AF358 at YYZ report to come out
Take a look at how close they came to a far more serious accident.
The GTAA needs to fill in the two ravines at the threshold of 05 and the 06's. The approach lighting towers may be "frangible" but what they're mounted on certainly is not. A dozen knots more groundspeed off the end of 24L and the outcome would have been substantially different.
The GTAA needs to fill in the two ravines at the threshold of 05 and the 06's. The approach lighting towers may be "frangible" but what they're mounted on certainly is not. A dozen knots more groundspeed off the end of 24L and the outcome would have been substantially different.
Join Date: Sep 2000
Location: Bielefeld, Germany
Posts: 955
Likes: 0
Received 0 Likes
on
0 Posts
Originally Posted by PJ2
The GTAA needs to fill in the two ravines at the threshold of 05 and the 06's.
So why don't they, even in relatively wealthy areas?
I think the answer is a combination of history, economics and blind hope.
History. Major airports have typically been in their present locations for decades, even as TO/landing distances for commercial aircraft have risen and the city has expanded out to meet the airport. Relocating an airport is prohibitively expensive. After Denver, no one wants to do it.
Economics. Land around the airport is increasingly expensive to purchase as time goes on; the people who pay for overrun areas are not the same people who benefit from their being there; the overrun area on a specific runway usually only earns its living infrequently (last one into the ravine was almost 30 years ago), so cost per life saved for the local administration is relatively high compared with, say, improving local roads.
Blind hope. They won't do *that* again here, will they? (Indeed, "they" usually don't. While overrun accidents occur regularly, overrun accidents at a specific airport and runway rarely repeat. And when they do, at least in relatively wealthy countries something is done: KLGA and KJFK).
PBL
The button of Vancouver's 08R, (departure on 26L), has far worse hazards:
The "overrun" area is the white concrete section (relatively new). There is a road at the end of the runway which leads into a tunnel. The road descends down into the tunnel on either side of the runway. The jpg below shows the relationships. The overrun area directly ahead has the approach lights for 08R but only after one goes through fencing and into the bog (or possible ocean, if the tide higher in winter).
The runway is the longer of the two parallels but the north runway is never used for departure due to noise complaints from residents across the water along Marine Drive.
On any overrun, one either goes very wide and risks the ditch, fence and bog, (unlikely due to inertia), or goes right down the centerline where there is not enough room to avoid the wings (of a 340) striking the approach lighting if one goes wide enough to clear the fuselage. The drop into the roadway which leads into the tunnel is, at it's deepest, more than the distance from the bottom of a 340 fuselage to the cockpit level.
Overrun areas of the other runways are less of a concern although an substantial overrun on departing 08R would cross a roadway which carries a lot of traffic.
The "overrun" area is the white concrete section (relatively new). There is a road at the end of the runway which leads into a tunnel. The road descends down into the tunnel on either side of the runway. The jpg below shows the relationships. The overrun area directly ahead has the approach lights for 08R but only after one goes through fencing and into the bog (or possible ocean, if the tide higher in winter).
The runway is the longer of the two parallels but the north runway is never used for departure due to noise complaints from residents across the water along Marine Drive.
On any overrun, one either goes very wide and risks the ditch, fence and bog, (unlikely due to inertia), or goes right down the centerline where there is not enough room to avoid the wings (of a 340) striking the approach lighting if one goes wide enough to clear the fuselage. The drop into the roadway which leads into the tunnel is, at it's deepest, more than the distance from the bottom of a 340 fuselage to the cockpit level.
Overrun areas of the other runways are less of a concern although an substantial overrun on departing 08R would cross a roadway which carries a lot of traffic.
Last edited by PJ2; 19th Dec 2007 at 04:04.
Rebel PPRuNer
Join Date: Aug 2000
Location: Toronto, Canada (formerly EICK)
Age: 51
Posts: 2,834
Likes: 0
Received 0 Likes
on
0 Posts
Re: covering the ravine - I seem to recall a Jeremy Clarkson article which suggested replacing safety belts and airbags in cars with a big spike in the centre of the steering wheel, thus acting as a deterrent to bad driving.
More seriously, I feel it would be better to use EMAS rather than adding more concrete to a fairly depressing area, containing as it does several major highways plus Pearson. Maybe the Government of Canada could release some of the security charges and airport rents they are grabbing from Pearson and invest in safety rather than demand it be self-financed.
