ATSB - Safety Actions for casa and airservices

[Up-into-the-air]

ATSB TRANSPORT SAFETY INVESTIGATION REPORT

Aviation Safety Research and Analysis Report 20050342 Final: Perceived

Pilot Workload and Perceived Safety of RNAV (GNSS) Approaches

The report made the following Safety Actions from the report:

Safety Actions

As a result of the findings of this study, and from feedback received during the consultation process, the ATSB has made a number of recommendations to enhance the safety of RNAV (GNSS) approaches.

Recommendations to Airservices Australia include:

• A study to determine whether the presentation of information, including distance information, on RNAV (GNSS) approach charts is presented in the most effective way;
• A review of the 21.5% of approaches with segment lengths different from the 5 NM optimum and/or multiple steps to determine whether some further
improvements could be achieved;
• A review of waypoint naming conventions for the purpose of improving readability and contributing to situational awareness; and
• A review of training for air traffic control officers for the purpose of ensuring clearances for RNAV (GNSS) approaches are granted in a timely manner.

Recommendations to CASA include:

• Further research to better understand factors affecting pilot workload and
situational awareness during the RNAV (GNSS) approach; and
• A review of training for pilots for the purpose of ensuring clearances for RNAV (GNSS) approaches are granted in a timely manner.

The question is:

What progress have casa made???

What progress have airservices made???

As one is a joint safety action, what has casa done to ensure that the airservices safety action has been met??

What is the time frame??? being undertaken??

[Selcalmeonly]

Good questions - these go to the bigger issue of transparency of CASA's responses to ATSB recommendations, Inquiry recommendations, Coroner Recommendations, etc. Do they post these on a web site somewhere?

If the ATSB are an important link in the aviation safety chain, their recommendations and subsequent responses and progress to completion should have transparency!

[flying-spike]

You will find if you go back to the safety recommendations to CASA arising from the Lockhart River investigations that some of those have not been carried out either.
It can be said that the only good thing that comes out of these tragic accidents is that they present us with the opportunity to learn. It is damnable that this opportunity is squandered.

[Flying Binghi]

Interesting reading...

"...Three thousand five hundred Australian pilots with an RNAV (GNSS) endorsement were mailed a questionnaire asking them to rate their perceived workload, situational awareness, chart interpretability, and safety on a number of different approach types...

...For pilots operating Category A and Category B aircraft (predominantly single and twin-engine propeller aircraft), the RNAV (GNSS) approach resulted in the highest perceived pilot workload (mental and perceptual workload, physical workload, and time pressure), more common losses of situational awareness, and the lowest perceived safety compared with all other approaches evaluated, apart from the NDB approach..."




Perceived Pilot Workload and Perceived Safety of RNAV (GNSS) Approaches (http://www.atsb.gov.au/publications/2006/20050342_RNAV.aspx)





.

[T28D]

CASA does its best to ignore the ATSB after all they don't want to be bothered by an organisation that deals in facts now do they, and an organisation that has more credible expertise than CASA is obviously a threat.

Best ignore them !!!!!!!!!!!!!!!!

[alphacentauri]

Tin hat on....an alternative view

So NDB approaches have a percieved less level of safety than an RNAV approach.....and yet no one bitches about them. Where are the calls to remove NDB reversal procedures or make them more safe? Everyone continues to fly them with no issues.

The only issue most people have with RNAV approaches is that they should show distance to missed approach point. Whilst I agree with this proposal, it cannot be done. The RNAV box can only show distance to one waypoint at a time....how do we get you to fly multiple legs of an approach whilst also showing you the distance to the missed approach point? The RNAV box just can't display this information. Alot of people have posted on this forum claiming reduced level of safety on this issue....but no one has come on here with a solution to the problem. It is more than simply publishing the figures because if you can't see those figures on the RNAV box, what use are they to you?

RNAV approaches (standard and non standard) are flown every day with out safety incidents. In fact lately most of the safety incidents seem to be coming from people trying to fly a VOR/DME approach (YMML and YBCG incidents come to mind) which according to the study have a perceived higher level of safety.

