Errors on departure boards

[enginedin]

How common are errors in the data that gets displayed on departure boards (and other customer information screens), and how are they introduced - are they typically human error or a technical glitch?

For example, this photo shows a departure board at York for a service to Edinburgh showing that Abbeyhill Junction is a passenger stop....

(n.b not my photo! Original photo taken from this Reddit post: https://www.reddit.com/r/Edinburgh/comments/1q1wjnc/abbeyhill_junction_station/)

 

[alistairlees]

This is most likely an issue with the data, not the departure board.

 

[enginedin]

This is most likely an issue with the data, not the departure board.

sorry, when I said "on the departure board" I was using that as the display medium, not an implication that it was an error with the actual departure board. I'll update my post to clarify

 

[D6700]

This is most likely an issue with the data, not the departure board.

Out of curiosity, what could that issue with the data be?

From what I can see, that train shows Activity Codes A, N and * at Abbeyhill Junction, which should fully supress any advertising of the stop. This is reflected in Realtime Trains, which shows it as an unadvertised pathing stop with no public times on display.

I know this sort of thing crops up every so often around the country - perhaps most commonly when there is engineering work affecting services.

The new departure boards at York can behave a little bizarrely. When I get chance, I'll post a recent photo of one such occasion.

Edit: From looking at the link under the photo, I see this example wasn't from today - in fact it was two weeks ago! Looking on RTT, the schedule has been changed since then, so probably was lacking the activity codes to supress the advertising of the pathing stop.

 

[jamieh27]

How common are errors in the data that gets displayed on departure boards (and other customer information screens), and how are they introduced - are they typically human error or a technical glitch?

For example, this photo shows a departure board at York for a service to Edinburgh showing that Abbeyhill Junction is a passenger stop....

(n.b not my photo! Original photo taken from this Reddit post: https://www.reddit.com/r/Edinburgh/comments/1q1wjnc/abbeyhill_junction_station/)

There can be data glitches or errors from time to time, I know GWR is known to have the odd faults now and then i.e. how long a train was delayed was incorrect on the announcements but data was correct. Coach formation was also at times incorrect announcing a IET as 15 coaches when they never done that. Last year the coach formation was wrong announcing a the train had one coach when GWR haven't got 153s.

 

[pokemonsuper9]

There was a discussion recently after Avanti started calling at Liverpool South Parkway and it called them Virgin Trains (presumably noone had updated the database).

 

[800001]

How common are errors in the data that gets displayed on departure boards (and other customer information screens), and how are they introduced - are they typically human error or a technical glitch?

For example, this photo shows a departure board at York for a service to Edinburgh showing that Abbeyhill Junction is a passenger stop....

(n.b not my photo! Original photo taken from this Reddit post: https://www.reddit.com/r/Edinburgh/comments/1q1wjnc/abbeyhill_junction_station/)

Happens quite frequently, usually a simple mistake where a location is selected as a calling point rather than a pathing or passing point when the timetable planners are writing the service.

Happens sometimes on LNER Carlisle diverts, where the train and platform screens show Carlisle London Road as a calling point.

 

[jamieh27]

There was a discussion recently after Avanti started calling at Liverpool South Parkway and it called them Virgin Trains (presumably noone had updated the database).

Also Bushey when LNR services were shown on the boards of London Midland.

 

[Stephen42]

Out of curiosity, what could that issue with the data be?

From what I can see, that train shows Activity Codes A, N and * at Abbeyhill Junction, which should fully supress any advertising of the stop. This is reflected in Realtime Trains, which shows it as an unadvertised pathing stop with no public times on display.

I know this sort of thing crops up every so often around the country - perhaps most commonly when there is engineering work affecting services.

The new departure boards at York can behave a little bizarrely. When I get chance, I'll post a recent photo of one such occasion.

Amended schedules can unintentionally introduce public stops at operational locations. For 2nd January the schedule was amended and lost the not advertised/stop for trains to pass designation with public times added.

This happens fairly frequently, looking at the rest of this month there are two schedules which at one point had a stop advertised there that have since been corrected.

