Green Line Extension to Medford & Union Sq

[Scalziand]

From the FHWA:

Construction Manager / General Contractor Project Delivery

The Construction Manager / General Contractor (CM/GC) project delivery method allows an owner to engage a construction manager during the design process to provide constructability input. The Construction Manager is generally selected on the basis of qualifications, past experience or a best-value basis. During the design phase, the construction manager provides input regarding scheduling, pricing, phasing and other input that helps the owner design a more constructible project. At approximately an average of 60% to 90% design completion, the owner and the construction manager negotiate a 'guaranteed maximum price' for the construction of the project based on the defined scope and schedule. If this price is acceptable to both parties, they execute a contract for construction services, and the construction manager becomes the general contractor. The CM/GC delivery method is also called the Construction Manager at-Risk (CMR) method by state law in some states.

http://www.fhwa.dot.gov/construction/cqit/cm.cfm

Essentially design-build, yes, but it's not supposed to be cost-plus, which is where it went all pear-after-Gallagher-has-been-through-it shaped.

https://www.youtube.com/watch?v=ErppAlOIGQE

 

[underground]

We've tried for 50 years the unlimited growth of a nameless, faceless, careless bureaucracy -- aka "LBJ's War on Poverty"

Gotta love those years of unlimited growth where HUD has cut funding to cities, states, and agencies every single year.

 

[F-Line to Dudley]

There's something slightly whimsical, no, about a troll who ducks a charge of dogwhistling by blowing harder and calling it muzak?


Protip for loose-moderation boards, guys: Counterpointing the bullshit sprinkler just succeeds at validating his own sense of vanity. Mocking him for saying transparently stupid things--and being much too lazy in his dotage to trollbait with nearly the skill and stamina as he used to--consistently succeeds at sending him running for cover to lick his wounds in the relative anonymity of the pretty crane pictures threads for days/weeks at a time. If you really want the Valtrex to work on a Westie flare-up: less devil's advocate, more Nelson Muntz gets results.

 

[whighlander]

There's something slightly whimsical, no, about a troll who ducks a charge of dogwhistling by blowing harder and calling it muzak?

F-Line I just found a dogtoy that squeaks at an ultrasound frequency -- you should try it some time

 

[bigeman312]

The Boston Globe has one of the best write-ups of the Green Line cost escalation I've seen:

Green Line project could be priciest effort for MBTA

With a price tag that’s ballooned to $3 billion, the MBTA’s planned Green Line extension is on pace to be the most costly rail expansion per mile in the transit agency’s recent history — and one of the most costly among similar projects around the nation.

The revelations in recent months about the rising estimate for the extension into Somerville and Medford have prompted state officials to consider scaling back the Green Line project or, in a less likely scenario, canceling it.

...

Here is a graphic that compares this extension to other recently completed light rail extensions across the country:

 

[itchy]

The Boston Globe has one of the best write-ups of the Green Line cost escalation I've seen

I agree it's the best I've seen as well ... but it was still woefully under-reported and raised many, many more questions than it answered.

The article dwelt more on what isn't causing the insane cost escalation than what is. It briefly suggested a few potential reasons without digging into any of them, quantifying any of them, or taking a stance on which may actually be driving the cost escalation.

What we're left with is a lot of comparator datapoints that leave the reader aghast at the cost of this project:
-Why is the GLX extension 7.5 times (!!!) more expensive, per mile, than the average light-rail project in the US?
-Why is the GLX extension 3 times more expensive, per mile, than any other light-rail project in the US?
-Why is the GLX - which is **above-ground light-rail** dozens of times more expensive, per mile, than Red, Orange, and Blue Line projects requiring underground tunnel constructions in the last few decades?

Again, the article doesn't really go into what is actually causing this insane cost escalation, but the main suggestion it (briefly) offers is that land is expensive.

It's overwhelmingly hard to imagine land in Medford or in the Somerville railyards - not exactly prime downtown real estate - is more than 3x more expensive than any other land that light-rail is going through in the US.

Moreover, it's hard to believe the cost to purchase land rights in those places is dozens of times greater than the cost to obtain land rights and bore huge underground tunnels throughout Boston and Cambridge in the 1980s.

There's something truly bizarre going on with the GLX project. I don't know if the Globe fouled up its numbers, if there is some funny business going on with the contracting process, or if Medford truly is built on stacks of gold.

