• Automatic train protection on the Green Line?

  • Discussion relating to commuter rail, light rail, and subway operations of the MBTA.
Discussion relating to commuter rail, light rail, and subway operations of the MBTA.

Moderators: CRail, sery2831

  by RailBus63
Uh, guys - if it was just 'the occasional bump', you wouldn't have the National Transportation Safety Board involved.

When accidents like these start becoming regular occurrences, train separation and safety is a serious issue:


http://multimedia.heraldinteractive.com ... ycrash.jpg

http://cache.boston.com/bonzai-fba/Thir ... 3_0851.jpg

  by FP10
RailBus63 wrote:Uh, guys - if it was just 'the occasional bump', you wouldn't have the National Transportation Safety Board involved.

When accidents like these start becoming regular occurrences, train separation and safety is a serious issue:
The point is they aren't regular occurrences. That accident on the D was an exception for a variety of reasons. The most prominent of them all that it was out on the D, not in the central subway where ATO would cause the most problems. I'm not sure what the first picture is of however, but it appears to be underground so perhaps my foot should be nearing my mouth about now...

As I said before, I would be in full support of some sort of positive stop mechanism on the D (especially if it ment the trains being able to go faster), and at blind curves underground. But NOT if it means further slowing of underground operations.
  by CRail
No, FP10, you've got it right.

Those 4 pictures are of two incidents which occurred over 1 year apart. One was the woman on the Riverside line who hit her leader at full speed (for no apparent reason), and the other is the GC texting incident. The first occurred at 38mph, and the other at about 20mph. 2 accidents over a year apart which have nothing to do with eachother are not regular occurrences. ATO would come at a great expense (both financial cost and efficiency), and would probably be more prone to failure than human error. Such failures would bring the entire line to a halt (if you're lucky), and that would be a regular occurrence.

These accidents are dangerous, and do cause injury, but have yet to be fatal (with the exception of the offending operator in the Riverside crash). Causing the operation to become too inefficient to sustain itself (it becomes not worth the trip, and people seek other forms of transportation; or, when the bus is faster, and the streetcar is useless...) because of the negligible chance that a crash might occur, would be like shutting down the airline industry because there have been numerous plane crashes over the years.

I feel that we have the means of preventing accidents of this nature with the current signal system in place. This would require the installation of some sort of device which would relay a stop signal to the car somehow, and the installation of a device to receive this signal on all of the equipment. A non-mechanical version of a trip arm is exactly what I propose!
  by Finch
Every serious accident on the green line (within the last few years) has been the result of red signal violations. The current signal system serves the line well, the only change (well, one of two changes, but I'll get to that) I really think would help the situation is giving the signal the ability to force a train to stop (like on the blue line and in NYC). I don't think mechanical trip arms are feasible, however an absolute stop feature is certainly possible. This would have no effect whatsoever on the line under normal operation, it would only come into effect when a red signal is violated. I don't see why a time delay feature couldn't also be implemented to allow cars to pass a red after a minute as the rule allows.
This is mostly what I was getting at when I referred to "following the rules to the letter." Seems there should AT LEAST be adherence to the most important rule of all...don't run red signals. Whether people or technology are the solution to this, I know not. I personally think the traveling public would prefer to have their operators stop when they are supposed to...even if they are just going to start moving again at a restricted speed or after a time delay.

I think the definition of "regular occurrence" is subject to debate here. I would suggest a simple solution. Any accident that involves a train running into the train in front of it as a result of a rule violation or operator error is unacceptable to the general public. Period. Nobody is going to forgive the T for that, nor should they. That it happened twice within a year or so, with dramatic results both times, is doubly unacceptable. "Fender benders" are one thing (and the T will still get no sympathy for those). Avoidable collisions hard enough to injure a bunch of people and/or kill the operator are another thing entirely. You can't tell me those should be accepted as a fact of life. Maybe once or a few times in a lifetime. Not once a year or once every five years. Again, I say avoidable. Like the operator was texting, or a red signal was ignored. There are endless unforeseeable failures that can lead to a serious crash. But it's the stupid crashes that kill people needlessly and ruin the T's reputation.
  by CRail
Regarding the accident which resulted in the death of the offending operator, one speculation is that there may have been a health issue involved making the operator unaware of her surroundings. You can't blame the T for that, it is an unfortunate event which took place and was presumably unpreventable by anyone involved.

