European Train Control System - Software issue End of Authority/Limit of Authority trip

[Daytona]

It looks as if dumb software design is leading to drivers triggering a Signal Passed At Danger procedure just because they've acknowledged an End of Authority/Limit of Authority trip, before giving the error message to a signaller.

I'm non railway ex-IT, and it screams software design issue to me. Since it's important, the message should not be capable of being cleared by the usual method of pressing acknowledge. There should be some other method - ideally the signaller doing it remotely.

This video shows the correct procedure for drivers and signallers following a trip when the train is travelling in ETCS Level 2

 

[OscarH]

Ah yes, that video. Absolutely insane software design, even if the driver could reset it like that it would be fine if the error could still be accessed.

One button press and gone forever is just dumb

 

[D365]

It doesn’t help that some groups of train operators are insistent on combining the ETCS and AWS acknowledgement into one button. I am especially concerned about the risk of messages being missed by inadvertent driver acknowledgement, when transitioning in and out of Level 2.

 

[bahnause]

ETCS in general is a horrible design for the user. This seems to be a procedural issue however as the rest of europe doesn‘t seem to have an issue with acknowledging a trip before contacting the signaller.

 

[Ashley Hill]

What a poor system. Messages past and present should be stored so they can be accessed as required.

 

[Camberman]

Could the message displayed be amended to include a warning not to delete it until the driver has spoken to the signaller and followed their instructions?

 

[Harpo]

For those unaware of it, the user interfaces to ETCS are bespoke and not standardised.

However, I’m not aware of how many supplier options there are for the drivers’ interface kit.

 

[a_c_skinner]

I too am non rail and from a risk averse profession and that is a very poor design. At the very least it should change to a contact signaller prompt and an OK/Cancel option

 

[ainsworth74]

This seems a classic case of "because we cannot make this more user friendly for reasons we will instead make the users life more difficult". Crackers. If you have to change the software then change the ruddy software to make it user friendly. Perhaps, for instance, by simply having a log that can be accessed of previous messages that have been displayed. That would be a start. What a barmy situation.

The tone of the video is also quite grating as everyone acts as if it's perfectly normally for us to be using such a rubbish piece of design as if it's actually all fine and just you silly drivers making silly mistakes. Rather than we've rolled out a rubbish piece of software design and now you all have to deal with it.

 

[HSTEd]

Shouldn't this message be recorded somewhere for analysis after an incident like an accident?

Or would this message simply be lost forever if the driver pressed a certain button between a quasi SPAD and an incident occurring?

This seems a very peculiar design choice.

 

[apbj]

Now, this nonsense IS the sort of a thing a union should not be allowing.

 

[D365]

Shouldn't this message be recorded somewhere for analysis after an incident like an accident?

One would expect that messages are recorded on the JRU (Judicial Recorder Unit) via a serial link.

 

[MarkyT]

One would expect that messages are recorded on the JRU (Judicial Recorder Unit) via a serial link.

No doubt, but that would not be immediately accessible to front line ops staff dealing with the incident. Teasing the driver with a flashing acknowledge button that immediately removes the vital message that would have enabled a speedy reset means the train is now stuck while incident investigation is escalated, which could be a major source of delay in the TL core for goodness sake. I hope this is 'just' a poor design decision on this particular family of trains that might be rectified in future software updates, rather than a specified functionality for future UK ETCS cabs! Is functionality the same for 158s and new trains on the Cambrian?

 

[TurboMan]

Is functionality the same for 158s and new trains on the Cambrian?

It's the same on HEx 387s on GWML and has been flagged to drivers in much the same way as in the video in the first post.

ETCS in general is a horrible design for the user. This seems to be a procedural issue however as the rest of europe doesn‘t seem to have an issue with acknowledging a trip before contacting the signaller.

Is that because RV mode is permitted elsewhere in Europe? It isn't in the UK.

 

[Fincra5]

Good news.... In v3.6.0 of the ETCS Software, the Trip Messages should remain on the DMI in most cases.

 

[Taunton]

The tone of the video is also quite grating as everyone acts as if it's perfectly normally for us to be using such a rubbish piece of design as if it's actually all fine and just you silly drivers making silly mistakes. Rather than we've rolled out a rubbish piece of software design and now you all have to deal with it.

Generally it's OK, but the two commentators, starting with the one that says sneeringly "if the driver does it wrongly", going on to imply what a nuisance it makes it for the back-office bureaucrats, would in any commercial business get picked up by HR for inappropriate commentary on fellow employees.

 

[MarkyT]

Generally it's OK, but the two commentators, starting with the one that says sneeringly "if the driver does it wrongly", going on to imply what a nuisance it makes it for the back-office bureaucrats, would in any commercial business get picked up by HR for inappropriate commentary on fellow employees.

And it's a pain for the driver who has to remain stationary possibly in a dark tunnel between stations until relieved somehow and then no doubt go through a series of interviews and tests. And for the signaller who has to set in train the tsunami of processes that a potential SPAD type event triggers and try to manage the inevitable major delays that will ensue. Stations may even have to be closed if this occurs at peak times to prevent overcrowding. Most of all its a total PITA for thousands of passengers now stuck in the tunnels or on the street if station entrances are closed, now all looking for alternative ways to get where they're going.

