Columbus, Ohio's metro transit agency, COTA, has now released a new network plan for public comment. As in the recently unveiled similar plan for Houston, I led the network design task on this project as part of a consulting team led by IBI Associates.
Again, the core idea is to expand the Frequent Network — the network of services that run every 15 minutes or better all day — so that more people have service that is highly useful. Here's the existing Columbus area frequent network :
And here is the Draft Proposed Frequent Network:
In Houston, we achieved similar expansion solely by reallocating existing resources. In Columbus, there was a small budget for expanded service, but still, 90% of what is achieved here is the result of reallocation: removing overlapping routes and deviations, removing duplication, and in some cases removing service that very small numbers of people were using.
Details of the plan are on the COTA website, here. The total proposed network is here. Note that color denotes all-day frequency: red is 15, blue is 30, green is 60. The plan does many other good things, including a major expansion of weekend service.
You can upload the existing network, for comparison, here: Download Existing System Frequency Map
Again, if you're in the Columbus area, please comment to COTA using this special email: [email protected]. At this stage there is no decision about whether to implement a plan such as this one. Any final plan will be revised based on public comment that comes in over the next couple of weeks. That means that if you like the plan, it's important to comment to that effect, as well.
Something I don’t see mentioned anywhere here or the COTA website – but that’s actually a Big Deal – is that for the first time there will be practical bus service to Port Columbus International Airport.
Right now, you have two options. You can ride the 52 directly to Ohio State (but only on about 30 days a year when it actually runs) or take the 92 crosstown. With this plan, there’s a half-hour-headway local and an hour-headway express directly from the airport to downtown.
> »Note that color denotes all-day frequency: red is 15, blue is 30, green is 60.«
How does one come up with this idea? The proper™ way to denote frequency is line width or continuous/dotted/dashed; things that instinctively point to how important a route is.
Red might stand out as the most important and frequent routes, but how is one supposed to see whether green or blue is the next step, without looking at the explanation?
(Obviously I’m biased, here they use red, blue, green and yellow for different routings through downtown.)
Sascha. The meanings of the colors is a standard we’ve evolved at our firm. It seems to work for most people, though we’re open to critiques of it. I use dashed lines only to mean short spans, such as peak only, because to me dashed lines give an instinctive sense of “ephemeral” or “not always there”. Sorry if this isn’t proper™ !
Sacha,I’d have to back up Jarrett on dashed lines. They usually mean a peak-hour express line or some part-time extension or turn-back on many maps I’ve read.
I feel like you’ve explained it before, but how did fifteen minute headways become “frequent?” I’m sure compared to whatever they were running prior to this it’s far more frequent, but I can’t help but think that anything more than 8 or *maybe* 10 minutes kind of strips “turn-up-and-go” of any meaning.
IMHO 12 minutes should be the threshold for “frequent.” On a recent trip to Las Vegas, I was very impressed with their excellent bus system, where many routes run every 12 minutes. Getting around the city was easy, convenient, and involved very little waiting (of course as a tourist, my schedule was particularly flexible.)
Jarrett, we spoke at PSU. I showed you my Trolleybus Reconfiguration design for Seattle. In your recent works I can now see similarities in viewpoint though I still praise couplets. Trolleybus lines arranged in couplets e/w to climb hill past Broadway to 12th Ave Turnarounds, similar to Lake Union, n/s on 1st/3rd Aves between Mercer and Jackson for frequency and curbside stops. (usure 1st Ave Streetcar with Left-lane Center-stations). Reduced overhead clutter, minimized turns. I hope to apply this ‘circulator’ theory in San Francisco.
(unsure 1st Ave Streetcar with Left-lane Center-stations)
I went with 5th & 4th Ave rail straight to King Station then straight to Alaskan Way on a Jackson and Main St(historic railbed) in a couplet.
The rail turn and route on 1st Ave will be high impact
and probably accident-prone. Waterfront Streetcar can
climb Queen Anne hill on W 3rd Ave to Mercer.
Final note: I dislike Mercer West design.
I doubt the ‘Drill-fill Sea-fence’ technique is credible.
I oppose Bertha DBT vehemently.
WRONG TYPE tunnel for unstable soil conditions.
Yeah, Seattle’s got me down.