Lot's of things are unacceptable, but this doesn't make every alternative acceptable. Fixing a problem with a solution which is worse is unacceptable. As a former daily commuter on a line operated automatically, I (a member of the public) regard the delays I encountered due to the automated system's numerous faults as unacceptable.

I feel that if I'm paying the T to get me to work on time every day (so long as I provide an adequate amount of time for them to get me there) they should do so almost every day (with the understanding that unpredictable events occur). If they can't meet those standards because of the complications of a system of which the duties would be better performed by a paid employee who is present but not allowed to perform such duties, then that is unacceptable.

Yes, I agree something should be done, but automated operation is not it!
  by Jersey_Mike
Does the Mattapan line have any kind of signaling? For the life of me I can't remember if it does. Smart move installing a pilot project where it will have the least impact. Cab signals may be effective on the Green Line branches (the D especially), but on the downtown trunk they would drastically cut capacity.
  by sery2831
The line only has 2 manually controlled starter signals at the terminals. Never has had signals. The cars didn't have radios until a serious rear ender accident back in the 90s.
  by Disney Guy
The original trip arm idea was to stop the train when it passed the red signal. This cannot work for much of the Green Line subway because, if a train passed a red signal at speed, a train occupying the block ahead might be stopped with its rear less than a carlength from the approaching train.

So a safety mechanism needs to slow down a train approaching a red signal and/or the train ahead. This means interacting with the approaching train before that train reaches a red signal. Granted, though, the operator should be slowing down the train anyway before reaching a red signal or another train.

For wayside sensors, multiple sensors would be needed for each signal, each being able to wirelessly interact with a train. Each sensor would need to send a specific message specifying the desired speed the train needs to slow down to. Sensors mounted on trains (and perhaps on the subway wall at curves) may suffice, without sensors at every signal.
  by jwhite07
How cute. They made it look kinda like a center headlight or a trolley retriever.

How's that low-mounted camera or whatever is in the square housing going to work when it's caked with snow and ice?
  by bellstbarn
Maybe the Boston Green Line should try out the type of warning devices that are now (as I understand it) fairly common in the automotive business. A friend's rather inexpensive new car will warn him if he dares to move right or left into an occupied lane. Stop signals already cover merging routes in Boston's underground. Unless I am wrong, new charter coaches come with a warning noise if the driver creeps too close to a vehicle in front of him. Maybe some of these warning devices also cut the accelerator or apply brakes. In Newark, Philadelphia, and Boston there have been injurious rear-enders. I doubt the systems developed by the automotive industry are fouled by ice cover. They could factor in the rapidity of moving into a collision.
  by jwhite07
It's certainly not groundbreaking technology... both of my cars (relatively inexpensive Toyotas) have frontal collision avoidance systems. One problem is it's perhaps too sensitive - it won't just slow you down, it will lock 'em up on you and make you pray that the car behind you and the one behind them too either have this system too or are acutely paying attention. That happened to me when someone abruptly changed lanes on the highway about 10 feet in front of me. I was paying attention and noticed what the crash test dummy was about to do and I'd let off the gas and was about to tap the brakes and give the bozo a little more room, but as soon as he greased his way in front of me, my car said NOPE and dropped the brake pedal out from under my foot right to the floor, and I went from 70mph to 20mph in about three seconds before the brakes released and I could accelerate again and avoid getting rear ended. Fortunately the roads were dry and it didn't cause me to go into a spin! Maybe too much of a good thing there...

I wonder how standees are going to like it when a car cuts a streetcar off on Huntington Avenue and activates this system? I have had the experience of being on a streetcar at low speed when the motorman hit the red "easy button" on the dash, and at low speed those magnetic track brakes will stop a streetcar almost instantly and send people tumbling. Better than the alternative? Probably, but still... I at least hope there are no false activation issues.
  by MBTA3247
bellstbarn wrote: Wed Feb 15, 2023 1:19 pmI doubt the systems developed by the automotive industry are fouled by ice cover.
You would be mistaken in that thought. The primary system in my car becomes disabled if there is ice, fog or excessive rain hitting the windshield. I had a rental once where the system became disabled because the sensor was covered in bug guts.
  by Disney Guy
"" I wonder how standees are going to like it when a car cuts a streetcar off ... ""

Perhaps the crash avoidance system needs to handle trolley problems (no pun intended) using a different strategy. Sacrifice the errant motorist ahead instead of the standing passengers.. Plenty of automated cameras taking photographs would make it easy to reconstruct what happened and even prove that the car hit in the rear was at fault.

No, I did not say blindly slam into the car in front but rather not make complete avoidance the top priority.