Reasonable Transit Pitches

[bigeman312]

One of the biggest bang-for-buck projects in terms of concrete has to be Newton Corner and a second West Newton platform. Just building the stations and running diesel trains every 15 minutes at peak could kill the 501/504 (which pretty much overlap the 57, although the 501 works in a strange way) and cut back the 505 to West Newton. Even under BNRD 501, 504, and 505 are planned to be operated every 15 minutes at peak.

I'd also add OLX to Roslindale to the list. The Roslindale Village-Forest Hills segment is overserved, and not having to run all the bus routes on the southwest side could provide room for network rationalization.

View attachment 33771

I 100% agree with this and want to add onto this comment. OLX to Roslindale Village is such a great project proposal.

Here, I'm going to add on to this pitch with a pitch for an infill stop between Roslindale Village and Forest Hills.

Roslindale Village is 1.4 track miles past Forest Hills, which would make it the second longest segment between stops on the Orange Line (behind only the Wellington - Malden Center stretch that has regular calls for an infill stop). The area between Forest Hills and Roslindale Village is densely populated (except for the Arboretum, of course). In fact the census tract that spans the sub-neighborhood between Forest Hills and Roslindale Village has a population density of 18.7k per square mile. This is more densely populated than the census tracts for all Orange Line stops north of North Station (as of 2020).

This densely populated sub-neighborhood would still necessitate bus service in the case of an OL+1 to Roslindale Village, given the relatively long gap between Roslindale Village and Forest Hills. Maybe that just looks like the 32, 34, 35, and 36, for example. Still a big win to be able to cut the 30, 34E, 40, 50, and 51 back to Roslindale Village, for example, but without much ability to keep trimming. By adding an infill stop at South Street (I'll name this station 'Aroretum' here), you'd have more flexibility to free up more bus resources. Arboretum Station would be 0.7 track miles from each of Roslindale Village and Forest Hills, which is still appropriately longer than most Orange Line stop segments, including every one from Green Street through Community College.

Arboretum Station could also serve as a much more effective access point to the Arboretum than Forest Hills Station.

 

[F-Line to Dudley]

I 100% agree with this and want to add onto this comment. OLX to Roslindale Village is such a great project proposal.

Here, I'm going to add on to this pitch with a pitch for an infill stop between Roslindale Village and Forest Hills.

Roslindale Village is 1.4 track miles past Forest Hills, which would make it the second longest segment between stops on the Orange Line (behind only the Wellington - Malden Center stretch that has regular calls for an infill stop). The area between Forest Hills and Roslindale Village is densely populated (except for the Arboretum, of course). In fact the census tract that spans the sub-neighborhood between Forest Hills and Roslindale Village has a population density of 18.7k per square mile. This is more densely populated than the census tracts for all Orange Line stops north of North Station (as of 2020).

This densely populated sub-neighborhood would still necessitate bus service in the case of an OL+1 to Roslindael Village, given the relatively long gap between Roslindale Village and Forest Hills. Maybe that just looks like the 32, 34, 35, and 36, for example. Still a big win to be able to cut the 30, 34E, 40, 50, and 51 back to Roslindale Village, for example, but without much ability to keep trimming. By adding an infill stop at South Street (I'll name this station 'Aroretum' here), you'd have more flexibility to free up more bus resources. Arboretum Station would be 0.7 track miles from each of Roslindale Village and Forest Hills, which is still appropriately longer than most Orange Line stop segments, including every one from Green Street through Community College.

Arboretum Station could also serve as a much more effective access point to the Arboretum than Forest Hills Station.

The area south has its walkshed SEVERELY constrained by the Northeast Corridor cut and lack of any spanning streets south of Ukraine Way. It's a 1-block radius around Washington when the NEC and Arboretum are factored, and that's it. Don't go by the census tract; go by the walkshed or lackthereof, which is how things like transit stops get their access heft measured. Since there will still be some buses plying the Washington corridor after the Rozzie cull-a-thon, the out-of-walkshed 'tweeners are still very well transit-served all being no more than 3 blocks from a bus stop at any given time.

Secondly...the Needham Line is perched high on an embankment through here. It's not going to be easy to construct a station around the midpoint without needing to pour fill over a part of the Arboretum and somehow get a permit for that. There was never historically any spanning stops there during 'peak' RR, and this is why.