More seriously, I feel it would be better to use EMAS rather than adding more concrete to a fairly depressing area, containing as it does several major highways plus Pearson. Maybe the Government of Canada could release some of the security charges and airport rents they are grabbing from Pearson and invest in safety rather than demand it be self-financed.
RBF re #38 “Survival odds going off the end at 80 some kt. are much better than in a failed go-around.”
Your statement is very much open to debate; there is no evidence that a go around would have failed. Furthermore, if taken at face value and used in other situations the statement could be hazardous. Compare the result of the YYZ overrun with the accident at Congonhas and then argue against a GA.
There are many ways of representing the YYZ situation; I choose TEM / decision-making.
A critical point was the decision to make the approach. The threats at this time were either ill-defined (information not available) or known but not adequately considered (risk assessment). Contributions to the latter activity would involve human behaviour and industry norms. Thus in hindsight the decision to make the approach in the prevailing conditions was an error, a weakness in judgement, also made by several crews preceding the accident.
The report observed that it is industry standard to make such approaches, i.e. we bear the risk as a normal operation whereas perhaps we should not – industry complacency. Perhaps this is because we don’t understand (or wish to acknowledge) how close to the edge of the safety boundary we operate (yet statistically still remain a very safe industry). For example the 1.96 wet runway landing distance factor may have to be 2.2-2.4 to maintain a level of safety consistent with a dry operation (TC report on landing on wet runway).
The conclusion from a wide range of successful wet operations is that crews adapt and correct for the lower level of safety. However, occasionally crews either do not adapt or get caught out, and it is the solution to preventing these situations which the industry must concentrate on.
During the approach up to the point of touchdown, there was opportunity to correct the ‘error’ by flying a go around. The reason for this option being rejected and continue with the landing reside deep in human behavior, but probably these are similar behaviors to those in the original decision to make the approach, ‘press-on-itis’.
All of the crews landing in the stormy conditions were doing their best, but were let down by the limitations in the availability and accuracy of information describing the situation, the limits of human behavior, and the safety norms and values of modern operations.
The comprehensive report provides a well balanced view of the accident and the human involvement. However, the recommendation for training to help pilots to make better landing decisions in deteriorating weather still begs the question what exactly do we train?
Refs:
Safety aspects of aircraft performance on wet and contaminated runways.
Safety aspects of tailwind operations.
Aircraft Braking Performance on Wet Concrete Runway Surfaces.
Wet Runway Friction: Literature and Information Review.
Managing threats and errors during approach and landing.
Your statement is very much open to debate; there is no evidence that a go around would have failed. Furthermore, if taken at face value and used in other situations the statement could be hazardous. Compare the result of the YYZ overrun with the accident at Congonhas and then argue against a GA.
There are many ways of representing the YYZ situation; I choose TEM / decision-making.
A critical point was the decision to make the approach. The threats at this time were either ill-defined (information not available) or known but not adequately considered (risk assessment). Contributions to the latter activity would involve human behaviour and industry norms. Thus in hindsight the decision to make the approach in the prevailing conditions was an error, a weakness in judgement, also made by several crews preceding the accident.
The report observed that it is industry standard to make such approaches, i.e. we bear the risk as a normal operation whereas perhaps we should not – industry complacency. Perhaps this is because we don’t understand (or wish to acknowledge) how close to the edge of the safety boundary we operate (yet statistically still remain a very safe industry). For example the 1.96 wet runway landing distance factor may have to be 2.2-2.4 to maintain a level of safety consistent with a dry operation (TC report on landing on wet runway).
The conclusion from a wide range of successful wet operations is that crews adapt and correct for the lower level of safety. However, occasionally crews either do not adapt or get caught out, and it is the solution to preventing these situations which the industry must concentrate on.
During the approach up to the point of touchdown, there was opportunity to correct the ‘error’ by flying a go around. The reason for this option being rejected and continue with the landing reside deep in human behavior, but probably these are similar behaviors to those in the original decision to make the approach, ‘press-on-itis’.
All of the crews landing in the stormy conditions were doing their best, but were let down by the limitations in the availability and accuracy of information describing the situation, the limits of human behavior, and the safety norms and values of modern operations.