Box Cut-Cover Tunnel/Seawall still possible,
and makes too much sense.
@Jarrett & Dexer, I agree that dashed and dotted lines ought to be reserved for part-time services; but I regard colors as an un-intuitive (and thusly improper) choice for showing frequency. Why no differing line-widths?
Oh, and are there any routes even more frequent as Low Headways’s “low headways” of 15min? Would a frequent network of a 12min threshold be have any frequent lines (outside of downtown) at all?
(Hmmm… the longer I’m staring at the map, the more obvious becomes the connection between colors and frequency: The red lines in the center are the most frequent (obviously), then a ring of infrequent blue lines and an outer ring of more infrequent green and dashed yellow lines.)
Jarrett, how many annual revenue hours and boardings does COTA currently provide? How many annual revenue hours and boardings did you forecast in the plan? If you can provide passenger miles in addition to boardings, even better, given the increased emphasis on transferring.
12 vs 15 minutes is not that much of a difference. If you found 12 minute service to be effective in Las Vegas and 15 minutes to be ineffective elsewhere, maybe it’s because Las Vegas buses adhere better to their schedule, or because the network is better designed, or serves your personal destinations better.
15 minutes is the ‘cut-off’ wait period for most people.
Beyond 15 minutes subtracts ridership. 12 minutes is reassuring.
5 minutes is the urban center ideal.
Streetcar lines, ‘circulators’ (Denver 16th St Shuttle)
can displace or work with traffic, especially on couplets.
Pedestrian travel patterns benefit ‘connecting street’ small business.
Trips can be orderly planned, especially with circulators.
The Seattle Streetcar line planned will turn from Jackson north on 1st Ave in the Left lanes with Center stations; high impact, least stations. A Trolleybus Circulator on 1st to Queen Anne instead. My route for Seattle Streetcar is Jackson to Alaskan Way to Coleman Dock, sooner or later extend to Interbay/Ballard.
At Left is WSDOT’s Concept ‘D’ CRC I-5 Interstate/Hayden Island.
Oh thank you so much, Warshingtonians?
ODOT’s version Concept#1 was a hushed-up rejection.
Stop Bertha and Mercer West. SDOT/WSDOT totally misled.
15 minutes is a reasonable headway for ‘frequent’ service, IMO. When I lived in the Boston area, I considered any route than reliably operated every 15 minutes or less as ‘walk up’ service. Maybe that’s just me, though.
I think that 15 minutes maybe okay service, but that their should be another category for 8 minutes (or something like that). Does a city like Columbus or Houston have any lines that are better than 10,8, or five minutes? This would be enough to influence my housing behavior compared to an every 15minutes bus.
15 minutes is considered the upper end of turn-up-and-go service. While relatively frequent, 15 minute service can still cause long waits if transferring, and in Europe, any headway over 10 minutes is generally considered to not be considered frequent enough to enable non-scheduled connections.
But overall 15 min service is a good starting frequency, and it has been shown to increase ridership and make transit more usable. Just like the minimum service frequency for a sort of attractive service is 30 minutes.
I am sorry to sound critical Jarrett, but I have to voice a concern with this plan, as with the Houston one.
The issue being that once again, instead of providing a total transit service, this plan like Houston, is telling people in some areas that they can’t have good quality transit service, because of the developments they live in.
As we know from Paul Mees’ work, and experience in other cities like Toronto, transit can succeed in suburban settings. But it will not succeed if we tell people they only deserve a bus every 60 minutes.
Another issue with this plan as with Houston, is that I think the idea of coverage vs ridership is a flawed concept.
If you are working towards providing a viable transit service, then transit must provide a basic attractive service level to all people in the built up areas of the service area. Picking and choosing areas based on development or density (which more and more is being proven to not be the sole factor in transit ridership) I think does us no good.
Potential and current riders could care less if 25% of the service area has outstanding bus service, if 75% of the service area has poor service or little service at all. People need to be able to travel from anywhere to anywhere, and they can’t do that if some areas of the city have poor service or no service.
American transit planning really needs to understand the need to provide a basic attractive service to all residents and workplaces, and then adding service in high ridership sections.
This idea that we have to choose between the two just does not hold up.