Finally...the Arboretum path dumps out right bloody across the street from FH . How much more convenient can you possibly get for access???

 

[The EGE]

Or why can't the signal crossing technology simply give priority of a F-ing bus coming out of the tunnel. That drove me crazy for the three years that I commuted down there.

Unfortunately, it's not so simple as "just turn the light green when a bus approaches." Signal priority and preemption aren't magic bullets; they have to work within the engineering constraints of the modes.

For example, you can't cut off the walk cycle arbitrarily - pedestrians have to have time to finish crossing. The crosswalks crossing the SL busway are about 30 feet long; that's a 10-second countdown plus a 3-second yellow plus an all-red clearance. Thus, you have to have at least 15 seconds before the light needs to be green - which is at the decision point where the operator decides to brake or continue through the intersection, several seconds before they reach the intersection. If buses are going faster than 10-15 mph, there isn't enough time after they leave WTC or SL Way to guarantee a green light. You might be able to tighten up the timing somewhat, but it will always be very close.

Also, even if you don't care whatsoever about auto traffic on D Street, you don't want to make pedestrians wait too long (especially since visibility from both directions on the SL isn't great, should those pedestrians cross against the light). Additionally, if you back up D Street enough, you will start impacting SL buses on Congress, or even the 7 on Summer.

This isn't to say there's not an opportunity for more aggressive TSP at D Street, but don't expect it to fix everything. There are some problems that only concrete will fully solve, and this is one of them.

 

[Riverside]

I 100% agree with this and want to add onto this comment. OLX to Roslindale Village is such a great project proposal.

Here, I'm going to add on to this pitch with a pitch for an infill stop between Roslindale Village and Forest Hills.

Roslindale Village is 1.4 track miles past Forest Hills, which would make it the second longest segment between stops on the Orange Line (behind only the Wellington - Malden Center stretch that has regular calls for an infill stop). The area between Forest Hills and Roslindale Village is densely populated (except for the Arboretum, of course). In fact the census tract that spans the sub-neighborhood between Forest Hills and Roslindale Village has a population density of 18.7k per square mile. This is more densely populated than the census tracts for all Orange Line stops north of North Station (as of 2020).

This densely populated sub-neighborhood would still necessitate bus service in the case of an OL+1 to Roslindale Village, given the relatively long gap between Roslindale Village and Forest Hills. Maybe that just looks like the 32, 34, 35, and 36, for example. Still a big win to be able to cut the 30, 34E, 40, 50, and 51 back to Roslindale Village, for example, but without much ability to keep trimming. By adding an infill stop at South Street (I'll name this station 'Aroretum' here), you'd have more flexibility to free up more bus resources. Arboretum Station would be 0.7 track miles from each of Roslindale Village and Forest Hills, which is still appropriately longer than most Orange Line stop segments, including every one from Green Street through Community College.

Arboretum Station could also serve as a much more effective access point to the Arboretum than Forest Hills Station.

I think F-Line raises reasonable critiques but I think this is a good example overall of a Reasonable Transit Pitch — builds on an existing well-known proposal, lays out the case for a modest addition, doesn’t blow up the bank on sight. (Even if further analysis does reveal higher costs.)

Unfortunately, it's not so simple as "just turn the light green when a bus approaches." Signal priority and preemption aren't magic bullets; they have to work within the engineering constraints of the modes.

For example, you can't cut off the walk cycle arbitrarily - pedestrians have to have time to finish crossing. The crosswalks crossing the SL busway are about 30 feet long; that's a 10-second countdown plus a 3-second yellow plus an all-red clearance. Thus, you have to have at least 15 seconds before the light needs to be green - which is at the decision point where the operator decides to brake or continue through the intersection, several seconds before they reach the intersection. If buses are going faster than 10-15 mph, there isn't enough time after they leave WTC or SL Way to guarantee a green light. You might be able to tighten up the timing somewhat, but it will always be very close.

Also, even if you don't care whatsoever about auto traffic on D Street, you don't want to make pedestrians wait too long (especially since visibility from both directions on the SL isn't great, should those pedestrians cross against the light). Additionally, if you back up D Street enough, you will start impacting SL buses on Congress, or even the 7 on Summer.

