Feedback on SORTA’s proposed 2013 service changes

SORTA recently wrapped up planning for service changes that will take effect in August and December this year. The changes they’ve proposed represent a fairly substantial(though certainly not huge) change to the shape of the network. As they work to finalize their plans, they’ve asked for feedback from the public; as a member of that group I intend to give them some, even though it’s now a little later than I might have liked.

I’ll write first about the changes themselves before going on to comment on the planning process more generally.

Basically there are a four major things changing.

There are also a number of smaller changes, but their effect is more local so I won’t go into it. You can see the complete list here.

The thing SORTA has been touting the most is the the thing I was most wary about before I attended the public meeting on the plan last week: The ‘Metro Plus‘ service, or as I’ll call it, the m+1, is to be their tentative step toward what might be called ‘bus rapid transit‘.

Metro Plus route map

Sneakily snapped from a binder left open by one of the planners. The map they’ve shared with the public is a little *ahem* …quirky. The loop at the top right is less practical and more a result of apparently not being able to find a good place for a layover.

For months it was unclear what kind of service this might provide or how fast it could be expected to go, but the rough details are now out in the open and almost all of my initial hesitation about the service has been wiped away. It will2 run from 6am to 10pm on weekdays with no weekend service, and have the surprisingly high frequency of every 15 minutes between 6am and 6pm before dropping to every half-hour for the rest of the day. That’s roughly the same level of service as the #4, #17, or #33 but for the lack of weekend and late night service.

As I understand it, it will be pretty much like every other local-running line in the system except with fewer and better-marked stops. I assume fares will be the same and will depend on the zone you’re travelling to or from.

The line will not, at least for now, have any designated rights-of-way or signal prioritization(though that one may be coming a bit later) and relies on the simple fact that it will be stopping less for any speed improvements. At the first public meeting in November, SORTA said the m+ would be about 15% faster than a comparable route with many more stops.

Metro Plus Bus

Source: SORTA
10 specially branded buses have been ordered for the line. They’ll be unusable for other services.

One thing about this service is still little unclear to me, though there may be a good justification for it: Why, if we recognize that lines in our most important transit corridors(like Mongomery Road, also served by the #4, or the others where ‘pre-BRT’ is proposed) are slowed down by too many stops, don’t we just strategically remove some of the stops thereby speeding up all service on that road? It seems like we’ve recognized a problem(too many stops in places) and rather than trying to fix it for the #4, or #17 or wherever else ‘BRT’ is planned, SORTA is overlaying those corridors with a substantially different kind of service that avoids the problem altogether. As such, we’ll eventually be getting two very distinct services on our most important corridors rather than one service that could be some sort of a mean between lots of stops and slow speed and what’s being proposed for limited-stop, faster BRT. One main line per corridor would be much easier for customers to understand. Two lines isn’t necessarily a bad strategy, but it’s not one that leaps out to me as obvious either and I’d like to hear SORTA’s reasons for it. TANK is working on strategically eliminating stops on many of their lines right now. Why isn’t SORTA doing the same even though they seem to acknowledge the same problem? What accounts for the difference in approach?

On to the next item, there’s effectively a shift in uptown service toward Vine St. This is a little hard to picture right now without a new frequency map(which will eventually be made), so I’ll do my best to explain it. In a nutshell, there are three main ways transit gets up the northern hill of the basin to the dense neighborhoods around UC: West Clifton Avenue, Vine Street, and Liberty Hill/Highland/Auburn.

Current(May 2013) routing

Current(May 2013) routing

The proposal is for the #19 to depart from West Clifton and replace the route of what’s now the #24(brick-red line above) up Liberty/Highland/Auburn before rejoining the #17 at Ludlow. The #24 and #39(brown line) would no longer go to Downtown at all, so the #19 effectively takes over for both in that neighborhood. The #51 also won’t come as far down as the Clifton Heights business district leaving Clifton Heights with just the #17 and #31. Meanwhile the m+ will be added to Vine Street with stops at the top, middle and bottom of the hill. When the m+ is combined with the #78 and #46 already going up Vine street, there will effectively be a very very high frequency of service between Corryville(East side of campus) and Downtown. Whereas now service to either side of campus is pretty evenly split, it looks like in the future it will be pretty heavily weighted to the Corryville side. This is a little inexplicable because the Clifton Heights business district at the corner of West Clifton and McMillan is currently the busiest cluster of transit stops outside of Downtown. Corryville is also a major hub for transit(fourth largest after Walnut Hills and Northside), but the cluster of activity around their business district is only about 65% of the size of Clifton Heights.

