Sunday, November 12, 2017

Anchorage Bus Restructuring (And Why Staten Islanders Should Care)

So a restructuring of the Anchorage bus system went into effect on October 23rd, 2017.

Why Should Staten Islanders Care?

Basically, they consolidated the routes with the idea of providing 15 minute headways on their busiest routes (similar to other cities such as Houston that had a major restructuring of their bus system). In my opinion, they went too far in cutting service to the lower-ridership areas.

In a report they released, they seemed to be torn between 2 options, an 80% plan and a 100% plan (referring to the percentage of resources put into higher ridership routes). The majority of people seemed to be in favor of the 80% plan, but it looks like they bumped the percentage up to 90% despite public opinion (I saw a map that said something like "90%" in he URL). The options start on page 65.

The general trend nationwide lately has been for agencies to redesign their networks to offer more frequent service on the busiest routes. They've done it in Cincinnati, Houston, and Omaha just to name a few cities. In some cases, they've gotten additional funding (I have to double-check, but I believe Cincinnati & Omaha gave their agencies more money to run an expanded system. I know Houston didn't, and Anchorage didn't). I haven't analyzed Houston's system (it's a lot larger than Anchorage's system), but from what I understand, they redirected a lot of their bus routes to feed into their light rail system, so they were able to put more resources into the frequent routes without reducing coverage). For those of you who don't know, the MTA tries to be cost-neutral in its service proposals, and from what the project manager said, this is no exception.

The express system restructuring proposed on Staten Island does follow the general trend of consolidating the routes into a frequent network. They didn't really reduce the coverage of the system, because on the Staten Island end, pretty much all the areas that had express service before still have it, and on the Manhattan end, there are technically still local buses and subway routes that cover the areas that had express bus service taken away from them, even in cases like the West Street corridor where the routes are significantly longer, which will hopefully be remedied thanks to public outcry). However, this does get me thinking about what this means for the local network. After all this fighting I have done to make sure isolated areas like Grymes Hill & Cottage Hill get their weekend service back, I don't want to see it brushed away under the guise of consolidation into a frequent network.

The thing is that Staten Island routes are still busier than routes in most of these cities. The busiest route in Anchorage (the old #45) carried around 2,600 riders a day. All of the full-time routes out of St. George carry more than that. So all of the areas that would be considered "coverage" areas on Staten Island would be considered high-ridership areas in other cities. Additionally, Staten Island isn't a city of its own, but rather a borough of another city, and on top of that, we still have the crutch of infrequent (and slow, but that's another topic) ferry service. So the most frequent network in the world won't help riders using the ferry to get to and from Manhattan. Additionally, a lot of service is consolidated onto major streets already, due to the topography and development patterns on the island.

Long story short, one of the major issues on Staten Island is the lack of coverage in areas throughout the island, and I hope they aren't blindly following the lead of other, smaller cities throughout the country. Hopefully this is just a rant about nothing and the local study gives us a comprehensive and usable system that we deserve.

Now, for those of you who are interested, here's my analysis of the Anchorage plan.....

What Specifically Is Wrong With the Anchorage Plan?

I'm going to start by saying that one of the things I don't like about the plan is that the people who designed it violated some of their own principles in the design of this network. The general idea of the plan was to consolidate the resources towards the busiest routes, to provide "Less Waiting" and "More Hours". Alright, that's a logical principle to follow. So then how come on Saturdays, they essentially cut the last round-trip of the #10 (old #3), #20 (old #45), #40 (old #7), and last outbound trip of the #30 (old #15)? I understand that they added some service in the morning, but I would've much rather they add less morning service in exchange for keeping the system end time the same. (They could've had the buses run every 60 minutes from 8-9AM or 8-10AM like they had before, and used the extra hours to keep the last trips the same, or even add a trip an hour beyond the original span)

In terms of the corridors/neighborhoods that should've had some type of service retained, these are the major ones
* Old Seward Highway
* 36th Avenue
* Turnagain
* South Anchorage (more than just rush hour service)
* Eagle River

In terms of routes, I'm going to start by saying that the worst decision was to eliminate the #60 bus north of Old Seward Highway. That route is a mile away from the #35 on Artic and a mile from the #55 on Lake Otis Parkway....assuming there's a direct road between them and you don't have to first detour to that road (e.g. Dowling Road). According to page 38 of the report, ridership the #9 & #60 between the Dimond Center & Downtown was very similar (as a matter of fact, the #9 had 30 minute service added in July 2008 (page 25), so had it been the #60 that saw the increased service, it would likely seen the increased ridership, and it might be the #9 that was cut.