This isn't to say there's not an opportunity for more aggressive TSP at D Street, but don't expect it to fix everything. There are some problems that only concrete will fully solve, and this is one of them.

You’ve done a good job educating us on this topic recently — I saw @dhawkins’s post, and my first thought was about the pedestrian signal cycle. Appreciate you adding the further detail beyond that!

 

[737900er]

If memory serves, I think one of the old Blue Books listed the number of vehicles per route — could be a useful reference for quantifying this.

Yes, they did used to publish that data back with the PDF Blue Book, but seems to have been stopped as they transitioned to the new reporting system. Given that data is now nearly a decade out of date, it's a bit difficult to use, especially with all the post-COVID network changes at the outer bus hubs.

That analysis certainly could be done, but it's kind of a PITA given that a lot of the benefit of the rail extensions is being able to cut buses short. \ Some buses would get zapped entirely with rail replacement -- 501, SL2/SL4/SL5 -- but most, like 35 or 450, would get shortened. Going into all the bus schedules to allocate the savings would be tedious, and some of the bus schedules don't exist anymore.

Instead, I decided to look at eliminated VMT because I am lazy and I had the data readily available. This means that I only counted miles east of Lynn on the 450 or miles on the Pike for the 505, for example. I also included a handful of other high-frequency routes as a benchmark. The miles are per direction as of 2019.

Now, in most of these scenarios, some amount of surface buses would remain, like we saw with GLX/BNRD in Somerville, but I didn't account for that.

VMT isn't exactly a great proxy because these routes operate at very different speeds, but it's easy to calculate.

It's also worth reconsidering in the context of the original question; LRTing the SL4/5 won't generate that much labor savings (this is also why I didn't include the SL2); it's mainly replacing bus operators with LRT operators. The heavy rail extensions probably generate better labor savings.

All that said, the more I look at OLX the more I like it. I'd be fascinated to see the T's internal analysis of ROI -- the labor savings make it look really appealing and like a good way of turning capital funds into reduced operational expenses that actually benefits riders.

BLX doesn't rate well partially because of the poor levels of service in the Lynn area to begin with. OLX is more about being able to allocate bus resources to other parts of the network, whereas BLX is more about being able to increase bus service into the rapid transit network. The Newton routes should also rate pretty well because of the low marginal cost beyond other planned improvements on the corridor.

 

[Teban54]

Yes, they did used to publish that data back with the PDF Blue Book, but seems to have been stopped as they transitioned to the new reporting system. Given that data is now nearly a decade out of date, it's a bit difficult to use, especially with all the post-COVID network changes at the outer bus hubs.

That analysis certainly could be done, but it's kind of a PITA given that a lot of the benefit of the rail extensions is being able to cut buses short. \ Some buses would get zapped entirely with rail replacement -- 501, SL2/SL4/SL5 -- but most, like 35 or 450, would get shortened. Going into all the bus schedules to allocate the savings would be tedious, and some of the bus schedules don't exist anymore.

Instead, I decided to look at eliminated VMT because I am lazy and I had the data readily available. This means that I only counted miles east of Lynn on the 450 or miles on the Pike for the 505, for example. I also included a handful of other high-frequency routes as a benchmark. The miles are per direction as of 2019.

View attachment 33793

Now, in most of these scenarios, some amount of surface buses would remain, like we saw with GLX/BNRD in Somerville, but I didn't account for that.

VMT isn't exactly a great proxy because these routes operate at very different speeds, but it's easy to calculate.

It's also worth reconsidering in the context of the original question; LRTing the SL4/5 won't generate that much labor savings (this is also why I didn't include the SL2); it's mainly replacing bus operators with LRT operators. The heavy rail extensions probably generate better labor savings.

All that said, the more I look at OLX the more I like it. I'd be fascinated to see the T's internal analysis of ROI -- the labor savings make it look really appealing and like a good way of turning capital funds into reduced operational expenses that actually benefits riders.

BLX doesn't rate well partially because of the poor levels of service in the Lynn area to begin with. OLX is more about being able to allocate bus resources to other parts of the network, whereas BLX is more about being able to increase bus service into the rapid transit network. The Newton routes should also rate pretty well because of the low marginal cost beyond other planned improvements on the corridor.