My applogies for the scribbles. This is Just to give you a rough idea of the intensity of transit ridership by neighbprhood...white to black is a linear scale and the coloured parts jump to a log scale. That means downtown is totally off the charts compared to the rest of the city.

My sincere apologies for the scribbles…for all the complaining I do about bad maps! This is Just to give you a quick idea of the intensity of transit ridership by neighbprhood…white to black is a linear scale and the coloured parts jump to a log scale. That means downtown is totally off the charts compared to the rest of the city and the brown of Clifton Heights is very significant. Exact data is here.

Corryville is also less densely populated than CUF which can probably explain most of the difference in ridership, but it’s also less well connected to places that are themselves popular transit destinations like Clifton and Northside. That factor at least will change somewhat with the changes to the #19 and certainly the addition of the m+.

I think as a result we’ll see a decline in transit use in Clifton Heights that will be nearly but not quite matched by a rise in transit use in Corryville. That in and of itself might be fine, but because of the difference in population density between the neighborhoods I suspect the distribution of service after the changes will be slightly less ‘fair’ with fewer people receiving higher-frequency service.

One of the planners at the meeting suggested the 17 could see slight frequency improvements in the future(not officially part of the plan), but I worry about the effective service cut to Clifton Heights because the #17/#19 (which I ride frequently) is already often standing-room-only.

The next big change is the merging of and additions to a few east-west running lines that do not/will not stop in Downtown. We’ll take these one at a time since there are only a few:

  1. The #24, will lose it’s downtown leg, instead ending in Fairview where the #51 currently turns around.
  2. The #51 and the #39 will merge at MLK north of UC’s main campus and lose their legs into Downtown. The new line will keep the number 51.
  3. The #41, a long low-frequency route through the northern suburbs will gain a substantial new segment and drop all the way down to the Glenway transit center.

Th first two generally seem like a pretty good idea. If people can rely on transfers to high-frequency lines running between the ends of these lines and Downtown, then there’s no need for the lines themselves to make their own foray into Downtown. That cut service will be applied to other parts of the system where it can be more useful.

Next on the list is the simplification of the routes of quite a few lines. This sort of falls into two categories. First is the elimination of what I call ‘only-sometimes’ route segments. These are portions of a route that are only used for a small minority of trips. For example the #31 currently makes 104 trips each weekday. Most of those trips are along the dark portion of the route shown in the map below,

The current #31 schedule

The current #31 schedule

but 25 of them, scattered pretty randomly(though I’m sure with intention) throughout the day, are one of four variations on that route. Effectively, this means that even if you know the basic route of the #31, you still probably can’t know exactly where you’re going unless you have a schedule on you or the ability to view the online PDF map. Such deviations are usually done for some purpose such as to serve a factory that only needs service at the end of the day, but it does make things more difficult for the general public. I’ve always found this sort of route-capriciousness extremely frustrating both as a cartographer and as a transit user. It’s exactly this sort of thing that makes people so crazy for the perception of permanence that things like streetcar tracks are supposed to provide.

But anyway, the fantastic news is that most of the only-sometimes segments are getting cut from lines for one reason or another, to be either eliminated completely or replaced by another and more consistently running line. This will of course upset some people’s daily routines but it will also make the system as a whole much easier to comprehend. Losing their only-sometimes segments are the #33, #19 and #31.

Subject to more basic simplifications are the #1, #24, #46 and #38X. Most exciting are the changes to the #1 which is now so convoluted and hard to follow that I very nearly printed t-shirts implying that the line was drunk. It will still be a little tipsy perhaps, though most certainly rideable.

route #1 simplified in december service changes

Map from SORTA

These changes are great. Let’s never go back to complex routing if a simpler route can do the same job. It’s much easier to walk an extra block than to understand where the #1 goes right now.

On to the more general comments!

There are a few things left out of these proposals that I think might well have been considered during a comprehensive review of the system.

  1. Downtown routing of pretty much everything is still a giant mess and even the new routes like the rerouted #19 or the m+ just add to the confusion rather than starting to simplify things.
  2. The new cross/around-town lines enable trips that use transfers between lines in a more meaningful way, but SORTA’s fare policy still discourages changing lines with a $0.50 fee. More clarity of direction on transfer policy is needed for future development of the system, particularly if we want more of a grid-like system with trips that rely on multiple lines.