Since the whole selling point of the new system is better frequency, the new system really limits your options between Downtown & Dimond Center (two major hubs), by leaving only one direct route (the #9). Granted, the #2 & #7 (especially the 7A through the airport) had much longer travel times, but they were options nonetheless. The #60 was only a couple of minutes slower than the #9 (which means that with a slightly slower driver, or some slight delay such as a wheelchair passenger, it's pretty much a wash between the two routes).

On the #92, the reason why they run 2 minibuses per trip as opposed to one is because people were worried about being left behind. Instead of spreading them out every half hour (which would boost ridership), they did the bare minimum and essentially cut one trip in each direction from the old schedule (the 4:40PM and 6:22AM trips).  Also, they sneakily cut the Eagle River Connect service, which means midday passengers to/from Eagle River are essentially stranded.

There's one final issue with this plan is that, according to pages 3-4 of the Ride Guide (7-8 of the PDF), there's no free transfers between routes. You can purchase a day pass for round-trip travel, but that discourages single trips (e.g. If somebody got a lift from somebody to work, but needs to use the bus system to get back home, that $4 fare will be a turn-off for that one-time trip, and will be encouragement for them to use some alternative route). Though I will say, I do like the fact that they offer an annual pass on the routes.

The Good Aspects of the Plan

The system is aimed towards concentrating service in the denser portions of the municipality. The busiest route in the old system (the #45) had issues with overcrowding and leaving people behind... which is especially unacceptable on one hour headways . With four full-sized buses an hour for most of the day on the #20, and two minibuses per hour on the #21 supplementing it, there's definitely more capacity available for bus riders. (On top of that, the #21 & #31 provide a quick link from the Northway Mall & Penland Park towards Downtown, as opposed to the old system where the #8, #15, and #45 all meandered their way towards Downtown). It looks like they made some type of attempt to coordinate the two routes to provide approximate 15 minute headways to/from Downtown (though it gets a little messed up due to the #21 having uneven headways, likely due to interlining it with another route). Between UAA/Providence Hospital and Downtown, riders have frequent, direct service on both the #10 & #20, as opposed to having some of that frequency going towards meandering routes like the #13 & #36 under the old system.

The areas with the highest percentage of autoless households (the "bread and butter" riders of the transit system), include Downtown, Mountain View, Fairview, Penland Park, Government Hill, Muldoon, and to a lesser extent, Midtown and Spenard. There's development plans for the Northern Lights & Spenard Road corridors, which would be supported by frequent transit. Those areas generally see better service under the new plan compared to the old system (with the major exception being in the Taku-Campbell neighborhood, which lost the #60). You could argue that consolidating the service from 36th Avenue in Midtown onto Tudor Road & Northern Lights also represented a service cut for a decent amount of riders.

Additionally, routes are generally straighter compared to the way they used to run, as they don't meander down neighborhood streets as much (this can especially be seen on the old #45 vs. the #20 in the Russian Jack Park area, and the overall routing of the old #8 vs. the #31).

How I Would Improve The Plan

So at the very least, north of the Dimond Center, I would keep the hourly service on the #60 (or some equivalent route covering the northern part of Old Seward, since they renumbered the #9 to #35 to fit the scheme of having the standard routes end with "5"). Perhaps some savings can be realized through interlining it with the #35 at both terminals).

I would also keep at least hourly service on 36th Avenue, by way of extending the #55 to Downtown Anchorage via the following routing: Turn east on Tudor Road, serve the Alaska Native Medical Center & Providence Hospital/UAA, then continue on 36th Avenue to Spenard Road, and follow the #40 route to Downtown.