An added benefit to OLX to Roslindale is that it likely doesn't require new rolling stock (thus also no or minimal new train operators and dispatchers), whereas BLX does.

While Newton Corner station also doesn't incur additional operating costs, I wonder if it's engineeringly feasible for all trains (including express trains) to stop here using diesels. Even if it is, technically that still doesn't achieve 15-min frequencies during peak with the current schedule like buses do, so you may still need to run express buses at some (reduced) capacity.

 

[dhawkins]

Unfortunately, it's not so simple as "just turn the light green when a bus approaches." Signal priority and preemption aren't magic bullets; they have to work within the engineering constraints of the modes.

For example, you can't cut off the walk cycle arbitrarily - pedestrians have to have time to finish crossing. The crosswalks crossing the SL busway are about 30 feet long; that's a 10-second countdown plus a 3-second yellow plus an all-red clearance. Thus, you have to have at least 15 seconds before the light needs to be green - which is at the decision point where the operator decides to brake or continue through the intersection, several seconds before they reach the intersection. If buses are going faster than 10-15 mph, there isn't enough time after they leave WTC or SL Way to guarantee a green light. You might be able to tighten up the timing somewhat, but it will always be very close.

Also, even if you don't care whatsoever about auto traffic on D Street, you don't want to make pedestrians wait too long (especially since visibility from both directions on the SL isn't great, should those pedestrians cross against the light). Additionally, if you back up D Street enough, you will start impacting SL buses on Congress, or even the 7 on Summer.

This isn't to say there's not an opportunity for more aggressive TSP at D Street, but don't expect it to fix everything. There are some problems that only concrete will fully solve, and this is one of them.

Understood; but there have been studies that point to transit priority as an easy fix. Here is a 2017 study by Northeastern with some insight of a simpler system. Simple and Strong” Transit Signal Priority for the Silver Line at D Street I always concluded seaport planning probably falls under state / Mass DoT rules for traffic that favors automobiles priority rather than urban transit. (Commonwealth Flats is State land??) I wish the fix were as simple as a students thinking. PS Reading thru the study, I never knew I was waiting 100 sec at that intersection, it just felt like forever. Also, its my understanding that the silver line can't go under D street because there is a massive storm drain /outfall tunnel that basically drains the length of D street..

 

[Teban54]

In addition to whether Silver Line can go under D St, another question is whether it should.

In the BNRD, SL1 and SL3 are rerouted to turn at D St to/from the highway ramps, skipping Silver Line Way station, in order to speed up the trip. Even if T under D is built, making the detour via SLW may still end up being slower than the proposed change, especially with some sort of signal priority (though it may not need to guarantee green light for every bus).

 

[Aprehensive_Words]

The area south has its walkshed SEVERELY constrained by the Northeast Corridor cut and lack of any spanning streets south of Ukraine Way. It's a 1-block radius around Washington when the NEC and Arboretum are factored, and that's it. Don't go by the census tract; go by the walkshed or lackthereof, which is how things like transit stops get their access heft measured. Since there will still be some buses plying the Washington corridor after the Rozzie cull-a-thon, the out-of-walkshed 'tweeners are still very well transit-served all being no more than 3 blocks from a bus stop at any given time.

Secondly...the Needham Line is perched high on an embankment through here. It's not going to be easy to construct a station around the midpoint without needing to pour fill over a part of the Arboretum and somehow get a permit for that. There was never historically any spanning stops there during 'peak' RR, and this is why.

Finally...the Arboretum path dumps out right bloody across the street from FH . How much more convenient can you possibly get for access???

All this said, it appeals to my love of dark humor that this station would be at the site of one of Greater Boston's worst rail disasters: https://www.structuremag.org/?p=17909

Reason enough to build the station, no?

 

[Riverside]

In addition to whether Silver Line can go under D St, another question is whether it should.

In the BNRD, SL1 and SL3 are rerouted to turn at D St to/from the highway ramps, skipping Silver Line Way station, in order to speed up the trip. Even if T under D is built, making the detour via SLW may still end up being slower than the proposed change, especially with some sort of signal priority (though it may not need to guarantee green light for every bus).