 

[citylover94]

I would suggest looking back at the last few pages. It is pretty clear that a lot of the cost increase is associated with the contracting process that was used and as such all phases other than those that are already ongoing are being rebid and there will be tighter controls on costs. The builder appeared to have been inflating costs to about twice the estimated amount.

 

[bigeman312]

Itchy: Yes. The write-up does a good job with the "what" but doesn't really address the "why." I wonder if anybody knows the full depth of they why. Probably not.

No, the Globe did not foul up the numbers. This light rail project is over six times more expensive than the average light rail project around the country and multiple times more expensive than the most expensive completed light rail project (per mile). The absolutely absurd price tag can not be overstated! GLX has merit. But, a gallon of milk has merit, and if your grocery store charges you $30/gallon, you do not buy it. You get the milk from somewhere else. Wasting unnecessary money is not ideal. Coming home to your family without the food they need is not ideal. Time to buy some cheaper (not necessarily worse) milk. Maybe we should ask our neighbors how they buy milk so much more cheaply than we do. Heck, I'd gladly throw them a few dollars to buy some milk for us if we are truly incompetent to the point where we will end up spending $30/gallon.

 

[Arlington]

Let's look at a typical station on the San Diego Mission Valley East project (the SECOND most expensive in the Globe's list)

The Prosecution Rests. What more need the jury see? "Light Rail" = slab stations, fully-ADA-compliant wheelchair-across-rubber-track-inserts (like at Park St) grade crossings and "Silver Line Gateway"-style amenities. Pittsburgh, Portland, Hiawatha are all predominantly this style, and most of LA MTA is too (and I assume the rest are, like Norfolk, that even when running on reserved former-rail ROW, they still to slab stations).

By powerful contrast, the GLX has "heavy rail" stations: overpasses and elevators instead of crossing tracks at grade, and headhouses and faregates. At this point, building stations to Blue/Orange standards is the single biggest reason that it is unlike its peers. Typical "light rail" peers have cheap slab stations. The GLX is far fancier--and as a result has costs that are fairly comparable (but above) "heavy rail" peers.

Here, from San Diego, is near-exactly what GLX "Ball Square" should have looked like if we'd wanted to contain costs:

Instead, we propose this:

See anything that's gotta cost an extra $50m per station? I do. (Assembly on Orange was $60m)

 

[JumboBuc]

Let's look at a typical station on the San Diego Mission Valley East project (the SECOND most expensive in the Globe's list)

The Prosecution Rests. What more need the jury see? "Light Rail" = slab stations, grade crossings and "Silver Line Gateway"-style amenities.

the GLX has "heavy rail" stations: overpasses and elevators instead of crossing tracks at grade, and headhouses and faregates. At this point, building stations to Blue/Orange standards is the single biggest reason that it is unlike its peers. Typical "light rail" peers have cheap slab stations. The GLX is far fancier--and as a result has costs that are fairly comparable (but above) "heavy rail" peers.

But the recently completed Assembly Orange Line station only cost $29 million. Build the 7 GLX stations to those specs (on average) and it adds up to $203 million. Divide that by 4.3 miles and you get an extra $47.2 million per mile. The GLX all together (at latest budgeting) costs $642 million per mile. Thus, the cost of Orange Line standard stations (on top of the cost of basic light rail stations) should only make up about 7.4% of the total per mile cost. Even if you assume the cost of each GLX station should be double the cost of Assembly (which they shouldn't be) it doesn't come anywhere close to explaining why the GLX is over twice as expensive as its most expensive peer.

 

[F-Line to Dudley]

That type of construction is unavoidable when all stations are either in a cut or on an overpass requiring vertical access from the street. And if you have to design elevators to handle the kinds of ridership these stations project to have, you have little choice but to do fare lobbies. Fundamentally the design of the stops isn't changeable. You can reexamine the headhouses and surrounding streetscaping for minor savings, but design itself is logical and more or less what it has to be to do the project at all. There's nothing to reinvent on basic station design when the change orders for making any change to final design end up costing more than keeping it as-is.

