PAT posts Transit Development Plan (Pittsburgh, Wilkinsburg: transfer, live in, versus)

[BrianTH]

Originally Posted by Tinare

This may be a stupid question, but I don't speak transit and didn't have time to read everything, but does anyone know what "average headway" means? Is that the estimated time to make a roundtrip on the route?

In this case, it is how often buses on that route will come by a given stop. So, for example, if they say for Route X the average headway in the peak AM time would be 15 minutes, then the Route X Bus will come by an average of every 15 minutes during that period.

[JoeLeaphorn]

Originally Posted by BrianTH

It looks like the total all-day ridership from Summerset was something like 12 people. You can't really justify running a bus route for that many people.

It is a classic chicken and egg problem for transportation planners. You need to have the ridership to justify the route, but you need a reliable route and schedule to get the ridership.

The problem that Pittsburgh has always had is the migratory nature of the neighborhoods and sprawl. Sprawl, in particular, has allowed business and employees to decentralize by making it cheaper to locate outside the city limits.

One of the reasons that rail is better than busses and even BRT is the fact that it suggests permanence. People know that if you are going to the trouble of laying rails down then you are serious about a long term future.

[groar]

Originally Posted by BrianTH

I definitely agree Summerset should be supported, but apparently the ridership from Summerset on the 61D was very low. In fact looking at the chart in this link:

http://tdp.portauthority.org/paac/portals/1/pdfs/Eval/61D.pdf (broken link)

It looks like the total all-day ridership from Summerset was something like 12 people. You can't really justify running a bus route for that many people.

i remember reading that when that route was introduced, summerset residents complained. i'm not surprised that the route had low ridership, although i think it's crazy that people living there wouldn't take advantage of a bus going into the squirrel hill business district, oakland & downtown.

[JoeLeaphorn]

Originally Posted by BrianTH

In this case, it is how often buses on that route will come by a given stop. So, for example, if they say for Route X the average headway in the peak AM time would be 15 minutes, then the Route X Bus will come by an average of every 15 minutes during that period.

The problem is that the use of average headway in public transportation planning is based upon the assumption that the person waiting is able to board the very next vehicle that comes by. Thus, the average wait time is 1/2 the average headway plus a coeffcient which represents the variability in headways. If that variability is high, people won't choose public transportation because they become uncertain of the wait time. Most people who use public transportation don't want to memorize a schedule, they just want to know that if they show up at a stop they won't have a long wait.

If the variation is too high, these people won't choose public transportation, even if the average is relatively low.

This becomes especially important if the deviation means that some people will miss their connections some of the time since this can be the deal breaker in terms of choosing the public transportation option.

[BrianTH]

By the way, when thinking about comparable cities for Pittsburgh for transportation purposes, I would suggest it is useful to look at not just total population but also average weighted density. Average weighted density basically tells you how concentrated the population is, and obviously a concentrated population can be more easily served by public transit, and conversely may also lead to more highway congestion--and, in fact, there is a robust statistical correlation between average weighted density of cities and the ratio of people using public transit versus driving for commutes (more on this below).

Anyway, here is a calculation of average weighted densities for the 34 largest urbanized areas using 2000 Census data:

Austin Contrarian: Density calculations for U.S. urbanized areas, weighted by census tract

Even though Pittsburgh had the 22nd largest urbanized area by population, we only had the 29th highest average weighted density. In other words, we are less concentrated than some cities the same size or smaller--which one could probably have guessed, since our topography creates some population concentration challenges.

What this means in turn is that Pittsburgh is actually doing pretty well given both its size and density when it comes to public transit use for commuting. Here is a link from the same source analyzing this issue:

Austin Contrarian: The association between density and mode of commute

If you look specifically at this chart, you can pick out Pittsburgh ("PIT") as being well above the fit line, meaning we have relatively high public transit use for commuting after controlling for weighted density:

http://austinzoning.typepad.com/.a/6...4c1ca8c970c-pi

In fact, although other cities also outperform the fit line, Pittsburgh is the third biggest outperformer, after DC (way out in front) and then Boston.

On the other hand, this analysis found an even better correlation with the ratio of weighted density to average density. It is hard to describe that measure intuitively, but the best I can do is to say it captures how locally peaky the concentration of the population is. And you might well expect this better fit because now you are really talking about better circumstances for public transit (i.e., you can schedule transit routes through these local peaks in concentration).

Anyway, by that measure Pittsburgh actually moves back up the density charts--again, thinking about the topography, you can visualize how although we are a little bit spread out overall, on a local basis we are a bit more peaky, meaning we have little pockets of higher concentration. Consequently, although we still outperform the fit line, we don't stand out quite as much anymore:

http://austinzoning.typepad.com/.a/6...4c1c601970c-pi

Still, we are fifth in outperformance, after a new #1 in SF, then DC a close #2, and then Portland, then Seattle, and finally us.

