but also two very long corridors, Western and Ashland Avenues, one of which is probably the most urgent BRT project in the city.

The study appears to be silent on whether these are envisioned as open BRT or closed BRT. Open BRT means that the infrastructure can be used by bus lines that flow onward beyond it to other destinations. To take an obvious example, an Irving Park BRT that ends at Ashland, short of the Red Line's connection opportunities and the high density of the lakeshore, is unlikely to be satisfying as a complete corridor. But if it's an open BRT, usable by buses that continue east, it could well be useful.

Unfortunately, a presumption of closed BRT, in which buses can't continue beyond the limits of the infrastructure, seems to be implied by the author's decision to discard super-dense Lake Shore Drive from the analysis simply because of the complexity of branching patterns that it requires; this assumption will have to be reconsidered in light of open-BRT best practice.

The study illustrates a common challenge in analyzing large, long transit corridors. Many of the key issues, including available right of way and "livability" impacts, are segment-by-segment affairs; if these dominated the analysis, the result would be a huge pile of largely disconnected short segments, which could not deliver the intended outcomes. So the author streamlined, discarding small segments and emphasizing larger continuous ones, which is quite right.

But issues of network integrity and completeness seem not to be fully considered. The report needed to step much further back and describe the underlying geographic structure of Chicago, which determines the type of services that could be relevant to citywide mobility needs.

Except near the lake, Chicago is an extremely regular grid of arterials spaced 1/2 mile (800m) apart. CTA follows this grid with a grid-pattern of long bus lines that attempt, as much as possible, to cover the entire length of an arterial all the way across the city. This achieves the important goal of grid completion. The purpose of each line is not just travel across that street but to complete a network in which people can travel from literally anywhere to anywhere else through a simple L-shaped movement:

For more on the high-frequency grid principle, see here. Obviously this structure only works, in its purpose to serve any origin-destination pair, if its constituent lines flow all the way across the grid to its natural edge. This is the problem with many of the proposed BRT corridors in the report.

Not everyone sees this grid, because Chicago also has an overlying radial system of rapid transit, which runs along diagonals pointing toward downtown. The two overlaid elements — radial trains and grid buses — work well together, but if you focus too much on the trains, which are mostly about going downtown, you miss the power of the underlying grid to complete trips on any origin-destination pair by a reasonably direct path. (The report discusses "network integration" only in the form of integration with rail. Confusingly, too, it gives heavy emphasis to connections with suburban commuter rail — whose poor frequency makes connection difficult — and little to the bus-bus grid connections that are the essence of the network's anywhere-to-anywhere versatility.)

Given Chicago's grid structure, and how well it already works, BRT needed to be understood as a system of grid accelerators, just like the Metro Rapid and proposed Wilshire subway in the similar grid of Los Angeles. Obviously, if you can concentrate particularly heavy demand on a few elements of the grid, you can justify an overlay of much faster service, stopping only at the grid connection points every half-mile.

On that score, Western Avenue is clearly a winner. It is the longest arterial in Chicago, running north-south the entire length of the city. Its extreme length creates reliability issues on a local-stop service, which has caused CTA to break it into three lines thus reducing its usefulness for continuous movement. BRT would be an opportunity to recombine these three segments to offer a service that would be understood as an intrinsic feature of Western Avenue over its entire length. Western is also far enough out of downtown that the direct paths it serves are much faster than riding rail into downtown and back. A vast range of trips between many parts of Chicago would find a Western BRT line useful.

None of the other corridors identified in the study can match Western in the utility that arises from extreme length with lots of connection opportunities. Ashland is obviously close. Most of the other proposed segments are simply too short, and would be useful only as open BRT segments used by buses that run further. Effective BRT has to serve long corridors, because the tradeoff that BRT requires of the customer — walk further in return for faster service — makes sense only for a fairly long trip.

To sum up, the report is very useful and highly recommended. But it misses (or at least downplays) two points that are missing in many similar studies, and that really matter:

Open or closed BRT? They're totally different, and if you're not clear which you mean, it's impossible to envision the service patterns, and thus the mobility, that your proposal will offer.
Integration with the total network, not just rail. This requires seeing how the whole mobility flow of the city works, and how each corridor would contribute to that flow. Localized analysis that asks where BRT would be easy to create or locally beneficial can easily lose this "forest" in its obsession with the trees. Understanding this principle would have required a much firmer focus on complete corridors that traverse the grid and make many connections, rather than the small fragments that are frequently proposed.

Still, the report can do a lot of good, and bravo to the Metropolitan Planning Council for sponsoring it. Chicago really needs to start accelerating its bus grid, especially on its busy, high-stakes, versatile corridors like Western. I hope this study helps to move that along.

Older posts Newer posts