General MBTA Topics (Multi Modal, Budget, MassDOT)

Hey MBTA, what the [insert four-letter word of your choice] happened?
Screenshot 2023-09-13 at 7.34.01 PM.png

The Sullivan-Community southbound slow zone is now back up to a 3 minute travel time after being at line speed for all of 2 days a week ago before increasing to nearly 2.5 minutes.

I think this is the finest example of slow zone repair being one massive game of whack-a-mole for the T.
 
Another thing that’s incredible is the poor state of RL Alewife-Harvard both directions. After 10 years of off & on weekend diversions for the Floating Slab project, where are the results? Or are we just “lucky” it runs at all.
 
Some big yikes: https://www.bostonglobe.com/2023/09...ummer-after-feds-rapped-agency-sloppy-safety/

According to the report, a [Red Line] work crew was located beyond a curve in the tracks in the shadow of a bridge, and overgrown vegetation next to the track area was partially obstructing the flagger. Moreover, the train operator was unaware of the work crew because they didn’t hear a radio call, the report said.

“Based on the report from the radio department, it does not appear that the train radio was in the on/active position at the time of the event,” the report said.

Then, on Aug. 28 at 2:14 a.m., two workers in the track area on the Green Line between the Brookline Hills and Brookline Village stations saw a train traveling toward them at about 10 miles per hour, according to a preliminary report of that incident obtained by the Globe. The area was under “level 1 protection,” according to the report, meaning no trains were supposed to enter the area.

“The dispatcher lost track of the crew on the [right of way] that had been given level 1 protection,” the report said.

Then, on Sept. 6 at 4:14 a.m., workers on the Red Line reported that they were in a “level 1 area with equipment coming at them,” according to a preliminary report of the incident obtained by the Globe. The report did not make clear exactly where the incident occurred.
 

I don't know what the "norm" is for rail procedures, but having sometimes listened to Air Traffic Control, I'm curious here.

There the norm seems to be to require to require acknowledgement for most/safety-critical instructions and if the acknowledgement isn't received as expected/there's radio silence for a response the controller starts acting as if they're going to ignore it (warning others, directing them to move them out of the way, etc).

Is rail/MBTA procedure to just make a radio call and assume others hear it?

(edited for minor wording).
 
Last edited:
357 vacancies for bus operators as of Aug 30th., changed from 368 vacancies in late July.


1694702953935.png

1694702990762.png
 
I really want to know what the estimated risk of a derailment is and whether that is worth slowing all the trains down to jogging speed for months. It's barely worth having a mass transit system when it runs like this and I really question the need for ALL the slow zones given that trains ran over them for years. If Eng can't get it together to make actual progress or tell people when progress will be made or what they would need from the legislature or municipalities to MAKE progress, then what is the point. There is enough money in the capital budget and they can and do hire consultants to help figure things out. Do it.
 
Because of recent near misses on the tracks, the FTA is putting new restrictions on how MBTA can work on the tracks.

and the letter from the FTA

Any ideas how big these restrictions are? What they mean practically for repair work?
 
Because of recent near misses on the tracks, the FTA is putting new restrictions on how MBTA can work on the tracks.

and the letter from the FTA

Any ideas how big these restrictions are? What they mean practically for repair work?
I’m interested in the Globe’s claim that the FTA is tacitly threatening to shut down service:
if the agency doesn’t take immediate actions to prevent safety failures on its subway, they could cut off track access — effectively shutting down service — until officials are confident that the T can keep workers safe.
The closest I can find in the FTA letter is this:
Failure to comply with these requirements may result in suspension of all activities that place workers on the ROW, including maintenance and inspection, until FTA is confident that the MBTA can ensure workers are adequately protected from collisions on the ROW
Which doesn’t explicitly mention shutting down service. Do people “in-the-know” interpret “suspension of maintenance and inspection” as tantamount to “suspension of service”?
 
i am sure it's a lot of work - but only 11 net new bus operators?

We won't see the impacts of the increased application rates/hiring from the recent pay bump until ~October/November. MBTA hiring process is glacial, and bus operators sign up in advance for a scheduled training program ~2 months out, I believe... Perhaps they've sped this up given the gravity/new leadership, but there was some insight posted on reddit a few months ago about getting onboarded as a bus operator.
 
We won't see the impacts of the increased application rates/hiring from the recent pay bump until ~October/November. MBTA hiring process is glacial, and bus operators sign up in advance for a scheduled training program ~2 months out, I believe... Perhaps they've sped this up given the gravity/new leadership, but there was some insight posted on reddit a few months ago about getting onboarded as a bus operator.