Anyway, I hope people found that interesting. The bottomline is that I think it is true that our public transit system usage looks pretty good when you control for things like these density measures. However, we can undoubtedly do better, and that is part of what this effort is about (more rationally structuring the transit system in order to serve the population as it is currently distributed).

[BrianTH]

Originally Posted by JoeLeaphorn

It is a classic chicken and egg problem for transportation planners. You need to have the ridership to justify the route, but you need a reliable route and schedule to get the ridership.

The problem that Pittsburgh has always had is the migratory nature of the neighborhoods and sprawl. Sprawl, in particular, has allowed business and employees to decentralize by making it cheaper to locate outside the city limits.

One of the reasons that rail is better than busses and even BRT is the fact that it suggests permanence. People know that if you are going to the trouble of laying rails down then you are serious about a long term future.

I'm not sure much of that applies to Summerset specifically, and there is no way we are actually going to be building a rail line there in the conceivable future. I suspect the problem with Summerset is just that it is actually too convenient for driving to Downtown and Oakland, and while it might work as an area you passed through with a route to pick up a few additional riders, it won't work as its own terminal route.

By the way, BRT in the form of actual dedicated busways, like the East Busway (soon to be Purple Line), captures much of that same sense of permanence. Again, though, Summerset is not conceivably going to be on a busway either.

[BrianTH]

Originally Posted by groar

i remember reading that when that route was introduced, summerset residents complained. i'm not surprised that the route had low ridership, although i think it's crazy that people living there wouldn't take advantage of a bus going into the squirrel hill business district, oakland & downtown.

I think the basic problem is that most of those people have money, can afford cars and parking, and are so conveniently located for driving they just don't take the bus much.

[JoeLeaphorn]

Originally Posted by BrianTH

\
Even though Pittsburgh had the 22nd largest urbanized area by population, we only had the 29th highest average weighted density. In other words, we are less concentrated than some cities the same size or smaller--which one could probably have guessed, since our topography creates some population concentration challenges.

What this means in turn is that Pittsburgh is actually doing pretty well given both its size and density when it comes to public transit use for commuting.

Be careful how you interpret this. We created the problem, first, by making the Interstates run through the city, rather than around, and we're adding to it by trying to do such foolish things as the Mon-Fayette Expressway. We made it cheap for people to move out to areas where there was little or no public transportation. Congratulations!

Statistics only go so far. The big statistic is the number of vehicles/family. That tells you how good public transportation is doing.

[JoeLeaphorn]

Originally Posted by BrianTH

I'm not sure much of that applies to Summerset specifically, and there is no way we are actually going to be building a rail line there in the conceivable future. I suspect the problem with Summerset is just that it is actually too convenient for driving to Downtown and Oakland, and while it might work as an area you passed through with a route to pick up a few additional riders, it won't work as its own terminal route.

Summerset as a intermediate node in a commuter line from the East to town makes sense and could be done.

Originally Posted by BrianTH

By the way, BRT in the form of actual dedicated busways, like the East Busway (soon to be Purple Line), captures much of that same sense of permanence. Again, though, Summerset is not conceivably going to be on a busway either.

But there is more, including acceleration and deceleration times and in terms of the stimulation of transit oriented development. Also, rails attract investment. There are no good examples of where BRT can be shown to have single-handedly stimulated adjacent real estate development. That is not true for rail. The average lifespan of a bus is about 15-20 years. Railcars have lasted decades more.

[BrianTH]

Originally Posted by JoeLeaphorn

Be careful how you interpret this. We created the problem, first, by making the Interstates run through the city, rather than around, and we're adding to it by trying to do such foolish things as the Mon-Fayette Expressway. We made it cheap for people to move out to areas where there was little or no public transportation. Congratulations!

Yep, although of course this was a nationwide problem. Fortunately, I think the tide has turned--for example, I think the idea of actually completing the Mon-Fayette Expressway on the Pittsburgh side is pretty much dead, as it should be.

Statistics only go so far. The big statistic is the number of vehicles/family. That tells you how good public transportation is doing.

I'm not sure why that is better than other statistics that more directly measure public transit shares, but anyway it looks like Pittsburgh is very low by that measure:

Motor Vehicle Availability by US Metropolitan Area

That isn't organized ideally, but I think NYC was last at #285, and Pittsburgh was #282 (Brownsville and New Orleans were in between).

Another interesting statistic for measuring car usage is VMT per urban square mile, and by that measure we are in fact last among the big urban areas:

USA Urban Areas: Per Capita Travel & Traffic Intensity: 2005

As charted there we are also on the low side for VMT per capita.