I believe this is a quiet consequence of the shift toward BEBs -- no longer necessary to stop at SLW to switch from wire to diesel. (Why the switch couldn't have been done inside WTC this whole time seems like a related but now moot question.) But yes -- as far as I can tell, entering/exiting the transitway via D St is by far the fastest way to journey between the transitway and the TWT. The mess of ramps to/from the Pike means that every potential path is at least somewhat roundabout, but I believe that's the most direct route.

That being said, if/when the transitway is converted to LRT, then I think T-under-D absolutely makes sense, either extending to a terminus at Silver Line Way (which I suggest renaming Harborview after the very short Harborview Lane), or with continued street-running out toward the eastern edge of the Seaport.

 

[Wash]

Doesn't Montreal have some spare 25kV EMUs, or have those been scrapped?

They did - but their top speed was only 90 MPH and they only ever operated at a top speed of 75 mph (?) in service, so they wouldn't have done much good for the T in any case.

 

[Tallguy]

They did - but their top speed was only 90 MPH and they only ever operated at a top speed of 75 mph (?) in service, so they wouldn't have done much good for the T in any case.

They would be good for inside of 128

 

[Tallguy]

They would be good for inside of 128

I could see them on inner Fitchburg while platforms get installed.
Start them on Fairmount first until EMUs are available

 

[F-Line to Dudley]

They did - but their top speed was only 90 MPH and they only ever operated at a top speed of 75 mph (?) in service, so they wouldn't have done much good for the T in any case.

No...design speed was 75, top operating speed was 68. They're too slow for anything but Fairmount.

Exo hired a consultant to try to find a new buyer for the fleet. They found no takers after 2 years of trying, and it was the large cost of giving them a midlife overhaul that kept interested parties away. They were bid out for scrap last summer. On the slim odds they haven't been physically scrapped yet, they are very likely now owned by the scrap bidder awaiting disposition. Phoning up Exo tomorrow morning for a chat probably isn't going to do any good.

 

[Wash]

I've been looking a fair bit at how overnight trains work in Europe. Frequently, trains are split and re-combined multiple times en route to serve multiple points of origin and destinations with one train. I've been brainstorming a bit about how to improve night trains on the NEC, and here are a few ideas:

1): Currently, Virginia is served by the overnight NE regional trains in a very odd way. Trains run north to Washington (and points north) via Roanoke, Charlottesville and Manassas, but somewhat inexplicably run south via Richmond, Williamsburg and Newport News. One can get an overnight train from Roanoke to Boston, but not one from Boston to Roanoke! So, let's split/recombine the overnight train in Washington, so that there are through cars from NYP/BOS/WAS to both locations. Heck, if the timing of the overnight run was shifted backwards about five hours, it would allow for very convenient journeys from Virginia to New York.

2) Re-instate the Montrealer, but add a split in Springfield - the Springfield leg continues down the B&A and up the Grand Junction (or to Worcester and then Ayer) before backing into Boston North Station (arriving there at ~04:00). After a hold there, the train leaves for Maine at about 06:00, arriving in Portland at about 08:30 (doubling as an early morning Downeaster). If there's space at North Station, a couple of cars could even be set off at North Station to allow for a BOS-NYP overnight service.

3) There are now two trains running through Springfield in the middle of the night - one is timed for a morning arrival in New York (arriving in SPG at about 5 AM), and one for a morning arrival in Montreal (arriving in SPG at about 3 AM). Why not split the train arriving in Springfield from Portland and Boston so that there are through cars from POR and BOS to NYP and MTL? This timing would only really work for Boston if some sleepers were "set out" at North Station to be picked up by the train arriving from Portland, which would pick them up at ~01:00 or so.

 

[F-Line to Dudley]

1): Currently, Virginia is served by the overnight NE regional trains in a very odd way. Trains run north to Washington (and points north) via Roanoke, Charlottesville and Manassas, but somewhat inexplicably run south via Richmond, Williamsburg and Newport News. One can get an overnight train from Roanoke to Boston, but not one from Boston to Roanoke! So, let's split/recombine the overnight train in Washington, so that there are through cars from NYP/BOS/WAS to both locations. Heck, if the timing of the overnight run was shifted backwards about five hours, it would allow for very convenient journeys from Virginia to New York.