 

[D6700]

Amended schedules can unintentionally introduce public stops at operational locations. For 2nd January the schedule was amended and lost the not advertised/stop for trains to pass designation with public times added.

This happens fairly frequently, looking at the rest of this month there are two schedules which at one point had a stop advertised there that have since been corrected.

Indeed - and thank you for the extra info. I edited (added to) my reply whilst you were writing yours, as, when I viewed the link, I discovered it was an old photo, rather than being one from today.

Regarding my other point about strange behaviour of York's new departure boards, here's the photo I mentioned, which was taken on the 19th of December 2025. The 00:16 to Newcastle is listed 10 times - and with 6 different expected times - 00:23, 00:24, 00:25, 00:28, 00:29 and 00:30!

 

[Gerald Fiennes]

Perhaps passengers were being asked to vote which one they preferred...?!!!

 

[D6700]

Perhaps passengers were being asked to vote which one they preferred...?!!!

It looks like 00:28 was winning the vote!

At least they wouldn't be short of seats, with 90 coaches in less than 10 minutes

 

[NSE]

It often seems to happen when a change is made. It happens on the Marshlink every so often. When a set swap takes place the platforms screens can sometimes show 6 car as the system seems to have had the new one added but not removed the old one.

 

[Stephen42]

Indeed - and thank you for the extra info. I edited (added to) my reply whilst you were writing yours, as, when I viewed the link, I discovered it was an old photo, rather than being one from today.

Regarding my other point about strange behaviour of York's new departure boards, here's the photo I mentioned, which was taken on the 19th of December 2025. The 00:16 to Newcastle is listed 10 times - and with 6 different expected times - 00:23, 00:24, 00:25, 00:28, 00:29 and 00:30!

View attachment 196024

Yes that one seems more likely a York/local CIS problem than data. The train location gets closer to York the further left it is and it's unlikely a train would be duplicated that many times with each having different running information. There's a large gap before the next service after that one. My suspicion would be it's software ghosting, as services move leftward if there's no eligible train to replace it a stale version of the data is left in place rather than being cleared. Without delays it would be a lot harder to tell they were stale with the clock continuously updated.

Would agree that a lot of passenger facing systems don't seem particularly well designed for edge case behaviour.

 

[D1537]

Not just on departure boards ...

 

[D6700]

Another recent oddity at York, but not so visually comical, was on the 3rd of January, when I caught the 09:55 London service. It was running about 10 minutes late, but all of the station departure boards were showing it as expected much later than that. As I had a connection at Doncaster, this made me nervous...

I wondered what they knew that I didn't - but the answer was that they didn't! The train rolled in many minutes before the expected time - so I wondered if it was going to be held for some reason, but the stop was just 4 minutes and we were away. As we left, I looked at one of the platform indicators - and, by this point, the expected time had been brought forward, but was still showing a time in the future, even though the train had already set off.

In the past, the expected time quoted was a decent guide as to when the train would arrive, as long as it didn't get held outside, but on that occasion, it was miles out - and a wrong side failure.

Aside from these oddities, the replacement of the arrivals board with a clock is very disappointing. Other than that and the painfully slow screen changes on the summary departure lists, which mainly affects the smaller board on platforms 6&7, I do actually quite like the new screens!

Not just on departure boards ...

View attachment 196026

Now that is really weird! I can't even think how that could come about...

 

[structos]

Few days ago, a GWR from Cardiff to Portsmouth was stated to call at Cosham on the departure boards, and by automated onboard announcements.
However, the timetable does not have Cosham scheduled as a stop for the 13XX trains.

 

[saismee]

I've seen a few freight services pop up on Blackbox displays recently. Some of their stops show as blank with a time, and display as a comma in the list.

09:44
Acton T.C.
Calling At
(10:18)
(11:11)
Acton T.C. (12:03)

On small boards it says

, , Acton T.C.

And sometimes the destination is blank, so it just appears with the time.

Here's an example from Ipswich, note that platform TL is the through line with no platform face adjacent.

 

[transportphoto]

Another recent oddity at York, but not so visually comical, was on the 3rd of January, when I caught the 09:55 London service. It was running about 10 minutes late, but all of the station departure boards were showing it as expected much later than that. As I had a connection at Doncaster, this made me nervous...

