Euclid If this collision was caused as speculated here, how would have more advanced information presented to the crew prevented the collision? What would they have done different if they had more information?
If this collision was caused as speculated here, how would have more advanced information presented to the crew prevented the collision? What would they have done different if they had more information?
Nobody here is psychic. We don't know what they WOULD have done.
It's been fun. But it isn't much fun anymore. Signing off for now.
The opinions expressed here represent my own and not those of my employer, any other railroad, company, or person.t fun any
zugmann Euclid If this collision was caused as speculated here, how would have more advanced information presented to the crew prevented the collision? What would they have done different if they had more information? Nobody here is psychic. We don't know what they WOULD have done.
Maybe I was not clear, but I am not asking for a psychic prediction about what the crew would have done or did do. My point was directed at the suggestion that informing a crew of upcoming meets would make things safer. So my question was to ask this:
Is it thought that, had the crew been informed of the upcoming meet, they would have been less likely to run the red board?
EuclidIs it thought that, had the crew been informed of the upcoming meet, they would have been less likely to run the red board?
Who knows. I've heard of crews that were told they were going to be stopping at CP-Whatever, and still got by the signal.
Communication is always good - but you can't be dependent on the dispatcher telling you everything. You're usually not the only crew out there.
Euclid zugmann Euclid If this collision was caused as speculated here, how would have more advanced information presented to the crew prevented the collision? What would they have done different if they had more information? Nobody here is psychic. We don't know what they WOULD have done. Maybe I was not clear, but I am not asking for a psychic prediction about what the crew would have done or did do. My point was directed at the suggestion that informing a crew of upcoming meets would make things safer. So my question was to ask this: Is it thought that, had the crew been informed of the upcoming meet, they would have been less likely to run the red board?
If a crew is unable to communicate - for whatever the reason - how do they get notified? The unabated speed of one train, would be prima facia evidence that they were unable to communicate with either the Train Dispatcher or the signals displayed for them.
Never too old to have a happy childhood!
Well, at some point, they were able to communicate. If they, in fact were asleep, is it being suggested that, had they been informed of the meet before falling asleep, it would have been more likely that they would not have fallen asleep and ran by the board?
Understand that I am not advocating such notification. On one hand I can see the argument that any added communcation about the operating plan can't hurt and will probably help. But I also see zugmann's point about the danger of crews relying on that extra communication. I agree with that point, and think that adding information redundancy can backfire by a greater tendancy to inadvertantly omit the redundant information, which can catch a crew that relies on it.
you're asking the wrong question, twice.
First, as noted already, the whole premise of CTC is to take dispatcher communication and train orders out of the picture. You run your train by signal authority only ... and, to be safe, the blocks are long enough, and the relay setup reliable enough, that you never see a red without having enough time to stop before hitting or splitting anything.
Second, there needs to be automatic train stop on absolute red. Any time, anywhere. One of the four functions of mandated PTC is to provide this, but it was just about as simple before WW1 -- magnet on the track that 'fails safe', that applies full (safe level of) service brake and, on a diesel-electric consist, either eases throttle and DPU off or drops the ground relay to the traction power.
In answer to your question: Yes, it would be nice to have the dispatcher radio the crew to alert them when they're meeting opposing traffic, that they'll be crossing over, etc. But on a proper and safe CTC system this is not adding anything new to safety -- and in the present accident, it is likely the dispatcher could have said all kinds of things that wouldn't have made any difference.
If no one is listening, it doesn't matter what a dispatcher says. If no one is following train orders, it doesn't matter what the train-order procedures are. If no one is controlling the throttle and brake, it does not matter what the signals are set to. So no, it likely wouldn't have mattered if the dispatcher had 'informed' that eastbound crew about an impending meet a little in advance of what the signals were telling them, or explained what they were going to encounter over the next couple of miles, or at the Panhandle crossover.