Why a split/combine instead of just a plain old second train filling that gap? That seems like an overly complicated solution to a simple problem. Splits/combines are very schedule- and labor-intensive for anything but niche cases like the Lake Shore Ltd. and future-considerations tinker toys @ Albany. Amtrak is not going to be keen on doing that for its moneymaker Regionals while it's actively seeking to trim schedule fat like the D.C. engine swaps in favor of dual-modes and way more thru-running to Virginia. Plain old schedule expansion fills this need.

2) Re-instate the Montrealer, but add a split in Springfield - the Springfield leg continues down the B&A and up the Grand Junction (or to Worcester and then Ayer) before backing into Boston North Station (arriving there at ~04:00). After a hold there, the train leaves for Maine at about 06:00, arriving in Portland at about 08:30 (doubling as an early morning Downeaster). If there's space at North Station, a couple of cars could even be set off at North Station to allow for a BOS-NYP overnight service.

The NNEIRI study proposed pairing the existing Vermonter at Springfield with a Boston-Springfield-New Haven Inland Route slot so cross-platform transfer could complete the legs of each individual route on one ticket (the same was proposed for the Boston-Montreal direct and a connecting Inland or Regional). That obviates the need for doing any split/combine trickery for Boston, since the Inland part would be load-bearing on ridership for that corridor. Right now the Vermonter isn't an overnight, though. It's mid-morning to mid-evening both directions. NNEIRI proposed the BOS-MTL train for late-morning to mid-evening, and for a third New Haven-or-New York short-turning Montreal train to be mid-morning to late-afternoon. Adding a late-night train would require coming up with a whole-cloth 4th train. If the official studies didn't ID that timeframe for ridership, it's probably not going to fare too hot on ridership.

While NNEPRA does have its dreams of doing an NYP-POR thru train with Grand Junction trip and reverse at North Station, that's probably only going to be a one-a-day during the day because the thru audience for that is not large and it has to live inside the ridership margins of de facto paired stock Inland and stock Downeaster slots because of the hugeness of the ridership overchurn at Boston. The Downeaster's full schedule spread is not hard most hours to coordinate via Orange Line transfer from a Northeast Regional at Back Bay, and that'll only get easier when OLT pares Orange headways down to 4 minutes at peak.

3) There are now two trains running through Springfield in the middle of the night - one is timed for a morning arrival in New York (arriving in SPG at about 5 AM), and one for a morning arrival in Montreal (arriving in SPG at about 3 AM). Why not split the train arriving in Springfield from Portland and Boston so that there are through cars from POR and BOS to NYP and MTL? This timing would only really work for Boston if some sleepers were "set out" at North Station to be picked up by the train arriving from Portland, which would pick them up at ~01:00 or so.

Springfield isn't a big enough Amtrak crew-and-equipment base for doing splits/combines internally. Not like Albany is a major-major base. And it would be decidedly less ops-efficient than having a cross-ticketed/cross-platform meet of two diverging routes like NNEIRI proposes.

North Station doesn't have any Amtrak facilities base whatsoever. The Downeaster sets layover in nooks-and-crannies around Boston Engine Terminal, and crews are remoted from Southampton Yard. There's no means of doing set-outs there unless you pay a T switcher to do it (and they'll never agree to that), and Amtrak trucked its sleeper amenities restock to Somerville from Southampton (they'd never agree to that).

 

[Riverside]

Bus fare elimination has other benefits, particularly with the T's archaic fare collection policies -- all door boarding, shorter dwell times, etc.

Just because they're collecting bus fares doesn't mean they're actually making that much money from the endeavor. I would love to see an analysis of what the net income from collecting bus fares on local buses is. The T's bus network is mainly designed to funnel people to the rapid transit network and the transfer structure means that rapid transit riders get free bus trips anyway. From the 2015-2017 CPTS survey 69% bus riders have some kind of rail or reduced fare pass. If we guesstimate that a similar ballpark of the pay-per-ride riders are reduced fare or transferring to rapid transit, that means they're probably only collecting meaningful revenue from 15%-25% of bus riders who have a monthly local bus pass, are paying cash on board, or are tapping a stored value CharlieCard without connecting to a rail service. Eliminating bus fares would also result in equipment savings, labor savings, etc.

