The cost of implementing Regional Rail

I just finished watching the video replay of Monday’s MBTA control board meeting, which was mostly about buses and bus drivers, and I found very little notable about it other than the fact that the MBTA’s Chief Engineer hates catenary, and this seems to be driving the (crazy to me) push towards replacing perfectly good trolleybus lines with battery-powered buses that must sit in an environmentally-controlled depot for large parts of every day to be charged (and which in any case are still much less efficient because you’re still lugging around that weight of batteries when you could have permanent infrastructure in place that doesn’t have to be dragged around with the passengers all day. I emailed some comments to the FMCB asking that, in public outreach about the bus program, they do a better job of both quantifying the costs and of documenting the crossover point between when carrying your power source is worthwhile compared to running wires (especially on high-frequency lines that really should be trams anyway). But that’s not really what I wanted to write about. Instead, this post is a followup on Monday’s post about scheduling Regional Rail on the Framingham/Worcester Line.

You’ll recall that I figured out a way to maintain eight trains per hour on the line during peak periods with a total EMU inventory of only 24 units. Totally coincidentally, MIT planning student Ari Ofsevit tweeted a thread about what it would actually cost to get level boarding throughout the commuter rail network, which is the first step of Regional Rail and, as I’ve advocated, should be the state’s number one priority after Positive Train Control is fully implemented (which is a Federal Railroad Administration mandate, and is also scheduled to be done by the end of 2020, which is the earliest any Regional Rail construction could plausibly start anyway, just based on the MBTA and state capital planning cycles). Ari counted 148 high platforms that needed to be built across the system, and 20 stations that need elevators (which cost much more) or other sorts of vertical circulation for accessibility. He conservatively estimates a construction cost (including design) of $3 million per platform, and then another $20 million for each elevator installation (some stations may need more than one). This alone would improve schedule adherence and reduce the staffing requirements that exist currently due to the need for conductors to operate door traps at each low-level station. Ari estimates the total cost of this capital improvement at $900 million.

For Regional Rail, we need two other components: rolling stock and electrification infrastructure (primarily catenary and substations). Pretty much everything else that is absolutely required, from double-tracking on the Franklin line to reworking overpasses to rehabilitating drawbridges to building a second platform at Worcester, is either already programmed or already planned. If the average MBTA commuter line can be served with 24 of those 75-meter Stadler FLIRT EMUs that I’m so fond of — the Providence and Fitchburg lines will require more, the Fairmount line (assuming it isn’t rejoined to the Franklin Line) and the Lowell Lines will require fewer (and the Old Colony is probably the last one to get Regional Rail anyway, because it’s new construction and the single-tracking through Dorchester limits service levels substantially) — then overall the entire system would need 300 of them. They cost about $8 million each, maybe less, call it $3.2 billion total for rolling stock and high platforms. Having a single, uniform fleet of rolling stock would be a tremendous cost savings to the MBTA, and with the cars delivered over the course of a 10-year contract, there would be no future huge hits as overhauls and replacements could be done over a similar time frame, 15 to 25 years after delivery.

Then comes the question of electrification. Other places can do it more cheaply, by quite a large measure, but let’s be pessimistic and assume that electrifying the MBTA system would cost as much per mile as the Amtrak Northeast Corridor extension from New Haven to Boston did 20 years ago, adjusted for inflation. That would work out to about $1.5 billion. Some of the infrastructure doesn’t need to be built out, because the catenary already exists along the Providence Line for most of its length, but yard tracks would need to be wired, and there’s additional complexity at turnouts; the Commuter Rail Maintenance Facility would need some upgrades as well. But all told, we’re talking about a $5 billion capital investment that would dramatically improve service, improve long-term reliability and state-of-good-repair, reduce operating costs, reduce carbon and particulate emissions, and make the whole commuter rail network fully ADA compliant. It would also make it possible to allow bikes on trains at all times, not just when it’s dark out.

How can we pay for this? Well, for starters, cancel South Coast Rail phase 1. That doesn’t quite free up the $1 billion, because some of the phase 1 components are also part of phase 2, like the real estate and rehabilitation of tracks into New Bedford and Fall River. But commit to Regional Rail and you can commit credibly to SCR phase 2 as a part of the Regional Rail program. Secondly, don’t build South Station Expansion. With Regional Rail, station dwell times should be 15 minutes or less on all lines, because nearly all trains turn around and run back out to the other end of the line (where they are less constrained on turnaround time). That saves another $1 billion. Finally, don’t build the unnecessary midday layover facility in Allston: for the Framingham/Worcester Line, you can send the trains to Framingham, as I’ve shown in my schedule, and in general, Regional Rail allows you to lay over trains at Southampton St. again, because they’re no longer emitting diesel smoke into the neighborhood when they are starting their trips. (My best guess is that upgrading the Agricultural Branch and building the two new stations and a layover facility should come in at around $35 million; a layover facility at Fountain St. would be much cheaper, possibly as little as $3 million. I would not build a parking structure at Framingham Tech Park, but encourage nearby businesses to either charge for parking in their surface lots or construct a privately owned parking structure and a pedestrian bridge to the station.)

Again, it’s $5 billion, to make a real, substantive improvement in our commuter rail system. It doesn’t depend on North-South Rail Link. (The other way around, actually: doing NSRL requires electrification, although not necessarily the full Regional Rail treatment, and switching to an all-EMU equipment roster would make some aspects of NSRL cheaper to construct.) It would avoid at least $2 billion in currently planned capital costs. We need to just do this, and we need to make it clear to the MBTA that we’re not taking “no” (read: “I don’ wanna!”) as an answer. Over the course of ten years, it would be less than $500 million a year to implement, possibly quite a bit less depending on how quickly we (1) commit to doing it, (2) get those orders in for rolling stock and electrical equipment, and (3) stop investing in obsolete diesel locomotives and locomotive-hauled coaches.

Time for me to go to bed. (This post is scheduled for delayed publication when people will hopefully be awake to read it.)

UPDATE 2018-09-26: I spent time tonight digging through the MassDOT FY2019-2023 Capital Investment Plan, trying to identify programs that should be cut in favor of early action implementation on Regional Rail. I found $27 million budgeted for locomotive overhauls, which need to be done to maintain state-of-good-repair, and likewise another $112 million to overhaul the Kawasaki bi-level coaches. But there’s also $27 million to buy new locomotives and $245 million to buy new bi-level coaches, and that is just throwing good money after bad. A combined $272 million of already programmed funds would complete electrification on the Providence/Stoughton Line, build the required expansion of Sharon substation, electrify the Fairmount Line, build all the necessary high-level platforms, and fund the acquisition of at least 20 250-passenger EMUs. (By the way, the highest-capacity bilevel coaches have only 180 seats, but of course replacing bi-levels with level-boarding EMUs implies operating more trains, because the improved boarding speed and reduced dwell time of the EMU comes at a cost of taking more space on the platform. But certainly all off-peak service on all three lines could be replaced with EMUs even without increasing frequency, and the diesel equipment moved to other services.) On the other hand, the Allston midday layover facility that I mentioned in the original post doesn’t turn out to be much of a budget hit; the issues there are more to do with the impact of that facility on West Station and eliminating the Turnpike viaduct.

This entry was posted in Transportation and tagged , . Bookmark the permalink.

One Response to The cost of implementing Regional Rail

  1. Pingback: Rolling stock for Regional Rail: What and how to buy | Occasionally Coherent

Comments are closed.