I'll leave it to the practical railroaders to comment on whether continuous information that may compromise the 'absolute meaning' of CTC displayed signals is a safe 'generally accepted procedure' for dispatching. In my opinion (as an IxD person who knows cognitive science) some aspects of that approach are beneficial, and some decidedly not beneficial ... in specific circumstances, with particular appropriateness in 'anomalous situations' or critical safety contexts (as in assisting untrained pilots in landing heavy aircraft, or responding to emerging failures in nuclear power-plant accidents). Here, the only real 'warning' the dispatcher could have given was to the westbound crew, to give them some few seconds advanced warning -- or to advise the conductor on the westbound to jump with her engineer, as death was bowling for her otherwise.
zugmann oltmannd Why not just give the train crew the dispatcher's track line view of the railroad... No more room for another screen in these engines.
oltmannd Why not just give the train crew the dispatcher's track line view of the railroad...
No more room for another screen in these engines.
It'll be even worse with PTC. Still, there are ways. Or you could let them have it on their phones...Which opens up a whole 'nother can of worms.
More info of this nature increases situational awareness. None of it is "bullet proof". That's what the signal system's for...
-Don (Random stuff, mostly about trains - what else? http://blerfblog.blogspot.com/)
Overmod Euclid Is it thought that, had the crew been informed of the upcoming meet, they would have been less likely to run the red board? you're asking the wrong question, twice. First, as noted already, the whole premise of CTC is to take dispatcher communication and train orders out of the picture. You run your train by signal authority only ... and, to be safe, the blocks are long enough, and the relay setup reliable enough, that you never see a red without having enough time to stop before hitting or splitting anything. Second, there needs to be automatic train stop on absolute red. Any time, anywhere. One of the four functions of mandated PTC is to provide this, but it was just about as simple before WW1 -- magnet on the track that 'fails safe', that applies full (safe level of) service brake and, on a diesel-electric consist, either eases throttle and DPU off or drops the ground relay to the traction power. In answer to your question: Yes, it would be nice to have the dispatcher radio the crew to alert them when they're meeting opposing traffic, that they'll be crossing over, etc. But on a proper and safe CTC system this is not adding anything new to safety -- and in the present accident, it is likely the dispatcher could have said all kinds of things that wouldn't have made any difference. If no one is listening, it doesn't matter what a dispatcher says. If no one is following train orders, it doesn't matter what the train-order procedures are. If no one is controlling the throttle and brake, it does not matter what the signals are set to. So no, it likely wouldn't have mattered if the dispatcher had 'informed' that eastbound crew about an impending meet a little in advance of what the signals were telling them, or explained what they were going to encounter over the next couple of miles, or at the Panhandle crossover. I'll leave it to the practical railroaders to comment on whether continuous information that may compromise the 'absolute meaning' of CTC displayed signals is a safe 'generally accepted procedure' for dispatching. In my opinion (as an IxD person who knows cognitive science) some aspects of that approach are beneficial, and some decidedly not beneficial ... in specific circumstances, with particular appropriateness in 'anomalous situations' or critical safety contexts (as in assisting untrained pilots in landing heavy aircraft, or responding to emerging failures in nuclear power-plant accidents). Here, the only real 'warning' the dispatcher could have given was to the westbound crew, to give them some few seconds advanced warning -- or to advise the conductor on the westbound to jump with her engineer, as death was bowling for her otherwise.
Euclid Is it thought that, had the crew been informed of the upcoming meet, they would have been less likely to run the red board?
Overmod,
You respond as though I am advocating this extra communication. Yet I explicitly said above that I am against the idea. However, there is support here for the idea of extra information such as telling crews of upcoming meets. I am asking my question of those advocates of extra information, so it is not a wrong question. I want to hear why they think such meet notification would add safety. You yourself have answered the question when you said this:
"In answer to your question: Yes, it would be nice to have the dispatcher radio the crew to alert them when they're meeting opposing traffic, that they'll be crossing over, etc. But on a proper and safe CTC system this is not adding anything new to safety -- and in the present accident, it is likely the dispatcher could have said all kinds of things that wouldn't have made any difference."
So how do you conclude that such reminders cannot possibly change the outcome?
Since I am not getting any answers to my question from those who favor the extra information, I will answer for them based on what I think their reasoning is as follows:
Informing a crew of an impending meet maybe 30 minutes ahead of time would reinforce the matter into their consciousness, and motivate them to be more alert. So maybe they would be less likely to fall asleep or lapse into some other kind of inattention.