The only place where a design change might help is with the carhouse if the Brickbottom opposition that rejected the other alternatives can be tamed for a revisiting. But I'm not sure that's going to do very much either since the ramp supports for the junction are proceeding full-speed ahead as a non-canceled contract (due to impact on commuter rail ops and adjacent Northpoint development), and the Walker property they own was one of the sunk-cost pieces they ID'd as near-worthless for resale in a cancellation scenario. Plus the yard/maint facility needs to be the size it is--Reservoir-equivalent--to run the north end at acceptable headways. There's no sleight of hand they can pull busting it down to a Lechmere-size yard and playing load-balance tricks with nooks and crannies at North Station Yard and other parking spots. Odds are pretty low you're going to find more than a few dozen $M in savings shifting the carhouse around. Likewise, the Community Path has nothing to squeeze since the non-optional retaining walls reserve its right of way, and the El pegs for the Lechmere-Brickbottom path segment are being poured right this second next to the GLX supports for that non-canceled contract. Design change orders for cutting it and shifting the retaining wall positions at this stage would cost more than proceeding to build the path as-designed.


Ultimately, all of that is a distraction from the real problem here. The contractor corruption is the source for vast majority of that bloat. It's a political-structural process ripoff, not an engineering-structural ripoff. They have to re-bid it right. That's where the comparison to other recent and ongoing LRT builds is instructive. Those are true per-mile costs with bidding done completely above-board. GLX is still going to be at the more expensive end by-design because of the cut or elevation of all stations and associated drainage + wall work on the ROW only indirectly related to the project. It may still clock in at #1. But a re-bid shouldn't put it off-scale at #1, where it warps the whole graph. We're talking a healthy bit >San Diego as its natural state of engineering, but with contracting process done right the bar shrinking 3-4 ticks closer to San Diego than it is now.

 

[datadyne007]

Let's look at a typical station on the San Diego Mission Valley East project (the SECOND most expensive in the Globe's list)

The Prosecution Rests. What more need the jury see? "Light Rail" = slab stations, grade crossings and "Silver Line Gateway"-style amenities.

the GLX has "heavy rail" stations: overpasses and elevators instead of crossing tracks at grade, and headhouses and faregates. At this point, building stations to Blue/Orange standards is the single biggest reason that it is unlike its peers. Typical "light rail" peers have cheap slab stations. The GLX is far fancier--and as a result has costs that are fairly comparable (but above) "heavy rail" peers.

The stations only account for $40 million. They can be cut down, yes, but that's not the problem. The core problem, as described a million times before, was the CM/GC procurement which allowed WSK to go unchecked, as no one could meet the standards to compete with them. By cancelling the initial contracts, they are taking steps in righting their wrongs and they are being cognizant of price as they prepare the rebid process.

A side problem was that all the drainage modifications and mitigation got lumped into the GLX cost instead of the local towns (namely Somerville) paying for them as well.

Edit: Interestingly, Kiewit is actually doing that San Diego job. I noticed the img url when I quoted. Kiewit was part of the mega JV WSK (White-Skanska-Kiewit) for the GLX CM/GC.

 

[Arlington]

The stations only account for $40 million. They can be cut down, yes, but that's not the problem. The core problem, as described a million times before, was the CM/GC procurement which allowed WSK to go unchecked, as no one could meet the standards to compete with them.

I'd say there's a multiplier. Every time you added $40m in unnecessary station costs, was an invitation to the CM/GC to double it corruptly. Cut the $40m in station costs, save $80m in total bloat.

 

[citylover94]

Arlington one difference between the new San Diego line is that the expected ridership per mile is much lower so the stations don't have to handle the same level of crowding. Even in Boston a good example is based on the predicted ridership of the extension the ridership per mile of the medford branch is around 10,000 people per mile by comparison the B line gets around 6,400 riders per mile on the surface portion. I just don't see how there is a better way to design the stations especially to make them easily accessible by people with disabilities. Also it is clear the bigger issue is not the direct cost of the project but the bidding and contracting system that allowed the construction company to drive up costs the best response is not to decide to build a worse and cheaper product to save money the best response is to prevent companies from inflating costs.

On a side not the Expo Line Phase 2 is expected to cost 1.3 billion dollars and is more similar to the green line extension even including the station design and the cost is similar it just isn't inflated by the construction company.

here is a picture of a station rendering. It looks pretty similar to this to me;

 

[cden4]

Indeed, many of the stations are accessed from streets above the tracks, so stairs and elevators become necessary. But this is not true for all of the stations. One of the other main limiting factors of station design is that the T refuses to allow for any pedestrian crossings of the Green Line tracks. I can understand that this is better from an operational perspective, but it certainly adds to cost of the stations.

 

[Randomgear]

...One of the other main limiting factors of station design is that the T refuses to allow for any pedestrian crossings of the Green Line tracks. I can understand that this is better from an operational perspective, but it certainly adds to cost of the stations.