View attachment 34752

Based off this, a quick sketch of a fare reform plan:

Let's assume for the moment that most daily journeys involve at least one leg on a rapid transit line (Red, Orange, Blue, Green, SL1/2/3). With the current fare transfer system, the revenue from a bus+subway+bus journey is equivalent to the sole subway journey. Let's also assume that a majority of daily journeys involve travel to/from stations within the "Inner Belt", i.e. within a circle of Maverick - Sullivan - Harvard - Kenmore - Ruggles - JFK/UMass.

Let's also assume that there isn't political appetite to implement fare-free service systemwide (due to revenue concerns). And let's further assume that piecemeal implementation of fare-free service is seen as risky due to equity lawsuits (a concern that may or may not have merit, but is still too dicey to gamble on).

Proposal: implement a two-zone fare system, "Inner" and "Outer". Eliminate fares when boarding buses and rapid transit in the Outer Zone (i.e. beyond the Inner Belt listed above); raise fares somewhat within the Inner Zone (exact amount TBD). Potentially also eliminate fares when boarding buses in the Inner Zone (with exceptions, see below).

This would mean that the majority of services that current pay fares at the doors (buses, surface Green Line stops, GLX) would immediately be able to implement all-door boarding, speeding service and reducing dwell times. Depending on the fare increase within the Inner Zone, revenue to the T, as well as costs to the rider, could be kept largely constant. A "Residential Pass" could be made available to people living within the Inner Zone, to ensure that they aren't disproportionately impacted by "double dip fare increases".

The idea behind this proposal is that it's likely that many riders pass through a faregate at least once during their day; so, let's focus fare collection at that point in the daily journey.

The best way to make this work would also include adding faregates to the busways at Maverick, Sullivan, Harvard, Kenmore, Ruggles, JFK/UMass, Haymarket, and potentially some sort of prepayment system in Longwood (ever the difficult exception). Particularly for, e.g., Harvard, this would ensure that commuters whose journeys end at these "satellite hubs" still pay a fare once during their daily journey. (This would also handle commuters who ride the bus to Ruggles and then walk to Longwood -- they would enter busway faregates on their afternoon trip and pay then.) In the interim, these locations could maintain payment-at-front-door-boarding, and aim to have doors open a few minutes ahead of departure to avoid a last minute crunch.

Within the Inner Belt, there aren't actually that many bus routes, so the impact of making these fare free as well would probably be light. SL4/5, the Charlestown routes, the 111, and the Southie routes would each need to be examined: the 111 could recover many of its fares via outbound gates at Haymarket, but the Silver Line would be harder to find a "faregate point", which is unfortunate because it would of course benefit the most from fare-free boarding in terms of speeding things up.

It's not a perfect solution, but at first glance I think it could be implemented with reduced infrastructure cost (and could be partially implemented purely via operational changes) and with little impact to MBTA revenue or most riders' pocketbooks.

(And yes -- this would potentially mean eliminating the faregates at, for example, Porter, Davis, and Alewife. I'd be curious whether those physical faregates could be repurposed and transferred to the busways mentioned above.)

 

[clam]

Based off this, a quick sketch of a fare reform plan:

Let's assume for the moment that most daily journeys involve at least one leg on a rapid transit line (Red, Orange, Blue, Green, SL1/2/3). With the current fare transfer system, the revenue from a bus+subway+bus journey is equivalent to the sole subway journey. Let's also assume that a majority of daily journeys involve travel to/from stations within the "Inner Belt", i.e. within a circle of Maverick - Sullivan - Harvard - Kenmore - Ruggles - JFK/UMass.

Let's also assume that there isn't political appetite to implement fare-free service systemwide (due to revenue concerns). And let's further assume that piecemeal implementation of fare-free service is seen as risky due to equity lawsuits (a concern that may or may not have merit, but is still too dicey to gamble on).

Proposal: implement a two-zone fare system, "Inner" and "Outer". Eliminate fares when boarding buses and rapid transit in the Outer Zone (i.e. beyond the Inner Belt listed above); raise fares somewhat within the Inner Zone (exact amount TBD). Potentially also eliminate fares when boarding buses in the Inner Zone (with exceptions, see below).