 

[bahnause]

Is that because RV mode is permitted elsewhere in Europe? It isn't in the UK.

As far as I understand it, there is no technical reason for the difference.

 

[MarkyT]

Good news.... In v3.6.0 of the ETCS Software, the Trip Messages should remain on the DMI in most cases.

Thank goodness for that! How quickly is that likely to be rolled out on current fleets.

 

[TurboMan]

Thank goodness for that! How quickly is that likely to be rolled out on current fleets.

I doubt it will be, if the software version on use on existing fleets is still compatible with the trackside. The HEx 387 fleet (introduced in 2020) uses v3.4.0, even though v3.6.0 has been around since 2016, so I can't see there being any rush to upgrade to 3.6.0.

 

[Fincra5]

I doubt it will be, if the software version on use on existing fleets is still compatible with the trackside. The HEx 387 fleet (introduced in 2020) uses v3.4.0, even though v3.6.0 has been around since 2016, so I can't see there being any rush to upgrade to 3.6.0.

Depends on the operator. As GTR are planned to be going to 3.6.0.

 

[Taunton]

And it's a pain for the driver who has to remain stationary possibly in a dark tunnel between stations until relieved somehow and then no doubt go through a series of interviews and tests. And for the signaller who has to set in train the tsunami of processes that a potential SPAD type event triggers and try to manage the inevitable major delays that will ensue. Stations may even have to be closed if this occurs at peak times to prevent overcrowding. Most of all its a total PITA for thousands of passengers now stuck in the tunnels or on the street if station entrances are closed, now all looking for alternative ways to get where they're going.

I only meant to comment on the attitude of the presenters, not the content, but if that all happens because of the way that the various different regulations have been pasted together inappropriately, it should really come to the attention of the media. The Daily Mail is normally good for stories against the railway management.

However did ASLEF allow this stupidity to happen to their members?

 

[JordR]

The tone deaf presentation of this is somewhat compounded by making the video publicly available on Youtube too. I don't think there's any good reason to do that?

I also agree with the above, unbelievably bad UI to have a green flashing button that you're not meant to press!

 

[godfreycomplex]

The tone deaf presentation of this is somewhat compounded by making the video publicly available on Youtube too. I don't think there's any good reason to do that?

It’s the quickest and easiest way to share the process with interested parties

 

[JordR]

It’s the quickest and easiest way to share the process with interested parties

Perhaps, but most large companies don't publish their training materials. Particularly modules that seem to have been created solely to address a presumably recurring problem that call their own staff out for doing things "wrongly" within.

 

[800301]

I doubt it will be, if the software version on use on existing fleets is still compatible with the trackside. The HEx 387 fleet (introduced in 2020) uses v3.4.0, even though v3.6.0 has been around since 2016, so I can't see there being any rush to upgrade to 3.6.0.

It’s taken them long enough to get the DOO camera’s to show when stationary on a 387 and even the ones that have been done sometimes glitch out so I can’t see them being in any rush to upgrade the software as no doubt it will introduce new bugs that will take time to work out, it took long enough to fix the fault that required an ETCS reset before departure at Paddington.

It’s a real shame in the grand scheme of things, I think ETCS has the potential to be a great system but there seems to be a endless problems at each hurdle and no doubt no one wanting to spend the money to implement it quickly. It’s clear it’s a bit rough around the edges for the driver interface but I haven’t heard of any drivers being consulted on what features they would like

 

[Horizon22]

It’s taken them long enough to get the DOO camera’s to show when stationary on a 387 and even the ones that have been done sometimes glitch out so I can’t see them being in any rush to upgrade the software as no doubt it will introduce new bugs that will take time to work out, it took long enough to fix the fault that required an ETCS reset before departure at Paddington.

Those issues seemed to go on for ages. TfL units seemed to have a similar spate when 345s went down towards Heathrow but they seemed to be fixed within a much shorter time.

 

[D365]

It’s a real shame in the grand scheme of things, I think ETCS has the potential to be a great system but there seems to be a endless problems at each hurdle and no doubt no one wanting to spend the money to implement it quickly. It’s clear it’s a bit rough around the edges for the driver interface but I haven’t heard of any drivers being consulted on what features they would like

BIB - I disagree. Operators and consulting firms are more than willing to spend money for as long as the ’taps’ remain open.

 

[cool110]

I also agree with the above, unbelievably bad UI to have a green flashing button that you're not meant to press!

It gets worse, multiple icons can appear in that space. Some like that you can't press without permission from the signaller, and others you do have to press immediately as they act like AWS warnings.

 

[Fincra5]

The tone deaf presentation of this is somewhat compounded by making the video publicly available on Youtube too. I don't think there's any good reason to do that?

I also agree with the above, unbelievably bad UI to have a green flashing button that you're not meant to press!

The Acknowledgement Icons (Flashing) are Yellow not Green.

You do need to press said Icon following a TRIP (not always a SPAD) and there's a procedure for the communication required with the signaller.