400% increase in bus operator applications in August. By the time they go through the hiring process, onboarding, and training, you're looking at 3.5 - 6.5 months? That means maybe a very tiny service increase around Christmas? But problaby the bulk of the impact won't make it until the March 2024 service change. MBTA fiscal cliff is around mid-2024 around June so maybe we'll only get the benefits for 2-3 months before that falls over.
 
Another pair of near miss incidents, apparently involving the same red line operator.

I will say that safety systems that rely on human performance and adherence to regulations can inherently be problematic. It's why the railroads now have to implement PTC/ATC, why aviation has so many safety systems designed to enforce compliance with the rules. Humans are invariably faliable, and even if mechanical systems are too there's a good area where they can complement each other. I realize that the MBTA heavy transit lines are equipped with ATC such that there are trip-stops at signals, but is there such a thing as a "portable trip-arm?" Something they can flip up and down as needed? I feel like there must be, and if there is, the T would be well advised to provide one to all of it's track crews post-haste. A flagger, by themself protecting a track crew is merely providing a signal to a driver, and lacks any means by which to enforce anything.
 
Last edited:
I wonder if adopting practices like the point & call system would be helpful? It’s used in transportation agencies throughout Japan and other places as well. It helps focus attention, and in vehicles where there are cameras in the cab it could be used to help evaluate operator attentiveness and adherence to procedures. It just seems like many things at the T and other peer agencies are done inside a bubble. Why not look at best practices around the world and learn from that?
 
I wonder if adopting practices like the point & call system would be helpful? It’s used in transportation agencies throughout Japan and other places as well. It helps focus attention, and in vehicles where there are cameras in the cab it could be used to help evaluate operator attentiveness and adherence to procedures. It just seems like many things at the T and other peer agencies are done inside a bubble. Why not look at best practices around the world and learn from that?

I was just checking out the new ho chi minh metro in vietnam and noticed that they were using this system, as japan had built the metro and trained the personnel. It stood out because I dont remember seeing it used outside of japan. If theres ever a system to model yours after its definitely japan. I definitely agree that the T should be willing to use whatever best practices there are from around the world, if it means a better functioning metro. It definitely couldnt hurt.
 
I was just checking out the new ho chi minh metro in vietnam and noticed that they were using this system, as japan had built the metro and trained the personnel. It stood out because I dont remember seeing it used outside of japan. If theres ever a system to model yours after its definitely japan. I definitely agree that the T should be willing to use whatever best practices there are from around the world, if it means a better functioning metro. It definitely couldnt hurt.
NYC does their version of point and call on the subway. They have since 1996, apparently.
 
Another pair of near miss incidents, apparently involving the same red line operator.

I will say that safety systems that rely on human performance and adherence to regulations can inherently be problematic. It's why the railroads now have to implement PTC/ATC, why aviation has so many safety systems designed to enforce compliance with the rules. Humans are invariably faliable, and even if mechanical systems are too there's a good area where they can complement each other. I realize that the MBTA heavy transit lines are equipped with ATC such that there are trip-stops at signals, but is there such a thing as a "portable trip-arm?" Something they can flip up and down as needed? I feel like there must be, and if there is, the T would be well advised to provide one to all of it's track crews post-haste. A flagger, by themself protecting a track crew is merely providing a signal to a driver, and lacks any means by which to enforce anything.
I’m frustrated by the Globe’s coverage here. There is an obvious follow-up question that must be asked upon the revelation that the same operator blew past a flagger twice: why was the operator allowed to continue working?

Is there no reporting mechanism internally? Is there such an operator shortage that people feel pressure not to report things? Was information siloed — reported, but not reported to the right people?
 
I’m frustrated by the Globe’s coverage here. There is an obvious follow-up question that must be asked upon the revelation that the same operator blew past a flagger twice: why was the operator allowed to continue working?

Is there no reporting mechanism internally? Is there such an operator shortage that people feel pressure not to report things? Was information siloed — reported, but not reported to the right people?
According to the article, the implication is that the track crew simply didn't report the first incident until the second one occurred, effectively reporting both at the same time. If the track crews didn't report it the first time it happened, that speaks to a different safety culture problem, one of laxity and risk normalization.

While speaking with the T’s safety department about the incident on Monday, the two-person crew reported another close call in the same area and involving the same Red Line operator on Sept. 11, the report said.
 

Back
Top