For service to Turnagain Heights, I would extend the #65 straight up Wisconsin Street to Northern Lights Blvd, and then take the #40 route to Downtown. Also, I would have the #65 take the old 7J detour up Raspberry Road, as it serves more people than staying on Sand Lake Road, without adding much travel time (it also helps some of the people who live closer to Northwood Street).

On the #92, I don't mind truncating the route to Downtown (as the #10 & #20 provide frequent service to the UAA/Providence Hospital, and the #20 & #25 provide frequent service to the ANMC). I do like the fact that there's one reverse-peak trip out to Eagle River in the morning, and returning from Eagle River in the evening (though to be fair, I remember seeing an old schedule of the #102 that said that passengers were permitted to stay on after the last stop to return to the Muldoon Transit Center). At the very least, I would at least keep that feature, and allow passengers to stay on until the Muldoon Transit Center as the bus deadheads back to the garage.

For South Anchorage, I would at the very least have the #91 run more than just 2 trips each way (though I will be fair, and say that those trips do seem perfectly placed. You have one that can get you into Downtown around 8am, and one around 9am, with return trips leaving around 5pm and 6pm. Not sure why the bus only run every 60 minutes, as opposed to every 30 minutes, as there is a 33 minute layover at the Dimond Center, while a round-trip only takes 27 minutes. Or they could have it run up to Tudor Road to at least cover the isolated part of Old Seward and give passengers a connection to the #25 and the destinations on that route in both directions). Come to think about it, while it would be a bit awkward, a shuttle from Old Seward & Tudor to Old Seward & Oceanview would be a bit awkward-looking (with Dimond Center in the middle of the route), but it would still provide the necessary coverage, and be enough to replace the coverage lost by the old #60. Such a route should run 7 days a week, similar to the other shuttle routes (#11, #21, and #31)

I understand People Mover has a budget to balance, and if the plan must be cost-neutral, then there should be service reduced somewhere else. The Debarr Road & Spenard Road corridors weren't in the original frequent network plan. Spenard Road has a transit-oriented development plan that's being pushed, so I think that (along with the fact that the route serves the airport) is a main reason why it's on the frequent network. In order to obtain the funds necessary to keep Old Seward Highway, 36th Avenue, and Turnagain Heights covered, I would look into reducing the frequencies on the #10, #30 & #40 routes (the old #45, which became the #20 had twice the ridership of the old #3, so I would leave it as-is). With the #55 & #65 routed onto Spenard Road, crowding heading into Downtown shouldn't be an issue if the #40 were reduced to every 30 minutes (especially since the #65 serves Spenard proper as well), especially if the trips are coordinated with the #40 so that the northern portion of Spenard Road sees 15 minute headways. The #10 and #30 I would reduce to every 20 minutes (the #10 is busier, but the #30 is a quick route from end to end). Keep in mind that under the old system, all of those routes ran every 30 minutes. The #55 extension would require 1 extra bus for hourly service (and that may be cutting it a bit close during rush hour) and the #65 extension would require 1 extra bus (but there would be plenty of slack in the schedule in terms of layover). In other words, if you interline the #55 & #65, you would be able to cover both the 36th Avenue corridor and Turnagain Heights with 2 extra buses. It looks like the #40 requires 4 buses for 15 minute headways (though it looks like they interlined it with another route to give themselves some more cushioning on the layover times), so between that, and possibly a slight reduction at certain times to either the #10 or #30, that should be enough to cover those 2 areas. Keep in mind, I wouldn't touch the late-evening service to the airport on the #40. 

....And With More Money....

I remember reading an old study that suggested extending some #45G short-turns to Tikahtnu Commons (or at least the VA Hospital). Since the #21 is basically the same thing as the old 45G (except it runs minibuses), perhaps it can run via Bragaw-Parsons-Lane-Mountain View Drive, then get on Glenn Highway at Boniface Parkway, and get off at Muldoon Road. This would provide a quick way to get there from Downtown (right now, the #25 takes the long way around. You can transfer to the #30 or #31, but the transfer takes time to make. Also, it's very circuitous to reach Mountain View itself (you have to either take the #20 down to the #25, or do a lot of walking to/from the #31). The only issue is, budget-wise, you might need more than just minibuses to handle the passenger loads. The runtime would be about 25 minutes, which would allow for a 5 minute layover and 30 minute headways if 2 buses were used.