I'm no safety expert but I would be leaning toward approach familiarity and complacency being a bigger issue, than the percieved complexity of RNAV approaches.

Just trying to offer an alternative viewpoint on the "RNAV approaches are unsafe" argument. I find this argument to be incorrect. I really do think that lack of education in conducting RNAV approaches is where the fault lies.

Alpha

[Capn Bloggs]

So NDB approaches have a percieved less level of safety than an RNAV approach.....and yet no one bitches about them. Where are the calls to remove NDB reversal procedures or make them more safe? Everyone continues to fly them with no issues.
NDBs were also created a decades ago. One would have expected that technology of the type used in GPS NPAs would have resulted in a "decades safer" approach.

In fact lately most of the safety incidents seem to be coming from people trying to fly a VOR/DME approach (YMML and YBCG incidents come to mind) which according to the study have a perceived higher level of safety.

That's because they are easier to fly. Get on the radial, there's the distance to the field (all the way from the IAF), just fly down the single profile and you're in. And don't confuse OS Intl crews with Oz pilots; it was a survey of Oz pilots, after all.

The RNAV box can only show distance to one waypoint at a time....how do we get you to fly multiple legs of an approach whilst also showing you the distance to the missed approach point? The RNAV box just can't display this information.
Of course it can. Ever noticed the flight plan miles to run? The boxes just needed to be programmed to display the distance to Mike on the main page as well as the next waypoint. Foxtrot doesn't need to be there. The scaling changes before Foxtrot, and it's a Flyby waypoint so IMO it's not needed. Not having it would solve all the dramas (Ozexpat incoming :ouch:).

[alphacentauri]

No, it can't. The Arinc code that is behind Rnav approaches doesn't support this functionality. Icao rules specific there must.be an intermediate segment, so yes the fix before the faf must be there.

The only way to implement your proposal is to make every Rnav approach have only one leg. From Iaf to mapt. The icao rules prohibit this and the current coding rules prohibit this. You then also lose the flexibility of being able to offer multiple initials.

Wrt your comments on ndb, that's exactly my point. In 40 yrs do you think we will be having this discussion about Rnav approaches? No, because we would have had 40 to get to know them

[Capn Bloggs]

Icao rules specific there must.be an intermediate segment, so yes the fix before the faf must be there. The only way to implement your proposal is to make every Rnav approach have only one leg. From Iaf to mapt. The icao rules prohibit this and the current coding rules prohibit this. You then also lose the flexibility of being able to offer multiple initials.

No, India is not the IAF. India is the Intermediate Fix. The IAFs would continue to happily exist as they are now. VOR approaches have FAFs in thin air, so there appears no impediment to not having the FAF as a waypoint in the box on that score, just use a distance to Mike like any other FAF.

[JCJ]

From a different perspective, I think part of the perceived increase in workload also comes from the various types of equipment types out there. Many of us fly more than one machine, so are exposed to different operating procedures for different units. Although the basic information is the same, the way it is displayed and the user interface is different, and I know when I am operating the least familiar unit, my workload increases. and I trust the unit less.

NDB, and VOR presentations haven't changed since Adam was a boy, so going from one aircraft to another is more seamless.

Without more standards in user interfaces, and presentations this will not go away.

This all of course has nothing to do with the way the info is displayed on the charts. Just my 2c

[alphacentauri]

The rules for coding the waypoints on an RNAV, specify that you must have at least 1 IAF, then IF, then FAF...etc. You must have them. I understand your proposal, but by removing the FAF the box would only countdown to the MAPt from the IF and what then if the descent starts in between the IAF and the IF?

As I said before the solution is to have an IAF-MAPt leg (like the straight segment on the VOR/DME approach) But current coding rules prohibit this.

The reason for the current rules is because the RNAV performance changes for each segment and its performance tolerances decrease as you get closer to the MAPt, thus allowing us to get low/useable minimas. If you were to have an IAF-Mapt leg then the performance tolerance would be quite large all the way up to the MAPt and then minimas would inevitably go up.

Yes I agree that maybe the criteria for RNAV should be looked at amended....but its not going to happen. Too many of them have been designed under the current philosophy. And we are talking about a world wide coding rules change.