I’d like to end by commenting on the method of feedback and community engagement itself. While I’m glad that SORTA is making a point of reaching out for feedback through public meetings, online surveys and by email and I know that they’re doing it in good faith, the information they’re providing and particularly the questions they’re asking (online at least) seem significantly less useful and informative than they might have been, both for interested citizens and transit users and for SORTA’s planners.

Similar to the way SORTA seems to think about their system generally, so in presenting route and schedule changes we’re not given a coherent picture of how these changes fit together as a network but rather a list of separate individual routes and the mostly unexplained changes each will have. I’m an urban planner and a huge transit nerd, and it’s taken even me hours of scrutiny and re-mapping work to really understand the effect of the proposed service changes. Without providing people with a bigger picture to make sense of the changes, I’m not sure how they could be expected to understand what’s going on here or to provide comments that would actually be valuable to SORTA’s planners.

The questions they’re asking feel to me kind of like a vote for your favourite route:

SORTA_survey

Does this route meet your needs?” is totally the wrong question to ask, unless you can ask it of the entire population or a representative sample. Online surveys like this have an obvious self-selection bias; you can easily imagine one person asking everyone on their block to use this form to “vote against” any particular change that they don’t like. The people reviewing the feedback, if they’re smart, will recognize that type of comment for what it is and throw most of them out so they don’t skew the results. Even if someone does manage to weed through the details available and form some valuable opinion on the changes generally, the form isn’t really set up for that kind of comment. That’s why I haven’t filled out the feedback form at all, in favour of talking with people one-on-one and writing here.

I think a great opportunity for public education about how the transit system works was really missed here with both the feedback methods and the way the changes were presented individually. The Hamilton County Budget Office for example does a great job of this. Every year they put out a public survey on budget priorities, but it’s so thick with completely necessary details that you can’t read it without coming away knowing the difference between capital and operating funds. The way that survey is set up, there’s literally no way to respond with something inane like “Don’t raise my taxes!” and “No service cuts anywhere!” just as that’s no way to balance a budget.

After a lot of research and conversation with the planners, these changes make good sense for the most part. Indeed, I’m quite happy with them and think they represent a sensible and reasonable improvement to the way the transit system will work. But I think it would have been much more valuable for everyone if the reasons behind the changes had been made clear from the beginning and this process had been a bit more open to the public. Not only would that have put these changes in context but it would have started to establish some basic standards that people would be likely to follow when talking about more publicly contentious projects or changes.

Do these changes meet your needs? By and large they should, but again that’s not quite the right question to ask.

Show 2 footnotes

  1. It’s just much nicer and shorter I think. I don’t like route names with words when all the other ones are numbers. A letter and a symbol? That’s not so bad.
  2. We’re still in the planning phases, so perhaps not. But it will probably be something very much like this.
Comments: 5
Posted in: Analysis | Plans | Priorities
Tags: | | |

Mapping ridership on a linear corridor

I saw something on the internet a couple days ago:

graph

…that made me want to analyse ridership stop by stop, and well, hopefully make more sense of it than this chart did. It’s kind of pretty, and I’d probably get more out of it if I were familiar with the system, but it’s hard to see any interesting trends here, mostly because the vertical scales vary quite a bit1.

Being the techy that I am, I wrote up a little PHP script to manipulate SORTA’s not-quite-public ridership data  from 20092. Stops are paired up, as well as I could manage for each direction. The green bars are passengers getting on, the blue getting off. Grey bars between them indicate the day’s total. Each side of the center line is a direction of travel.

a day's ridership on the SORTA #33

View full size to read the little stop labels. One pixel represents 2.5 riders(in the full size image)

I made a few of these graphs and they got me thinking about what I should have expected the data to look like. I’d never really thought about what transit ridership should look like theoretically. Did you notice that the balance of riders getting on or off shifts completely from one end of the line to the other? It makes sense. If a transit line were to travel straight through an area with even density, we might expect it to look something like this:

even ridership on a linear transit corridor

This on-off aspect of the data is pretty interesting, and I’m glad it wasn’t all aggregated to totals for each stop.

We see a shift in boardings and de-boardings because a line is more useful if it has a long way to go yet, and not very useful at all if you’re at the second to last stop. Not many people will get on at 7th St to go two blocks to Government Square but if they get on at 7th to go the other direction they have the whole world ahead of them.