The high ridership routes rely just as much on lower ridership routes to feed them. As well as the idea that people can travel anywhere they want to go. If they can’t, because the planners have decided an area is not transit friendly. Then potential and current riders will be more inclined to get a car, if they are feeling cut off from mobility.
American transit systems generally to do not provide enough coverage, and promoting that they further cut coverage service is not good. It is especially not good when half or more residents in most American metropolitan areas don’t even have access to transit within walking distance of their homes.
Interessantes Artikel, Sehr Informative.
@Mike: The reason no American transit agency does this is not because its planners are bad, it’s because they don’t have the money.
Some land use patterns are more conducive to mass transit than others. You can throw as many bus hours as you like at much late-20th-century US suburbia and achieve very little ridership. Real transit agencies and their planners have real financial constraints, and given a choice, they’re going to spend their money where it makes a difference. Where it increases the number of trips people will actually make on transit because they can get there in a reasonable amount of time. Where it prevents people from being left at the curb. Not where it chases hypothetical riders with commute patterns that are nearly impossible to serve effectively.
I understand your answer. However there are plenty of suburban areas in American cities which are no different than Canadian, Australian, or other western suburbs. The only difference being American ones don’t have transit, and many have planners who think they can’t support it.
This creates two issues.
1: By not providing a uniform coverage, transit does not provide access to all employment and other regional destinations in a given area.
Brookings did a great report on how so few jobs in most American metro areas can be accessed via transit. And that is party due to planners who do not wish to think outside of the box, or actually try offering an attractive service to areas they deem not worthy of transit.
2: By telling some areas they are not worthy of transit, we are setting up transit to continue to be a second class travel option. While at the same time leaving whole segments of the population unable to even give transit a try, even if they wanted to.
This will lead to less support for transit as transit continues to be nothing but a last resort option of travel, and not a viable way to access all areas of the city and region.
I fully understand financial constraints. However this issue goes beyond financial constraints. It has more to do with the engrained nature of the vast majority of American transit planners, where transit is seen as just a last resort of travel for poor people, and that suburbs are evil places that can’t support transit. It is not always the planners fault as that is the environment they grew up in.
Providing transit does not always mean fixed route services. There are other ways to ensure everyone has basic access to transit.
Overall you need equity in transit planning. But there is no equity if we continue to cut off whole areas of our cities and suburbs from transit. It is an injustice for the residents not served, but also for the people who cannot access employment and other destinations.
This issue also brings up the fact that planners have to be better advocates. If we continue to tell agencies it is o.k. to pick between coverage or ridership, instead of demanding more funding. Then of course there will never be enough funding.
If transit is to become a viable travel option, it must provide access to almost all regional destinations and residents. Not pick and choose based on poverty and density.
Like I always say. Under the density and need logic, Finch Ave East in Toronto would have a bus every 60 minutes if following American transit planning. Instead of the bus every 90 seconds that it currently offers, going through low density suburbia.
In the Columbus model, the plan could have been done in a way that brings a high frequency core route to all areas of Columbus, filled in with feeders.
Under the plan presented here, the north part of the city basically gets everything at the expense of the south.
One could say the north side has more destinations, which it does. But that does not mean a high frequency spin service could not have been provided to the south as well.
Mike, bring it up with the politicians, not with the transit planners.
The politicians will not understand, if we have transit planners telling them it is o.k. to pick between coverage or ridership, or suburbs vs the inner city.
Unless politicians are told that the status quo is not good enough, then most unless they are planners themselves will not know to fund more simple service expansions.
The Transit Capacity and Quality of Service Manual has now eliminated letter grades for frequency. In the second edition, though, they gave an A for frequency of every 10 minutes or better, a B for 10-14 minute service, and a C for 15-20 minute service. On many American transit routes, 15 minute service would be a big improvement over what exists now.
Thanks for this very interesting blog post; I found it mainly because I’m already a Columbus, Ohio, area resident.
1) I wish COTA would post the information to their own website more prominently, however, thank you for doing that same in lieu of that. Your blog was actually the first I saw of COTA’s Transit System Review.
2) The second proposed service map was actually useful to me as a frequent user of the COTA service you discuss in the post (!), and it looks like I’ll be adding an additional route. More to point #1.
Anyway thanks for the useful post.
A very nice contribution I liked.