The comprehensive report provides a well balanced view of the accident and the human involvement. However, the recommendation for training to help pilots to make better landing decisions in deteriorating weather still begs the question what exactly do we train?
Refs:
Safety aspects of aircraft performance on wet and contaminated runways.
Safety aspects of tailwind operations.
Aircraft Braking Performance on Wet Concrete Runway Surfaces.
Wet Runway Friction: Literature and Information Review.
Managing threats and errors during approach and landing.
Apache for HEMS - Strafe those Survivors!
Join Date: Jan 2001
Location: uk
Posts: 120
Likes: 0
Received 0 Likes
on
0 Posts
Just 2 thoughts on this, big hand to the cabin crew, who did a superb job and stopped this from being a disaster; pleased to see AF changed there sops to either crew member being able to call a go around after this event, however slightly concerned about what the previous position says about the cross cockpit authority gradient that could easily exist in such circumstances.
Overall, yes a hull loss, but thank goodness due to a combination of fortune and excellent cabin crew, no lives lost
Overall, yes a hull loss, but thank goodness due to a combination of fortune and excellent cabin crew, no lives lost
Alph2z, Re #40, the problem with assumptions is that they are often wrong.
JAR-OPS 1.550 requires that the landing mass is sufficient for a landing within 60% of LDA (and for a wet runway the distance is 115% of the dry distance). Thus, provided the crew compares landing weight with the max allowable landing weight (wet) for a given runway, an operator meets the regulatory requirement; distance is a by-product of the calculation. (FAR 121.195 is similar ‘… weight of the airplane on arrival…’).
There is a further requirement in JAR for an in-flight check, which is probably more relevant to the circumstances of this accident, vis runway condition, safe landing, and missed approach.
JAR-OPS 1.400 “Before commencing an approach to land, the commander must satisfy himself that, according to the information available to him, the weather at the aerodrome and the condition of the runway intended to be used should not prevent a safe approach, landing or missed approach, having regard to the performance information contained in the Operations Manual.”
The problem which I think that you allude to is that most crews do not understand or cannot determine the margin of safety that is available during the landing which enable the safety requirement of JAR to be met (safety as such is not defined). However, considering safety as the need to minimize risk requires the Commander to have all pertinent information to evaluate the risks in the planned approach and landing, and thus confirm the safety of the operation.
The accident report refers to this as a margin for error; I prefer ‘margin of safety’, a margin in which risk is as far as possible quantified and controlled.
UK AIC 14/2006 refers to this margin as ‘field length factor’ which accounts for the normal operational variability that can be expected in day to day service such that the chances of a landing overrun are remote. Thus, any additional factor or error could reduce the margin of safety below that which we are familiar with in normal operations. The AIC concludes “It cannot be assumed that the scheduled landing distances can accommodate a landing in which all relevant parameters are at the limit of their tolerance in the adverse sense”; which again excludes additional error or adverse factors beyond the crews’ control (i.e. threats to a safe operation).
One of the fundamental problems which crews face is that they do not always have information enabling them to determine the runway condition, wind speed/direction, the precise weather (visibility/rainfall), etc, all of which affect the margin of safety.
Thus in this accident it was possible that from a crew perspective all of the parameters under their control were within normal operational variability (I am not saying that they were or were not, nor I as I interpret, did the report). However, the significant issues involved the late and rapid weather changes which invalidated the crew’s perception and plans, which reduced the margin of safety for operational tolerances to zero (220ft).
It might be debatable that a crew should have identified the potential for a wet contaminated runway at an earlier stage and reconsidered the landing distance (i.e. mass – too heavy for a short wet contaminated runway).
It may be appropriate for operators to require crews to adjust the planned margin of safety before landing in adverse conditions, e.g. tighten the stabilized approach criteria thus forcing an earlier go around, or reduce the landing weight. Unfortunately as illustrated in this accident, these changes still might not accommodate ‘normal’ human behavior (error prone) where the failure to perceive changing weather, runway state, visibility, or a long landing, may remove all of the margin of safety. In these rare events, the industry requires addition safety margins, preferably a good overrun area or increased landing distance requirements.