One idea I had was to run this route out to Eagle River (with a full-sized bus) to replace the #92 (ideally, with 7 day service). Service within the Anchorage Bowl (meaning, west of the VA Hospital) would run every 30 minutes all day, every day, and service to Eagle River would be every 30 minutes during rush hour, and every 60 minutes other times. That way, it provides Eagle River riders with a full-time connection to the #25 to get to East Anchorage, as well as a one-seat ride to Downtown Anchorage. (Keep in mind that "all-day, every day" refers to the general span of the system. The Anchorage system starts shutting down around 11pm on weekdays, and 6-7pm on the weekends. I would run it longer on the weekends, but not 24/7 yet.

Another idea I had was to split up the #10 route. I would have alternate buses run straight down Northern Lights Blvd without serving Providence Hospital/UAA, and alternate buses terminating there, from both directions. In addition, I would have the #55 run every 30 minutes via the route I described above, and coordinate it with the #10 to provide 15 minute crosstown headways towards Midtown & Downtown. (On top of that, there's still the #20 if your destination is Downtown, and the #20 is actually a bit quicker than the #10). For those in a hurry heading towards the Muldoon Transit Center, they can take the #20 to Northern Lights & Bragaw and catch any eastbound bus (one originating from Downtown, or one originating from Providence Hospital), and the same thing heading westbound in reverse. Granted, the #30 already provides the most direct route from Downtown to Northeast Anchorage and the Muldoon Transit Center, but this would provide a quicker option for those further south (also, the #30 doesn't serve Midtown like the #10 does). Not sure how the routes should be numbered, since they would be a combination of 30 minute headway routes (that should end with "5") that combine to offer 15 minute headways along their common trunk (which means the number should end in "0".

Additionally, I would also coordinate the #35 and #60 (or whatever that route is renamed to) to provide approximate 15 minute headways between the Dimond Center and Downtown. From the Dimond Center, I would have a circulator that covers both the old #60 route along John Road, and the old #2 route along Independence Drive). Service could depart the Dimond Center every 45 minutes, perhaps alternating between running clockwise & counterclockwise)

Additionally, since the #11 serves the dense Government Hill & Fairview areas, I would bump up the frequency to every 30 minutes. Fairview has the frequent #30 bus within walking distance of most able-bodied riders (either that, or at least the #31 for those in the northern portion of the neighborhood), but the #11 provides the only form of transit for Government Hill residents. Also, if the #11 were rerouted to run straight across 9th Avenue, it would result in a 15 minute combined headway for those living in the northern portion of the neighborhood looking to head Downtown (whether as a destination in and of itself, or to connect to another route). 9th Avenue is a 0.4 mile walk from 15th Avenue, and so somebody halfway down the middle would only have a 0.2 mile walk to get to a frequent route (and since the #15 runs down Cordova, people in the western section would have the option of catching the bus there). And frequency is especially good at attracting ridership if the trip is short (if somebody has to wait 30 minutes for a bus, to make a 1 mile trip that can be made in 20 minutes by walking, they'll likely just walk. If the bus runs every 15 minutes, they'll at least think twice). 

One final thing for People Mover to consider is that, since their depot is located around Tudor & Elmore (a few blocks south of the #20 terminal at the ANMC), I think there should be a few more trips that do a short-turn trip to/from their first/last stop (so some #10 buses that only run between Providence & Elmore and either the Muldoon Transit Center or Downtown, some #25 buses that run between the ANMC and either the Muldoon Transit Center or Downtown, and try to have more of the #20 trips run on and run off from the ANMC end). Additionally, some #15 buses can start/end at Alaska Regional Hospital, and deadhead via Lake Otis Parkway. The old schedule actually had a few #75 trips that did that (one of them only went to the Muldoon Transit Center, I guess to do a trip on another route). I'm for minimizing deadhead mileage wherever possible (whether it's in Anchorage, Staten Island or somewhere else in the world)

No comments:

Post a Comment