Sydney Airport problems again

[RodH]

From the ABC news web site.
Sydney domestic departure flights delayed reportedly due to air traffic control system
Updated 5 minutes ago

Some flights out of Sydney's domestic terminal are being delayed reportedly due to a problem with air traffic control systems.

All airlines have been affected, including Qantas, Jetstar and Virgin.

Hundreds of passengers are facing delays, including families heading out of Sydney for the start of the school holidays.

Arrivals have been able to land but departures have been delayed.

More to come.

First posted 25 minutes ago

[Checklist Charlie]

Flight Planning software not aware of the change to Sydneys ICAO designator.

CC

[RodH]

What is the change then ?

[RodH]

Air Services Australia said there has been a "system software failure".

Follow
Sydney Airport ✈️ ✔ @SydneyAirport
Flights are delayed due to an @AirservicesNews system issue. Please check with your airline for flight status. Thanks for your patience.
7:24 AM - Sep 25, 2017

The ABC understands the fault relates to the flight planning system. This feeds into the radar picture and without it, controllers cannot see which plane is which.

It is also believed the data connection between Sydney and Melbourne is down.

The Qantas departure board lists flights to Melbourne, Brisbane and Adelaide "delayed due ATC Radar Failure".

[Transition Layer]

Flight Planning software not aware of the change to Sydneys ICAO designator.

CC
Haha, very good 😂

[Advance]

IF an incident like this were to be properly investigated, a number of systemic faults would be revealed.
The whole reason for using radar is that you can "move" more aircraft in a given airspace with radar than you can do without it; you can reduce the separation standards to 3 miles because you can "see" the aircraft on the screen.
BUT precisely because you are reducing the separation standards, the safety case of the use of radar must have a whole raft of safety mitigations built in so that no single failure can make aircraft less safe.
Hence SYD is served by at least two different radar sensors which normally feed a common radar display processor but can also feed directly to each separate ATC console.
Duplicated power supplies, duplicated communications paths etc etc exist to protect against failure.
Airservices is supposed to maintain a safety case that demonstrates how safety is maintained in the event of the inevitable failures of component parts of the system. CASA are supposed to audit Airservices to ensure compliance.
It follows that this "incident" demonstrates a failure of the safety system and of its oversight.
But will ATSB, under the leadership of an Ex-ATC safety manager, truly get to the bottom of this foul up?

[parishiltons]

IF an incident like this were to be properly investigated, a number of systemic faults would be revealed.
The whole reason for using radar is that you can "move" more aircraft in a given airspace with radar than you can do without it; you can reduce the separation standards to 3 miles because you can "see" the aircraft on the screen.
BUT precisely because you are reducing the separation standards, the safety case of the use of radar must have a whole raft of safety mitigations built in so that no single failure can make aircraft less safe.
Hence SYD is served by at least two different radar sensors which normally feed a common radar display processor but can also feed directly to each separate ATC console.
Duplicated power supplies, duplicated communications paths etc etc exist to protect against failure.
Airservices is supposed to maintain a safety case that demonstrates how safety is maintained in the event of the inevitable failures of component parts of the system. CASA are supposed to audit Airservices to ensure compliance.
It follows that this "incident" demonstrates a failure of the safety system and of its oversight.
But will ATSB, under the leadership of an Ex-ATC safety manager, truly get to the bottom of this foul up?
The information provided so far does not necessarily suggest a failure of surveillance feeds, but is pointing to other factors. I can't see any safety failures here, ATC has prima facie responded to a reduction in system capability and capacity by moving traffic at rates that are safe commensurate with the status of the ATC system at any given point in time.

[Advance]

The information provided so far does not necessarily suggest a failure of surveillance feeds, but is pointing to other factors. I can't see any safety failures here, ATC has prima facie responded to a reduction in system capability and capacity by moving traffic at rates that are safe commensurate with the status of the ATC system at any given point in time.