[Sarcs]

This is one of the most concerning paragraphs in the findings:


Approach chart interpretability was assessed as more difficult for the RNAV (GNSS) approach than all other approaches by respondents from all aircraft performance categories. Unlike the non-directional beacon (NDB) and ILS
approach charts, ease of interpretation did not increase with the number of approaches conducted per year.


Kind of shows that even the most experienced respondents had stuffed up one of these approaches! So no longer does practise 'make the master'!:rolleyes:

[Sarcs]

The only way to implement your proposal is to make every Rnav approach have only one leg. From Iaf to mapt. The icao rules prohibit this and the current coding rules prohibit this. You then also lose the flexibility of being able to offer multiple initials.



So it is the ICAO rules that prohibit this, not that it can't be done?

Take an FMS aircraft conducting a overlayed ILS via a DME Arc, the aircraft will fly a perfect arc all the way to the Localizer intercept.

Presumably the arc is made up of lots of GNSS waypoints with straight line tracks between the waypoints, yet the only waypoints depicted are the intitial to the arc and the lead bearings with a 'track miles to run' via the arc and localizer. So surely that shows that it can be done! :ok:

[framer]

Icao rules specific there must.be an intermediate segment, so yes the fix before the faf must be there.


Maybe we are in the process (slow as it might be) of changing a rule that worked well for one era but has past it's use by date.
Counting down to intermediate way points destroys S.A. and reduces safety, simple as that.

[Capn Bloggs]

The reason for the current rules is because the RNAV performance changes for each segment and its performance tolerances decrease as you get closer to the MAPt, thus allowing us to get low/useable minimas. If you were to have an IAF-Mapt leg then the performance tolerance would be quite large all the way up to the MAPt and then minimas would inevitably go up.
As I mentioned earlier, on one of the boxes I flew, the scaling changed gradually 2nm before the FAF. The system does not need a physical waypoint to know when to change scaling.

Yes I agree that maybe the criteria for RNAV should be looked at amended....but its not going to happen. Too many of them have been designed under the current philosophy. And we are talking about a world wide coding rules change.
So despite unbelievable progress in electronics in the last few years, we are condemned to a fundamentally flawed design forever. The only saving grace will be that in 40 years time, most aircraft will be fitted with VNAV and Nav Displays so that pilot situational awareness won't be so bad...

[Sarcs]

So despite unbelievable progress in electronics in the last few years, we are condemned to a fundamentally flawed design forever

Perhaps if the ATSB had done the survey and report while the approach was being trialled, we wouldn't have ended up with a 'fundamentally flawed' approach. Kind of like the poms did (see link):

http://www.caa.co.uk/docs/33/2007_06.pdf

Instead it was rushed through and as a consequence has been causal to people being killed!:ugh::ugh:

[Up-into-the-air]

So despite unbelievable progress in electronics in the last few years, we are condemned to a fundamentally flawed design forever. The only saving grace will be that in 40 years time, most aircraft will be fitted with VNAV and Nav Displays so that pilot situational awareness won't be so bad...

This is really the key issue and the question is:

In it's desperate quest to be "world leader" in GPS, has casa put a band aid over a basically flawed system ???

[Sarcs]

The only saving grace will be that in 40 years time, most aircraft will be fitted with VNAV and Nav Displays so that pilot situational awareness won't be so bad...

So CB what you are saying is the world's aviation regulators are waiting for technology to overtake the 'fundamental flaws' of these approaches.

Gee that makes a lot of sense!:ugh::ugh:

[alphacentauri]

Bloggsy,
I think you'll find most boxes start the scale change before the faf, but the change in scaling occurs where it does because there is a faf there. Take the faf away, no change in scaling. That is how the protection area for rnavs are designed coded. With a reduction in scaling comes a reduction in protection area and usually a low/usable minina

[alphacentauri]

Sarcs, the coding rules for the Rnav box also prohibit this. The box can only go to one appoint at a time. Rnav coding is really only based around simple point to point navigation.

And no, for an arcing procedure it is not a series if small straight segments. It's similar to a fixed radius turn.