It would be like when you make a dentist appointment. Everything is locked into the schedule, yet they call you a day ahead of time to remind you. This issue is obvious. Sometimes crews fail to act as required by signals for a variety of reasons. So reminding them of upcoming action can’t hurt, and it may help.
Again: The part is blue is NOT my position. It is the answer to my question those who advocate the extra communication.
EuclidThis must be the video that you refer to where one of the trains is moving fast. It is the second video down in this linked story: http://www.mysanantonio.com/news/local/article/Several-videos-capture-moments-after-trains-8331885.php
No, that's still at least several seconds later than the one I saw -- the top of the 'cloud' is much higher and has turned a full gray color in this video, and I think the angle is different.
I referred earlier to a video by the New couple; I based that on it being the attribution given by the TV station to the video on its Facebook page. It now appears that video was changed TO the version provided by the News, which is the one that shows the last few seconds of the westbound train stopping. So the version I was commenting on initially is still not 'up' in media where I can link to it now.
There are problems in a safety-critical system when you start mixing reminders at different times or lags and there is no written or persistent backup of the "workflow". Reminding someone about an upcoming meet in 30 minutes is likely to be 'overwritten' by subsequent things that happen in foreground attention, unless there is a written reminder or prompt that flags it (with some hierarchical method of signaling as the time of the reminder's importance approaches). Meanwhile, if the meet changes for some reason, one of two things may happen: the latency or perceived primacy of the original communication may override the recency of the 'update'; and if there are too many changes the crew may start 'zoning out' the early warnings, and extend that over to later ones as well -- with increasing risk as they become tired, stressed, or concerned with other events or emergencies.
I would be highly in favor of Don's tongue-in-cheek proposal to 'give the crew the dispatcher's view of their piece of the railroad' -- in much the same format as the old Microsoft moving-map GPS did. This would give you a route line with your train 'transponders' indicated, front and rear, with telemetry, and all the signals, switch positions, anticipated crossovers, other train information, etc. displayed. It would also have a 'repeater' for all the civil restrictions, slow orders, and other transient things the crew would be expected to remember -- the immediate problem then becoming they will increasingly rely on the 'repeater' to show them the current state, without remembering and reviewing the actual paper communications that are the 'official' proof in writing on which the system rests.
An armchair Apple engineer would make this sort of display, with data fusion, the primary train=handling interface. The problem with that is that if the system goes bad -- and it depends on a number of radio-based systems any of which in failing may eliminate safety-critical inputs -- it will abruptly shift from mandatory to untrustable, a very poor thing for a safety critical system to do, and manifestly unfair to do (perhaps over and over in the course of an eight-hour workday) to a crew that has to depend on its connectivity for 'safety assurance'.
Note that it is not easy to predict when the system fails, or how reliable its information is, without using precisely the sort of foreground attention and complex activity that has been demonstrated to cause accidents when certain smartphone activities are used ... see the Air France crash with the pitot-tube freezing for an extreme, but perhaps very relevant example.
Frankly, I think it's much wiser to keep the system 'automatic' and predictable using nothing but rules-based signal indications, and use an overlay for the civil and dispatch 'refinements' within the same visually-based system. If you have dispatcher communications, have them read and text simultaneously where, if a crew shifts their eyes to read them, they will see an easily-comprehended instantaneous signal display at the same focus their eyes adopt to read them. With a simple control to play back part of the message if it isn't understood when heard ... but not a bunch of fiddly controls with no haptic feedback, that require a crew to look at them every time to figure out what they need to put their finger on.
I have my own thoughts on how an e-log system ought to contain its own record of slow orders and other civil stuff -- provided to a crew in defined electronc form at the time they leave to get on the engine, and then uploaded in equally defined electronic form to a standard system on the engine. That is then the time and place to compare the electronic version, step by step against a checklist, with any 'paper' or verbal instructions, to be sure everything is loaded correctly and everything on the engine that reads things is working as expected, or workarounds are satisfactory and understood. Etc., and there are a whole bunch of etc's that I'm not going to go into.
But don't mix media, and don't give a crew TMI, especially if it distracts them to receive or process it, and don't mix things up by giving them information out of good haptic sequence...