Sir, you are missing the point. It does not matter in the slightest what actually failed. What matters is that a failure has occurred such the the system is so degraded that ongoing capacity is now limited - as you point out. But at the time of failure, there could have been more aircraft in the airspace around Sydney than could safely be managed without the radar. Had the failure happened at 0820 instead of 0520 for example?
The safety management system and its oversight is intended to ensure that no single failure (or failure 'chain' consequent on a single failure) can prejudice the safety of any aircraft.

The very fact that traffic handling is now limited is proof that whatever fault did occur, so limited traffic handling that aircraft in the airspace between the time of failure and the time at which the traffic numbers were reduced suffered a reduction in safety assurance.

The safety system and its oversight is intended to preclude that situation.

[4 Holer]

Australia, the best International Airport in a 3rd world country...White Africa.

[parishiltons]

Sir, you are missing the point. It does not matter in the slightest what actually failed. What matters is that a failure has occurred such the the system is so degraded that ongoing capacity is now limited - as you point out. But at the time of failure, there could have been more aircraft in the airspace around Sydney than could safely be managed without the radar. Had the failure happened at 0820 instead of 0520 for example?
The safety management system and its oversight is intended to ensure that no single failure (or failure 'chain' consequent on a single failure) can prejudice the safety of any aircraft.

The very fact that traffic handling is now limited is proof that whatever fault did occur, so limited traffic handling that aircraft in the airspace between the time of failure and the time at which the traffic numbers were reduced suffered a reduction in safety assurance.

The safety system and its oversight is intended to preclude that situation.
Beg to disagree, sorry. As I wrote, the information available so far does not point to a failure of surveillance feeds. What happened was a system degradation (whose full nature and extent is yet to be revealed) at Sydney that resulted in a response to limit traffic to rates that could be safely managed commensurate with the extant state of the system. So of course capacity is reduced, what else could be expected? This response is all planned, documented and obviously has had the appropriate safety, risk and audit rulers run over them, both internally and by CASA. It's just a matter of pulling out the checklist and initiating the appropriate response to a system degradation.

Reduced traffic volumes do not mean a reduction in safety. In contrast, traffic volumes would be reduced as required to assure the maintenance of safety assurance in the absence of some element of ATC's technology.

As for time of day - please bear in mind that Sydney ATC only manages the traffic in the vicinity of Sydney itself and typically only deals with traffic in the flight phase that is within 15-odd minutes from Sydney. The overlying and surrounding airspace is controlled from Melbourne and Brisbane.

[parishiltons]

Australia, the best International Airport in a 3rd world country...White Africa.
We're talking about Sydney here, not Perth!

[Snakecharma]

RodH - your question didnt get answered in the flurry of posts - sydney has not been ASSY for more than 20 years.

Australian airports are Y airports - YSSY - Sydney, YBBN - Brissie etc..

[Advance]

Perhaps I am not explaining the situation correctly.
(Another) system failure has occurred at Airservices Australia ATC to which the ATC response has been to reduce the amount of aircraft being handled so as to restore an appropriate level of safety.
Consider the situation that would exist with the ATC system operating normally in the morning peak period. Then "this" (whatever) failure happens.
If it is then necessary to reduce the amount of traffic in the air to restore a safe operation then by definition, in the time between failure and the actual traffic reduction, taking effect, there was more traffic in the air than could be managed with the requisite level of safety.
If that statement was not true, then why reduce traffic?

[parishiltons]

Perhaps I am not explaining the situation correctly.
(Another) system failure has occurred at Airservices Australia ATC to which the ATC response has been to reduce the amount of aircraft being handled so as to restore an appropriate level of safety.
Consider the situation that would exist with the ATC system operating normally in the morning peak period. Then "this" (whatever) failure happens.
If it is then necessary to reduce the amount of traffic in the air to restore a safe operation then by definition, in the time between failure and the actual traffic reduction, taking effect, there was more traffic in the air than could be managed with the requisite level of safety.
If that statement was not true, then why reduce traffic?
I think this is going off track here. There are endless possible scenarios, of which you mention one. I'll stop and wait for the facts to emerge.