It's also a pedestrian safety issue since it separates the crossing, but also helps operations since the green line driver will not have to look out for pedestrians running across the tracks.
The fare lobbies are a good thing, too: prepayment equals easy boarding and deboarding as all doors can open. Sure, they, and the headhouses, could likely be simplified, and probably should be, to show our desire to make this work even on a budget.
As much as I would like to replace elevators (that require frequent maintenance and still tend to break down and smell of urine) with covered ramps, many handicapped people just don't have the strength to push themselves up the 20 or so feet from roadway to platform.
Unfortunately, we just don't have the year round pleasant weather that San Diego has, so I don't have issues with providing covered waiting areas.

 

[F-Line to Dudley]

It's also a pedestrian safety issue since it separates the crossing, but also helps operations since the green line driver will not have to look out for pedestrians running across the tracks.
The fare lobbies are a good thing, too: prepayment equals easy boarding and deboarding as all doors can open. Sure, they, and the headhouses, could likely be simplified, and probably should be, to show our desire to make this work even on a budget.
As much as I would like to replace elevators (that require frequent maintenance and still tend to break down and smell of urine) with covered ramps, many handicapped people just don't have the strength to push themselves up the 20 or so feet from roadway to platform.
Unfortunately, we just don't have the year round pleasant weather that San Diego has, so I don't have issues with providing covered waiting areas.

The track crossing issue is overrated in the specific instance of GLX because of the required up/down access. These are being designed as island platforms, so at least that elevator and stairs that have to be there are dumping onto a singular platform instead of requiring pairings. Islands are almost always better value for new-build prepayment stations or non- grade level commuter rail stations because of that. It's no accident that nearly every new rapid transit station they've built since 1970 has been island platform unless site constraints dictated otherwise: all of the Braintree Branch, Orange Haymarket-north, Orange SW Corridor, Davis and Alewife, relocated Green-Haymarket.


Too many egresses is something they can look at trimming back, but that's another instance of searching for nickels wedged between the couch cushions while you're tied up in your own house by the burgular who's robbing your safe and priceless jewelry collection. Wrong angle to be approaching the problem.

And gotta watch those change orders...if the egress requires shifting anything rather than just lopping it off clean, you end up fighting to a draw--or worse--on cost by reopening the design. Contractors love change orders (and love it when NIMBY's force change orders) because they can ding the project with huge premium penalties for the seemingly simplest of design alterations. It's change orders that really make downsizing stations a nearly useless exercise for recovering cost.

 

[bigeman312]

So, what should the price tag be for the remainder of the project?

 

[Randomgear]

And gotta watch those change orders..

F-Line - I'm quite aware of the amazing profits that a contractor can make on Change Orders. I'm pretty sure many contractors bid jobs with very little to no profit margin and try to make it up on CO's.

My understanding is that the project (including stations) will be rebid, so no Change Orders for any station design changes (unless the current contractor has already done some work on the stations as part of their current contract). Of course, you still have to pay the designers (Architects, Structural, Mechanical, Electrical, Plumbing, Civil Engineers and the specialized consultants) to change the design. That won't be cheap and in some instances it may not produce any cost savings. Most of the station egresses (number, location, spacing) are based on minimums required by the Building Code or user convenience and not necessarily for a grand design.

On the other hand, if the current contractor owns most of the Station work in their present contracts, we are pretty much stuck with what we have.

 

[a_tortoise]

Can someone please explain to me how erasing the entire institutional knowledge of the green line extension project by firing the design team, the construction manager, the project manager, and the cost estimator is going to save any money? The amount of money spent retracing previous steps and getting up to speed with the complexities of the historical decision-making process will easily outpace any savings achieved by programmatically downsizing the project. If Governor Baker had at least kept the design team or the project manager, there would have been a realistic means to maintain the continuity of the process and move it forward with a new set of budgetary constraints. However, starting over with the entire team makes it almost certain that the project will continue to hemorrhage money, and it also guarantees years of cost escalation that must be absorbed due to delaying the project another your or two. It would have been much simpler to just fire the CM, pursue some VE items with the same design team and cost estimator, and bid the remaining scope with a GC (rather than a CM) in one lump sum, rather than multiple bid packages. This locks in the budget at current rates and minimizes the potential for future escalation, while still allowing for meaningful cost cutting, as well as the retention of some institutional knowledge of the project.