‘Landing Performance of Large Transport Aeroplanes’ UK AIC, free but registration required; then select PUBS / AIC / Pink.
Also, see AIC 86/2007 ‘Risks and factors associated with operations on runways affected by snow, slush or water’.
Safety aspects of tailwind operations.
JAR-OPS 1.550 requires that the landing mass is sufficient for a landing within 60% of LDA (and for a wet runway the distance is 115% of the dry distance). Thus, provided the crew compares landing weight with the max allowable landing weight (wet) for a given runway, an operator meets the regulatory requirement; distance is a by-product of the calculation. (FAR 121.195 is similar ‘… weight of the airplane on arrival…’).
There is a further requirement in JAR for an in-flight check, which is probably more relevant to the circumstances of this accident, vis runway condition, safe landing, and missed approach.
JAR-OPS 1.400 “Before commencing an approach to land, the commander must satisfy himself that, according to the information available to him, the weather at the aerodrome and the condition of the runway intended to be used should not prevent a safe approach, landing or missed approach, having regard to the performance information contained in the Operations Manual.”
The problem which I think that you allude to is that most crews do not understand or cannot determine the margin of safety that is available during the landing which enable the safety requirement of JAR to be met (safety as such is not defined). However, considering safety as the need to minimize risk requires the Commander to have all pertinent information to evaluate the risks in the planned approach and landing, and thus confirm the safety of the operation.
The accident report refers to this as a margin for error; I prefer ‘margin of safety’, a margin in which risk is as far as possible quantified and controlled.
UK AIC 14/2006 refers to this margin as ‘field length factor’ which accounts for the normal operational variability that can be expected in day to day service such that the chances of a landing overrun are remote. Thus, any additional factor or error could reduce the margin of safety below that which we are familiar with in normal operations. The AIC concludes “It cannot be assumed that the scheduled landing distances can accommodate a landing in which all relevant parameters are at the limit of their tolerance in the adverse sense”; which again excludes additional error or adverse factors beyond the crews’ control (i.e. threats to a safe operation).
One of the fundamental problems which crews face is that they do not always have information enabling them to determine the runway condition, wind speed/direction, the precise weather (visibility/rainfall), etc, all of which affect the margin of safety.
Thus in this accident it was possible that from a crew perspective all of the parameters under their control were within normal operational variability (I am not saying that they were or were not, nor I as I interpret, did the report). However, the significant issues involved the late and rapid weather changes which invalidated the crew’s perception and plans, which reduced the margin of safety for operational tolerances to zero (220ft).
It might be debatable that a crew should have identified the potential for a wet contaminated runway at an earlier stage and reconsidered the landing distance (i.e. mass – too heavy for a short wet contaminated runway).
It may be appropriate for operators to require crews to adjust the planned margin of safety before landing in adverse conditions, e.g. tighten the stabilized approach criteria thus forcing an earlier go around, or reduce the landing weight. Unfortunately as illustrated in this accident, these changes still might not accommodate ‘normal’ human behavior (error prone) where the failure to perceive changing weather, runway state, visibility, or a long landing, may remove all of the margin of safety. In these rare events, the industry requires addition safety margins, preferably a good overrun area or increased landing distance requirements.
‘Landing Performance of Large Transport Aeroplanes’ UK AIC, free but registration required; then select PUBS / AIC / Pink.
Also, see AIC 86/2007 ‘Risks and factors associated with operations on runways affected by snow, slush or water’.
Safety aspects of tailwind operations.
there is no evidence that a go around would have failed
I beg to disagree. I experienced what I suspect was the same thunderstorm that AFR358 decided not to enter on a go-around. It brought highway traffic to a standstill; another cell that same hour washed out a major road bridge. I would not like to imagine flying any airplane in those conditions. I had also been following the weather radar updates; it's the only time I've seen a purple return.
Unfortunately TSB Canada has not yet seen fit to release the weather images that it obtained.
Join Date: Feb 2007
Location: Not here
Posts: 222
Likes: 0
Received 0 Likes
on
0 Posts
Originally Posted by alph2z
....... To come to a full stop they would have needed an extra 1460 feet of concrete beyond the threshold lights........
I just found some other data hidden away in the TSB report after reading almost all of it. Doing some simple subtraction of their 3 numbers and I get 1474 ft .