[Advance]

ParisHiltons, perhaps I have not explained my point clearly?
The Sydney ATC system has, this morning, suffered (another) failure to which the ATC response has been to reduce the volume of traffic so as to restore a level of safety.
Now consider the situation that had the fault (whatever it was) happened at 0820 instead of 0520, with maximum aircraft in the airspace.
By definition, in the period between failure and the reduction in traffic being achieved, there would be more traffic in the air than could be safely handled with the reduced functionality available to ATC.
IF THIS STATEMENT IS NOT TRUE, WHY REDUCE TRAFFIC NUMBERS?

Airervices are meant to maintain a safety management system that precludes this happening and CASA are meant to oversee the validity of that system.

[RodH]

RodH - your question didnt get answered in the flurry of posts - sydney has not been ASSY for more than 20 years.

Australian airports are Y airports - YSSY - Sydney, YBBN - Brissie etc..

Thanks snakecharma . I did know that but I made a typo so it's now corrected.
Anyway what is the reply from checklist charlie? Or was he alluding to my typo?

[4 Holer]

Maybe Rocket Man from North Korea hacked you..... Would not be to hard to do to White Africa ??

[RodH]

Managed to correct the typo in the post but not in the List of Aust.NZ etc.
How is that done?

[Snakecharma]

Rod -yes i think he was alluding to the typo

[mikethepomme]

ParisHiltons, perhaps I have not explained my point clearly?
The Sydney ATC system has, this morning, suffered (another) failure to which the ATC response has been to reduce the volume of traffic so as to restore a level of safety.
Now consider the situation that had the fault (whatever it was) happened at 0820 instead of 0520, with maximum aircraft in the airspace.
By definition, in the period between failure and the reduction in traffic being achieved, there would be more traffic in the air than could be safely handled with the reduced functionality available to ATC.
IF THIS STATEMENT IS NOT TRUE, WHY REDUCE TRAFFIC NUMBERS?

Airervices are meant to maintain a safety management system that precludes this happening and CASA are meant to oversee the validity of that system.

I really don't understand your point Advance. Are you suggesting that ASA only manage and allow enough aircraft into the airspace that, in the event of a system failure, those same number of aircraft could be handled in exactly the same way?

That just isn't possible to achieve without the delays you saw today. The entire point of radar and taaats is to allow more aircraft to be controlled than without... so clearly when something breaks, the capacity reduces dramatically and delays occur.

[QSK?]

I think what Advance is trying to convey here is that Airservices' SMS procedures should ensure that any single ATC system component failure should not result in the need for implementation of any reduction in airspace utilisation or capacity to maintain aviation safety. The fact that Airservices had to resort to implementing limits on traffic handling capacity indicates that their SMS approach had failed and that, in order, to continue to maintain adequate safety, traffic handling limits had to be imposed as a last resort. The question Advance is really asking is what was the level of safety available to in-flight pilots between the time the failure first occurred and the time that the traffic handling limits started to become effective.

[Advance]

NO mikethepomme. Rather the reverse. As I pointed out early in the chain, the whole point of having radar is to increase safe capacity beyond that available without radar. But in doing so it must be recognised that any failure may leave an unsafe, overcapacity situation. The safety case is intended to examine failure modes (FMEA) and ensure there are systemic redunancies and excess capacities so that a failure does not necessitate traffic reduction precisely because in the time it takes to actually reduce traffic there is an excess beyond safe capacity.

The FAA actually carried out an extensive study on the subject before they introduced ARTS I - what, back in the '50's.

The latest info in the Sydney newspapers (they must be right??) is that the system failed to "uncombine" from night mode when called upon to do so, leaving but one console working.

A software change is being blamed. So clearly the safety case justifying the change, if it was prepared at all, did not address the failure mode that eventuated.

Sure, systems fail but safety is about ensuing the failure does not increase risk.

[mikethepomme]

Well if the case was an inability to decombine... then there should be no increased risk... as one controller would have been able to handle the workload on one console before the problem. Traffic would then get metered to make sure he could handle it during the problem.