And so we see each trip slowly shift from “on”-heavy at the beginning to “off”-heavy at the end with a solid mix of boardings and de-boardings in the middle.

Here’s the #33 again, in map form, and only going out from Downtown:

ridership on the #33 outbound

Remember that this doesn’t include the inbound trips. Not everyone is going one way, I’m just not showing the other half.

Here it is zoomed in on the Glenway/Warsaw business district:

Glenway business district ridership on the SORTA #33

Here’s what just that looked like in the chart by the way:

Ridership outbound on the #33

It’s only the outbound trip, so it’s only one side of the graph. If you look at an aerial photo, you can see pretty quickly that the ridership is higher where there are businesses and dense side-streets going off perpendicularly to houses. The places where riders drop off rather quickly are places where hills have precluded development and one or both sides of the street have either a lot less development within a couple blocks, or even just woods. The hill up from Lower Price hill(East side of the image) demonstrates this effect most notably.

ridership of the #33 over satellites

Transit users don’t visit the woods.

Dense areas are good places to run transit. It’s also good to connect dense areas that are divided by a sparse area like Queensgate. The stops along the way won’t show it, but the huge number of people who loaded on at Government square are still on the bus heading west to the Warsaw Glenway business district and beyond.

There’s a ton more to discover in here but it’s actually a lot of work preparing the data. If you want to do a little exploring yourself, here’s the script and the data that I used. The full stop-level ridership dataset from March 2008 is available on the Data page of this site and a more up to date version will be available very soon if enough people email Kevin, SORTA’s senior system planner. If you haven’t seen it yet, you may also want to check out the full ridership intensity map I made from the same 2008 data. Here are a couple more charts just to whet your appetite:

The #24:

chart of ridership on SORTA's 24

This is the exact same scale(for full size image) as for the full chart of the #33 above.

And the #51:

51 ridership

Same scale here too. Notice that rider’s don’t start using the #51 in earnest until it hits Clifton. That’s why the dip into Fairview is dropped altogether in SORTA’s new routing proposals.

It’s important to note for these two that it’s not that they’re necessarily “performing” a lot worse than the #33, rather their frequency is significantly lower so less riders could have been expected from the beginning.

Show 2 footnotes

  1. As has been said of it. Not being original there.
  2. Spreadsheet software just has nothing on custom, dynamic, database-driven scaleable vector graphics!
Comments: 3
Posted in: Analysis | Logic | Math
Tags: | | | | |

2008 ridership data-a preliminary retrospective analysis

2008 preliminary stop analysis

A preliminary analysis of ridership by stop in 2008, not including TANK, and not properly displaying dense areas, therefore not properly indicating the intensity of the most intense stops. But it’s enough to see strong patterns, and that’s better than nothing!

statistics for 2008 one day boardings

~71,000 people boarded SORTA buses on one day in March 2008.
The mean stop boarded 9.5 people, the median only 1(!) person.
Individual stops boarded between 0 and ~1450 people, though I’m sure government square hasn’t been counted as a single stop, thus diminishing that larger number.

Mean SORTA bus stop location

The mean location of all 5,000+ stops was
39.155687,-84.500106
3825 Spring House Lane
Cincinnati Ohio 45217
Thanks to Google for the satellites

Here’s the data: all_stops_2008

There’s a data CD with the current ridership-by-stop info coming in the mail any day now from the friendly planners at SORTA so we can all look forward to a much more detailed analysis of the new stuff once I get it. I love public planning meetings. They’re always the perfect opportunity to ask questions that in email would likely be ignored(like “Can you please give me all of your data?”).

Rule #1 of urban planning is that if you don’t show up to the public meetings, you will have less franchise than if you do. You’ve all already missed (or didn’t) the public meetings for SORTA’s preliminary transit plan, but it’s not too late to send emails. I’ve been advised that Sallie Hilvers will field comments/suggestions/questions by email and pass them on to the right people.

More commentary on that preliminary plan coming momentarily.

Comments: Leave one?
Posted in: Math
Tags: | | |

When TANK crosses the river

There’s a lovely logic to the way TANK arranges their stops in Downtown Cincinnati.

Government Square transit diagram

Lines running all day through Downtown–doesn’t include peak-only/express routes

This means that it’s easy for anyone to remember where to catch their bus, and particularly that anyone who’s just taking a quick run from Downtown to either Newport or Covington can just go to the right spot and jump on any bus that comes by.