Extra runway required beyond threshold
= touch down pt + rolling distance required - runway length
1474 ft = 3800 + 6674 - 9000
In other words I was off by only 14 ft (or 1%) !!!!!
Not bad, not bad.
.
Join Date: Dec 2007
Location: france
Age: 75
Posts: 74
Likes: 0
Received 0 Likes
on
0 Posts
Join Date: Apr 2005
Location: uk
Posts: 260
Likes: 0
Received 0 Likes
on
0 Posts
As has been pointed out, the information regarding moderate to severe weather, is rarely transmitted to crews as accurately as it could be to make critical decisions. What the driving conditions are in the vicinity are never made known until after touchdown and I would dispute their total relevance. However, an improved method of communicating such vital information is required and essential.
This was a very experienced and competent crew at the outset of the approach. What part fatigue ( Captain's history of reduced flying due to fatigue and co-pilot's tiredness from a late-night simulator session, with minimum rest period) played in the ensuing mind-set to continue the approach with the intention of landing as opposed to going around and diverting with (uncormfortably) low fuel state, can be argued ad infinitum.
The difficulty in how to ensure a non-reoccurence by training, is how do we 'train-out' mind-set conditions? How many of us professional pilots would have fared any better, in making a difficult approach at the end of a long duty period as two-crew operation?
As for calculating landing distance in such conditions, the problem is that it only gives you a distance based on assumed conditions, including wear and tear of brakes and runway surfaces. This distance is than weighed up against the individual's previous experience of landing in similar conditions and the decision to continue or not is arrived at. Do not forget that in using such information, precise skills are required, such as speed control, touchdown point, application of brakes and selection of reverse.
I do not know the recommendations Air France have used to alter their SOP's, but will make the observation( which applies to Southwest's Midway accident and Qantas's accident in Bangkok) that if the PF is delegated as the reverse selector, then the PM must be prepared to call non-selection after touchdown immediately. Some airlines delegate the selection of reverse to the PM, on PF's command - it would not have prevented the overrun in this situation, primarily because of the water contamination and the deep landing.
This was a very experienced and competent crew at the outset of the approach. What part fatigue ( Captain's history of reduced flying due to fatigue and co-pilot's tiredness from a late-night simulator session, with minimum rest period) played in the ensuing mind-set to continue the approach with the intention of landing as opposed to going around and diverting with (uncormfortably) low fuel state, can be argued ad infinitum.
The difficulty in how to ensure a non-reoccurence by training, is how do we 'train-out' mind-set conditions? How many of us professional pilots would have fared any better, in making a difficult approach at the end of a long duty period as two-crew operation?
As for calculating landing distance in such conditions, the problem is that it only gives you a distance based on assumed conditions, including wear and tear of brakes and runway surfaces. This distance is than weighed up against the individual's previous experience of landing in similar conditions and the decision to continue or not is arrived at. Do not forget that in using such information, precise skills are required, such as speed control, touchdown point, application of brakes and selection of reverse.
I do not know the recommendations Air France have used to alter their SOP's, but will make the observation( which applies to Southwest's Midway accident and Qantas's accident in Bangkok) that if the PF is delegated as the reverse selector, then the PM must be prepared to call non-selection after touchdown immediately. Some airlines delegate the selection of reverse to the PM, on PF's command - it would not have prevented the overrun in this situation, primarily because of the water contamination and the deep landing.
Join Date: Jul 2007
Location: home
Posts: 100
Likes: 0
Received 0 Likes
on
0 Posts
What would be the rate of climb of an A340 at that mass? On TO they don't climb very well (like other quads but the 345). Could they have outperformed a 6000 ft/min downburst?
Join Date: Aug 2007
Location: London, New York, Paris, Moscow.
Posts: 3,632
Likes: 0
Received 0 Likes
on
0 Posts
Well firstly I will state my non flying pedigree, is that OK?
Looking at the report from a non expert point of view I think the whole accident can be surmised in the saying
"sometimes you get handed the ****ty end of the stick"
The crew were left to make snap and instinctive decisions that on face value led to an overrun. However once presented with the full facts as known (and importantly some unknown to the crew) they "almost" got away with it.
Perhaps it would be better to focus on how the information reqd can be better disseminated to crews approaching this airport in the future?
rgds
Glad Rag.