...except in emergencies. And when you call an emergency, be very careful to keep it well-managed as an emergency, neither too much or too little excitement and interference. Automatic systems that trigger on emergency conditions shouldn't bray alarm, either ... there are better ways to get a crew to respond, if they are able. And there better be ways to implement a response if the crew does not...
Overmod Euclid This must be the video that you refer to where one of the trains is moving fast. It is the second video down in this linked story: http://www.mysanantonio.com/news/local/article/Several-videos-capture-moments-after-trains-8331885.php No, that's still at least several seconds later than the one I saw -- the top of the 'cloud' is much higher and has turned a full gray color in this video, and I think the angle is different. I referred earlier to a video by the New couple; I based that on it being the attribution given by the TV station to the video on its Facebook page. It now appears that video was changed TO the version provided by the News, which is the one that shows the last few seconds of the westbound train stopping. So the version I was commenting on initially is still not 'up' in media where I can link to it now.
Euclid This must be the video that you refer to where one of the trains is moving fast. It is the second video down in this linked story: http://www.mysanantonio.com/news/local/article/Several-videos-capture-moments-after-trains-8331885.php
So with all the video and still shots of this wreck, the one video that is the most uniquely revealing and the most unlikely to get has been posted on line for a few days or so, and now has been removed.
EuclidSo with all the video and still shots of this wreck, the one video that is the most uniquely revealing and the most unlikely to get has been posted on line for a few days or so, and now has been removed.
I would say this a bit more carefully. But yes, I am saying that a video was available (on a TV stations' Facebook page, with 800-odd comments) that showed something very different from the video that shows now on that page. And I am further saying that I have not yet found the original video elsewhere on the Web.
I'm not saying anything about how that might come to be, other than to note that when I do find it I'll provide a link.
Euclid If this collision was caused as speculated here, how would have more advanced information presented to the crew prevented the collision? What would they have done different if they had more information? When I worked road trains I was always glad to get information about what was coming up ahead. If I knew about a meet in advance I could time my arrival at the meeting point to avoid having to stop thereby saving fuel and wear and tear on the equipment. Or I knew I would be stopping, giving some time for a bathroom break or time to check on a trailing unit. In this incident the dispatcher calling on the radio "may" have gotten the crews attention. Mark Vinski
schlimm Buslist Because as has been said before, a properly functioning CTC system's field logic would not allow a dispatcher to make such an error. So we would need a malfunctioning signal system and a malfunctioning dispatcher to satisfy you. Looks like Dave Husman and Jeff Hergert have discussed other ways that the accident could occur. Perhaps you can turn your snarkiness on them?
Buslist Because as has been said before, a properly functioning CTC system's field logic would not allow a dispatcher to make such an error. So we would need a malfunctioning signal system and a malfunctioning dispatcher to satisfy you.
Looks like Dave Husman and Jeff Hergert have discussed other ways that the accident could occur. Perhaps you can turn your snarkiness on them?
neither one of those constitute dispatcher error! And Zugman (I think) it is ultimately up to the train crew to obey the signals.
Buslistneither one of those constitute dispatcher error! And Zugman (I think) it is ultimately up to the train crew to obey the signals.
That's what I said on here from the start. Kind of what I do.
zugmannBut then you run the risk of crews getting too dependent on a dispatcher. I know most of our dispatchers always try to give a heads up of when you may be stopping, but they're in an office far away. We're the ones staring at the pretty lights along the tracks. Responsibility is all ours.
Buslist schlimm Buslist Because as has been said before, a properly functioning CTC system's field logic would not allow a dispatcher to make such an error. So we would need a malfunctioning signal system and a malfunctioning dispatcher to satisfy you. Looks like Dave Husman and Jeff Hergert have discussed other ways that the accident could occur. Perhaps you can turn your snarkiness on them? neither one of those constitute dispatcher error! And Zugman (I think) it is ultimately up to the train crew to obey the signals.
My typo mistake, should have been
And Zugman said it is ultimately ....
BuslistMy typo mistake, should have been And Zugman said it is ultimately ....
It's all good.
Our community is FREE to join. To participate you must either login or register for an account.