Going to Newport on the Levy? Go to Vine between 4th and 3rd and catch the 11, 12, 16, or 25, one of which will shortly whisk you across the Taylor Southgate bridge.

Going to Downtown Covington? Walk on over to 4th St between Main and Walnut and you’ll catch whichever of the 1, 3, 5, 7, 12, 25, or 33 that comes by first. They’ll take you over the Clay Wade Bailey bridge into Covington.

In either case it doesn’t matter which you catch and the buses are collectively coming often enough that you don’t need to check a schedule.

Except… did anyone remember the Southbank Shuttle?

The Southbank Shuttle goes back and forth between Covington and Newport through Downtown. It’s main stop, in front of Fountain square on 5th, boards in the same place for both directions of travel. It is a block away from the other two main TANK stops in Downtown: on 5th St between Walnut and Main, and south one block on 4th St.

The Southbank Shuttle sells itself as a quick way to get across the river. Indeed, ‘quick’ is in one important respect a very good word for the line. It’s TANK’s most frequent service, running every 20 minutes in both directions. And it does indeed cross the river as quick as anything.

TANK crosses the ohio river

Line thickness indicates the frequency of the service. The thicker the line, the more frequently transit runs. Where lines travel the same path, you can get a rough guess of total frequency by adding the widths–after all, for the portion of the trip where the lines run parallel, it doesn’t matter which one you catch since they all go to the same place!

But let’s take as an example a trip between Downtown Cincinnati and the Covington transit center to see how the shuttle fares as an option. A quick glance at the frequency map above shows two possible paths from Downtown: the Clay Wade Bailey bridge or the Roebling bridge. Now really, this shouldn’t matter, as long as you make it to Covington but the fact that the stops are different for each path is critically important. There’s no place where they all meet, and you can’t wait at multiple stops. You have to choose one, so which one do you pick? Rationally, you should pick the one where it’s most likely a bus will be coming very soon to take you across the river.

Look again at the map above.

You should wait at the stop on 4th to catch the 1, 3, 5, 7, 12, 25, or 33. Their total frequency combines to something much greater(if surely slightly less regular) than that of the shuttle.

That doesn’t sound so bad–two options, you might say–are better than one, right? Nope. Multiple options for the same trip make each option worse. Where transit lines overlap they complement each other. Their collective frequencies increase, meaning that for trips within the span of redundancy, people spend less time waiting for the next vehicle to arrive and relatively more time actually in motion.

SORTA recognizes this fact with their numbering. The 4, 17, 43, and 11 all branch out once they get a few miles from Downtown into sub-routes identified by a name after the number(’17-Mt Airy’ for example). The number common to what are truly different lines belies the importance of understanding overlapping transit lines as a unified service. You can also see this very clearly in the color scheme of New York City’s subways which come together in Manhattan before branching apart in the other boroughs into lines with both a color and a letter or number identifying them.

New York City Subway Map

One of at least a dozen versions of an iconic map.

The shuttle is effectively duplicating something that already exists: a reasonably quick and easy way to get right across the river. That duplication isn’t at all a bad thing as long as it builds on and amplifies the effect of other services. If the shuttle took the same bridges across the river and stopped at the same stops Downtown, it would likely have little if any immediate impact on the ridership of the shuttle but could increase the effective frequency of short trips across the river to Covington, saving time for all people making that trip.

Why does the shuttle feel the need to be a rebel? It’s clear that TANK is marketing it to a different audience. First of all, it has a different (and somewhat kitschy)vehicle.

southbank shuttle

Punk.

It has a different schedule format. It has a different schedule structure(every 20 minutes throughout the day rather than less regular times and tapering off in the evening). It has it’s own page on the website. It’s even $0.50 cheaper than any other line. And of course it has a different routing across the river and different Downtown stops.

Let’s assume that there are some legitimate psychological/marketing reasons for such differentiation. I’ll assume TANK knows their customers better than I do. Still, that differentiation could exist on top of other transit lines already going to the same places, allowing some people who don’t care about wooden seats and cheaper fare the option of catching a ride across the river on a different line with substantially less time spent waiting.

Overall, TANK’s stops in Downtown make a lot of sense, but the shuttle’s separate stops for the Downtown-Covington segment keep TANK from really making the trip across the river both quicker for short trips, and easier to understand for everyone.

Comments: 6
Posted in: Simplicity | Tips & Tricks
Tags: | | | | |