Stopping departures, holding aircraft out etc.

[Capn Rex Havoc]

YSSY is really very ASSY.

[morno]

Maybe they should have tried airplane mode instead

[Ivasrus]

Advance, TAAATS is designed to fallback to 'degraded mode' where the existing radar picture doesn't really change that much, but the controllers lose much of the system automation. This is why one of the the first actions in the quick reference guide is to "stop departures". Sorting out existing airborne traffic, even at maximum capacity, is safely manageable. The system in 'degraded mode' isn't able to easily cope with new flights hence the severe restrictions placed on departing flights as soon as the failure becomes apparent. Consider this akin to "land at closest suitable airport". Hope this helps.

As far as causal factors I dare say that will take some time.

[underfire]

Managed to correct the typo in the post but not in the List of Aust.NZ etc.
How is that done?

Go to edit your post...at the bottom, click where it says go advanced. (the other GA!)

You can edit the title there.

[RodH]

It does change it in the edit function but it still shows up as ASSY in the
Australia, New Zealand & the Pacific List of posts. It is corrected in the post itself but not in the master list.

[underfire]

Power failure in SYD was the cause of it all.

Brilliant!

[sierra5913]

Power failure in SYD was the cause of it all.

Brilliant!

Those damn renewables...

[underfire]

solar powered and shuts off at night.

[Nick_F]

I don't think Sydney is as bad as Brisbane. Or even Melbourne for that matter.

[Duane]

NO mikethepomme. Rather the reverse. As I pointed out early in the chain, the whole point of having radar is to increase safe capacity beyond that available without radar. But in doing so it must be recognised that any failure may leave an unsafe, overcapacity situation. The safety case is intended to examine failure modes (FMEA) and ensure there are systemic redunancies and excess capacities so that a failure does not necessitate traffic reduction precisely because in the time it takes to actually reduce traffic there is an excess beyond safe capacity.

The FAA actually carried out an extensive study on the subject before they introduced ARTS I - what, back in the '50's.

The latest info in the Sydney newspapers (they must be right??) is that the system failed to "uncombine" from night mode when called upon to do so, leaving but one console working.

A software change is being blamed. So clearly the safety case justifying the change, if it was prepared at all, did not address the failure mode that eventuated.

Sure, systems fail but safety is about ensuing the failure does not increase risk.

Advance, you dont have a clue what you are talking about, and obviously have no familiarity with ATC in Australia, in particular Eurocat, perhaps keep your mouth shut lest you make yourself look like a fool.

The issue was of the 2 LAN connectors, when one fails the other takes over, in this case, path A didnt fail gracefully, it kept sending messages, so many in fact that the Sydney partition became overloaded, then the melbourne system, then the brisbane system. It was a very unusual failure, one that hasnt happened before anywhere in the world that has run Eurocat.

What this meant was that positions couldnt be moved split or whatever; resulting in curfew setup continuing beyond curfew period.

Safety was never effected.
Radar was never effected.
The ability to process more than say 10 arrivals an hour was impacted.

Literally zero of your suppositions in this thread that you have raised are even close to correct.

[josephfeatherweight]

Best you learn to spell "affected" correctly: lest you make yourself look like a fool

[Duane]

Well at least I am not speculating and talking out my ass pretending to know ****.

I think that what he is trying to say is, when something fails, ATC should be able to process the exact same amount of traffic as if it hadnt failed. You know like when planes lose an engine they fly exactly the same profile right?

If path 1 (of 2) fails and you are now operating on your backup path, you dont use it the same as if you had redundancy, that is crazy, you dont know why it failed, and you are just going to assume its ok to keep using at max capacity, what if the same failure occurs and you are at max rate.. such a stupid position to maintain. The SMS is to reduce traffic levels until the system returns to normal so you can return to normal levels of traffic. I dont know how you have it in your head that the SMS failed. The SMS that was implemented worked, people got delayed, everyone landed safely who could get a slot, internationals got prioritized short of inconveniencing some people the failure was handled well.

