Deep inside the Atlantic magazine's Cities section, an article on restructuring a bus network! The city is Tallahassee, and the redesign team included Samuel Scheib of StarMetro, who comments here now and then. The redesign took an all-radial system, reduced the number of lines but increased their length, and introduced some new non-downtown connection points and even some lines that don't go downtown at all.
Few journalists would consider the topic interesting, but the article by Emily Badger takes the lay reader through the issues, highlights the relationship to city-building concerns, and generally helps it make sense. I wish every journalist — or even every New York Times reporter — took this kind of care to understand a planning issue. Given how well Badger explains the issue, I wouldn't have minded if she'd also interviewed some riders who are personally inconvenienced by the change. But most journalists cover only the latter, as though their goal is to maximize rage rather than understanding.
This quote from Scheib was interesting:
“If you talk to a land-use planner, typically they would want you to keep … service focused more on the downtown because they want more people to live downtown, in that dense environment. I’m all for that, I’m all for urbanization, I’m all for denser places,” Scheib said. “But the reality is that people need to get to work. And you’ve got to go where the jobs are."
I can assure you that this change won't damage downtown. I was hanging around Portland's TriMet in 1982 (in the indispensible role of teenage transit geek) when they totally restructured the inner city bus system, creating a grid pattern with many crosstowns that don't go downtown at all. Several of those crosstowns are now among Portland's most productive lines. But downtown Portland survived, to say the least.
If anything, Portland didn’t go far enough. There are major gaps in north-south service in the SE and east-west service in the NE. There are a still a lot of trips that require folks to go through downtown for no reason. This is especially aggravating when traveling between SE and NE, since it requires crossing over the river twice! In any case, it’s great to see more transit agencies realize that ridership downtown gets inflated by people needing to transfer there when it is not actually their destination.
Portland has light rail that radiates from downtown, that may well be the difference.
@francis. No, the Portland grid was created in 1982. The first light rail line opened in 1986. The grid, in fact, was an important process of setting the stage for light rail.
“If anything, Portland didn’t go far enough. There are major gaps in north-south service in the SE and east-west service in the NE. There are a still a lot of trips that require folks to go through downtown for no reason.”
That’s very true. As one solution, when idly putting thought to that problem recently, I came up with the idea of a frequent 8-15th/12th line. It would be equivalent to the 8 on 15th NE, then pass Lloyd Center MAX Station, and after that be equivalent to the 70-12th. People that took the 8 downtown could take the MAX there, ditto people that took the 70 to Rose Quarter.
I reckon that through the saving in resources by cutting a considerable amount of distance from both routes, a 15 minute all-day frequency could be possible on both the north and south ends of the routes, making the route useful for quite a lot of transfer trips.
I am extremely unimpressed by Tallahassee’s efforts. It takes some of the principles of the frequent grid, but misses out the “frequent” – a crucial part of the usefulness of a grid. Not only are the frequencies low, but they vary from service to service – such that few transfers will be dependable from one trip to the next.
Since not everyone wants to go downtown, many people had to transfer on the old network, and many people will have to transfer now. But whereas before now, they could enjoy the facilities of the downtown plaza, which I’m guessing included seating and AC, now some passengers will have no choice but to spend a long time at a street corner. Journeys may be a shorter distance geographically, but far less pleasant to make.
If you want to emphasise transfers in areas that merit low frequencies, there’s really no good solution but a pulse network. One could probably be created following the general principles of Tallahassee’s new network, but it would mean making a choice about what frequency to settle on.
I’d strongly advocate schedules that repeat from hour to hour, which would imply, on a network generally based upon 40- or 50-minute headways, deciding which routes merited a 30-minute headway, and which only merited a 60-minute headway. It looks like Tallahassee could also manage to standardise on a 40-minute headway on every route, but that’s just guesswork based on the present headways.
50 minute headways are worse than 60 minute headways… one of the bus routes near me runs hourly on Saturdays, and I know exactly what times it runs (10 minutes to the hour southbound, 5 minutes after northbound). If it was 50 minutes, I’d have to conult a timetable every time.
One transit agency fairly near me (Mississagua Transit) runs buses on a key corridor every 19 minutes, which is a stupid headway! if they ran every 20 minutes, you’d get an easy-to-remember schedule, at the cost of two fewer buses per day.
I absolutely agree. But what’s even worse than either a 50 minute or a 60 minute headway is needing to transfer from a bus at a 50 minute headway onto a bus at a 60 minute headway, so you never know whether you might have the best part of an hour to wait. This is the sort of thing that Tallahassee’s network features.
Zoltan, another idea thrown around is to merge the 6 and the 70, since the part of the 6 south of I-84 will be redundant with the new streetcar extension.
I think the 73 could run down 20th or 28th to form another crosstown.
A 28th crosstown would be lovely; the problem with it would be the lack of a connection with MAX. Which just goes to show that there really, really needs to be a MAX station at 28th, and preferably 20th too.
I’d configure such a thing as a combination of the 9 and 10 routes.
With every due respect, for someone to claim a 60min headway to be better than a 50min headway, implying a 20% lower frequency, just because it is easier to memorize, is to cater for laziness in a way that is just not justifiable altogether.
People have smartphones, Internet everywhere etc. to check departures.
A transit service is either too frequent (more than 6-8 services/hour) that it doesn’t require planning when reading to a stop, or not, case in which timed transfers are more important than an oddity of times of low computational power called clock-face schedules.
This reminds me of students whining that their university classes don’t start at the same time every weekday.
Not everyone has a smartphone. And not everyone with a smartphone can use it everywhere. My London experience would’ve been more pleasant with free data roaming.