This would mean that the majority of services that current pay fares at the doors (buses, surface Green Line stops, GLX) would immediately be able to implement all-door boarding, speeding service and reducing dwell times. Depending on the fare increase within the Inner Zone, revenue to the T, as well as costs to the rider, could be kept largely constant. A "Residential Pass" could be made available to people living within the Inner Zone, to ensure that they aren't disproportionately impacted by "double dip fare increases".

The idea behind this proposal is that it's likely that many riders pass through a faregate at least once during their day; so, let's focus fare collection at that point in the daily journey.

The best way to make this work would also include adding faregates to the busways at Maverick, Sullivan, Harvard, Kenmore, Ruggles, JFK/UMass, Haymarket, and potentially some sort of prepayment system in Longwood (ever the difficult exception). Particularly for, e.g., Harvard, this would ensure that commuters whose journeys end at these "satellite hubs" still pay a fare once during their daily journey. (This would also handle commuters who ride the bus to Ruggles and then walk to Longwood -- they would enter busway faregates on their afternoon trip and pay then.) In the interim, these locations could maintain payment-at-front-door-boarding, and aim to have doors open a few minutes ahead of departure to avoid a last minute crunch.

Within the Inner Belt, there aren't actually that many bus routes, so the impact of making these fare free as well would probably be light. SL4/5, the Charlestown routes, the 111, and the Southie routes would each need to be examined: the 111 could recover many of its fares via outbound gates at Haymarket, but the Silver Line would be harder to find a "faregate point", which is unfortunate because it would of course benefit the most from fare-free boarding in terms of speeding things up.

It's not a perfect solution, but at first glance I think it could be implemented with reduced infrastructure cost (and could be partially implemented purely via operational changes) and with little impact to MBTA revenue or most riders' pocketbooks.

(And yes -- this would potentially mean eliminating the faregates at, for example, Porter, Davis, and Alewife. I'd be curious whether those physical faregates could be repurposed and transferred to the busways mentioned above.)

What's stopping people from walking in via the bus way? Could see this as an incentive for people to do dumb things avoiding a fare gate at the bus terminal

 

[Riverside]

What's stopping people from walking in via the bus way? Could see this as an incentive for people to do dumb things avoiding a fare gate at the bus terminal

You mean like walking in via the street/roadway? I suppose that's true. I mean, that feels like it could be a solvable problem somehow. I also feel like, if riders are so inclined to avoid paying fares that they're willing to trespass through an active right-of-way, that feels like a bigger problem.

Also, depending on the numbers, it may not be worth putting gates at all of those bus terminals. Probably the only two that would see a large swath of bus-only commutes would be Harvard and Ruggles. Walking down the Harvard bus tunnel seems less likely, given its length. As for Ruggles, yeah I don't know. Like I said, this is a rough sketch of an idea.

 

[Koopzilla24]

You mean like walking in via the street/roadway? I suppose that's true. I mean, that feels like it could be a solvable problem somehow. I also feel like, if riders are so inclined to avoid paying fares that they're willing to trespass through an active right-of-way, that feels like a bigger problem.

Also, depending on the numbers, it may not be worth putting gates at all of those bus terminals. Probably the only two that would see a large swath of bus-only commutes would be Harvard and Ruggles. Walking down the Harvard bus tunnel seems less likely, given its length. As for Ruggles, yeah I don't know. Like I said, this is a rough sketch of an idea.

This could be the pretense for more covered/sheltered stops and stations. Though that would mitigate the reduced cost of implementation mentioned. For example, the Ruggles situation could be solved by making the station an entirely indoors concourse that has to be entered first in order to progress to the busways as well as current OL and CR platforms. It would be quite the infrastructure investment and would involving limiting access points and making it so that people with various mobility devices cannot just use the sloped sidewalk from Ruggles St. to enter the station and have to wait for the few slow elevators. Though I believe giving a major hub station like Ruggles an indoor climate controlled concourse is a must. Doing this becomes even more tricky at places like Nubian (which I think would benefit greatly by being transformed at least partially to and underground busway station), but in turn would be very straightforward to do at Sullivan.