The radar wasnt effected, btw SY gets a mosaic of about 7 radar feeds, but if the TAR gets lost or degraded things will get slow down.

Slowing the rate is a perfectly logical and safe way to implement the SMS, if you are serious that you think that regardless of fault the rate shouldnt change, you are being entirely unrealistic. Planes dont have a back up engine they can whack on the wing when one fails, you cant expect to operate like normal when dealing with a failure. If you need this explained to you any further Advance, you are just being obtuse.

[rjtjrt]

Best you learn to spell "affected" correctly:
He did spell "effected" correctly.
Perhaps you mean to correct his grammatical use of effected vs affected, not a spelling error.

[das Uber Soldat]

Best you learn to spell "affected" correctly:He did spell "effected" correctly.
Perhaps you mean to correct his grammatical use of effected vs affected, not a spelling error.
This boys and girls, is why I don't hang out with pilots.

[IsDon]

This boys and girls, is why I don't hang out with pilots.

And yet here you are. :confused:

[QSK?]

Duane

Was the aggressive tone of your emails really necessary? This forum is successful because, generally, it operates on mutual respect.

If Advance is who I think he is, then you might be surprised as to how much he knows about ATC and Eurocat.

Give respect and you will then receive respect. If you can't do that, then get off these fora

[Duane]

Duane

Was the aggressive tone of your emails really necessary? This forum is successful because, generally, it operates on mutual respect.

If Advance is who I think he is, then you might be surprised as to how much he knows about ATC and Eurocat.

Give respect and you will then receive respect. If you can't do that, then get off these fora

If he did have any clue about how Eurocat works or about how ATC works in Australia, he would not have made the speculative jumps he did, nor insist implementation of a SMS means 100% output capacity. I am sorry but as someone who works the system; to have people with little to no knowledge of what we do come here and put us on blast every couple of weeks gets old, and he is quite clearly in this case, talking about something he knows nothing about.

[parishiltons]

Duane

Was the aggressive tone of your emails really necessary? This forum is successful because, generally, it operates on mutual respect.

If Advance is who I think he is, then you might be surprised as to how much he knows about ATC and Eurocat.

Give respect and you will then receive respect. If you can't do that, then get off these fora

Duane is perfectly correct in everything he has said, and certainly does not come across as aggressive. I'm concerned that Advance progressed a fallacious argument, but respect his/her opinion and right to do so. I don't think that when a logically presented view is presented that it should be suggested that the poster 'get off these fora'. Give respect.

[LeadSled]

Folks,
There are those of us still around who remember when it was ASSY, and who think it should still be, seeing as how the country is called Australia, not Yustralia.
Tootle pip!!

[OzzyOzBorn]

Just curious ... having used around 400 airports worldwide, I can't think of any other which doesn't provide free transfers between passenger terminals. And SYD's options of four bucks for the 400 bus or six bucks for the train are expensive too. Arguably to the extent of being a national embarrassment. SYD must leave a terrible impression on pax caught out by this? Other airports offer all kinds of elaborate terminal transfer arrangements completely free to the customer. Is nobody concerned by the damage to SYD's reputation as a hubbing choice? There are alternatives ...

[thorn bird]

Ozzy, mate, really?
Good grief man, what are you suggesting that OZ airports should be customer-centric?
What about McBank executive bonuses? They have to be funded from somewhere.

[davidclarke]

Just curious ... having used around 400 airports worldwide, I can't think of any other which doesn't provide free transfers between passenger terminals. And SYD's options of four bucks for the 400 bus or six bucks for the train are expensive too. Arguably to the extent of being a national embarrassment. SYD must leave a terrible impression on pax caught out by this? Other airports offer all kinds of elaborate terminal transfer arrangements completely free to the customer. Is nobody concerned by the damage to SYD's reputation as a hubbing choice? There are alternatives ...

Well even the head of IATA say the airport privatisation was a mistake, would you expect anything less?

http://m.atwonline.com/airports-routes/aapa-iata-challenges-privatized-airport-model

[OzzyOzBorn]

Thanks for the link, David. Interesting read.