I wonder whether there was an accompanying “this train will be delayed at York and is expected to be X minutes late” Darwin message sent by the control room.

 

[zwk500]

From what I can see, that train shows Activity Codes A, N and * at Abbeyhill Junction, which should fully supress any advertising of the stop. This is reflected in Realtime Trains, which shows it as an unadvertised pathing stop with no public times on display.

The correct code for that stop should be *,A in TPS. The * functions as a suppression indicator to downstream systems that only read the first character, and only has any effect if in first position. The N must be in first place if being used, and is mutually exclusive with the *. If the A is in first place then some systems may not recognise it, and replace it with the default code T, for stops to set down and pick up passengers, which would show on the boards.

N is normally used for stations that haven't opened yet or stops where a specially booked group of passengers are getting on where a service wouldn't normally stop. If the service was making an unadvertised call and needed to be held for pathing reasons it should show N,A. However, 'N' still represents a station call and should not be used at a junction with no platform.

However the way that the TPRs are written (section 1.3.5 at https://www.networkrail.co.uk/industry-and-commercial/information-for-operators/operational-rules/, if anybody wants to look up) is not particularly clear and data errors like this are usually left to the last as planners are focusing on clearing timing conflicts. This is also a factor in why errors are more common around amended schedules.

It's also worth mentioning that some systems are set up to change certain codes on import or export, so codes that were correct can get mangled during bidding or publishing.

 

[Horizon22]

Few days ago, a GWR from Cardiff to Portsmouth was stated to call at Cosham on the departure boards, and by automated onboard announcements.
However, the timetable does not have Cosham scheduled as a stop for the 13XX trains.

Might have been added additionally. Did it stop there or were you not on that train?

 

[voyagerdude220]

The CIS at both Preston and Carlisle recently advertised the Avanti Evero trains running via the Settle and Carlisle line as having First Class in coaches E and L.

This is incorrect- it's in J and K on a 10-car Evero. E and L used to be the case with a 10-car Super Voyager formation.

 

[800001]

The CIS at both Preston and Carlisle recently advertised the Avanti Evero trains running via the Settle and Carlisle line as having First Class in coaches E and L.

This is incorrect- it's in J and K on a 10-car Evero. E and L used to be the case with a 10-car Super Voyager formation.

A mistake where someone has selected the wrong 'tick box' in back office when adding the information to CIS.

 

[Adam Williams]

Regarding my other point about strange behaviour of York's new departure boards, here's the photo I mentioned, which was taken on the 19th of December 2025. The 00:16 to Newcastle is listed 10 times - and with 6 different expected times - 00:23, 00:24, 00:25, 00:28, 00:29 and 00:30!

I think the British Sign Language (BSL) interpreter should be commended for his attitude for his commitment to signing the same service 9 times without complaint, and at such an unsociable time of day as well!

 

[James H]

last week at Waterloo. Image shows scrambled departure board.

 

[pokemonsuper9]

last week at Waterloo. Image shows scrambled departure board.

I'm not sure what's better, the 14.5 coach train to Hampton Court that thinks it's only 5, or the 2 carriage 09:09 to E that thinks it's 8.

 

[zwk500]

last week at Waterloo. Image shows scrambled departure board.

That looks like a column alignment/signal error to the boards themselves. Not a good thing at one of the country's busiest stations!

 

[Peter Mugridge]

That looks like a column alignment/signal error to the boards themselves. Not a good thing at one of the country's busiest stations!

Memories of the Solari boards getting their flaps stuck and as a result showing some fantastic routings!

 

[moogal]

Had a couple of bizarre formations displayed on LNR services on the same day last month. The 16.47 showing as an impossible 11 carriages (it was a triple-350, so 12) on the screen at Leighton Buzzard, then returning from Euston that evening the diagram and text on the same screen disagreed on the number of coaches.

 

[Bletchleyite]

That "formed of 11 carriages" looks like a manual note so someone has probably clicked on the wrong train when they were intending to put that note on a Pendolino service.

I've seen the diagram being different from the text before, it is probably from a different source. Which was correct out of interest, I suspect the diagram?