Looking at the report from a non expert point of view I think the whole accident can be surmised in the saying
"sometimes you get handed the ****ty end of the stick"
The crew were left to make snap and instinctive decisions that on face value led to an overrun. However once presented with the full facts as known (and importantly some unknown to the crew) they "almost" got away with it.
Perhaps it would be better to focus on how the information reqd can be better disseminated to crews approaching this airport in the future?
rgds
Glad Rag.
RBF, re #48, we hold differing views of the subject, which is often the case in aviation with both positive and negative aspects (e.g. good CRM cross monitoring vs poor situation awareness).
My views relate to a generic situation and not that of the accident; in addition, I could be biased by certification experienced where aircraft are designed, built, and tested to withstand the theoretical worst case cb, and also my experience of flying in cbs (briefly) during tests.
If future data associates the YYZ cb with a microburst (mauve on the radar?), and that the crew were aware of this, then I can reconsider my views.
My concerns relate to how pilots might use information acquired during training and thereafter with experience. An out of context or casual remark (web comment) may form a bias, resulting in misunderstanding a situation or choosing an incorrect course of action, particularly in a stressful situation.
I would not disagree with a view that all of the hazards of flight in cbs can be known, but at least some of the ‘cb accidents’ resulted from inappropriate crew actions and not due to the strength of an aircraft. Similarly, we cannot be sure of the results of an overrun, events usually precipitated by crew action.
Pilots do not choose to fly in cbs nor experience an overrun. However, the choice between the two when there are apparently no other options, might be represented a pilot’s perception of the level of risk s/he allocates to each. This judgment is influenced by knowledge, experience, situation assessment, and thinking behavior; the latter can be affected by bias and false knowledge.
The point of my previous post was to illustrate that it is unusual for any situation to be clear cut and a ‘this or that’ statement, relating to a specific accident provides opportunity for the unwary to form a (subconscious) bias, which may not apply to all situations; i.e. whilst accidents do repeat themselves, it is rarely a specific one, thus the choices of action should be judged on the situation at the time and not on any preconceptions.
How Good Pilots Make Bad Decisions.
Errors in Aviation Decision Making.
Explanations for Seemingly Irrational Choices.
My views relate to a generic situation and not that of the accident; in addition, I could be biased by certification experienced where aircraft are designed, built, and tested to withstand the theoretical worst case cb, and also my experience of flying in cbs (briefly) during tests.
If future data associates the YYZ cb with a microburst (mauve on the radar?), and that the crew were aware of this, then I can reconsider my views.
My concerns relate to how pilots might use information acquired during training and thereafter with experience. An out of context or casual remark (web comment) may form a bias, resulting in misunderstanding a situation or choosing an incorrect course of action, particularly in a stressful situation.
I would not disagree with a view that all of the hazards of flight in cbs can be known, but at least some of the ‘cb accidents’ resulted from inappropriate crew actions and not due to the strength of an aircraft. Similarly, we cannot be sure of the results of an overrun, events usually precipitated by crew action.
Pilots do not choose to fly in cbs nor experience an overrun. However, the choice between the two when there are apparently no other options, might be represented a pilot’s perception of the level of risk s/he allocates to each. This judgment is influenced by knowledge, experience, situation assessment, and thinking behavior; the latter can be affected by bias and false knowledge.
The point of my previous post was to illustrate that it is unusual for any situation to be clear cut and a ‘this or that’ statement, relating to a specific accident provides opportunity for the unwary to form a (subconscious) bias, which may not apply to all situations; i.e. whilst accidents do repeat themselves, it is rarely a specific one, thus the choices of action should be judged on the situation at the time and not on any preconceptions.
How Good Pilots Make Bad Decisions.
Errors in Aviation Decision Making.
Explanations for Seemingly Irrational Choices.
Join Date: Feb 2007
Location: Not here
Posts: 222
Likes: 0
Received 0 Likes
on
0 Posts
alf5071h Re.: #47 long post.
The pilots have to be given, as much as possible, very accurate data to make proper decisions.
Obviously, perfect data is not always possible.
In this case, extremely surprisingly, the management abandoned the pilots with missing, confusing, or wrong data.
I thought this had been already resolved with all the previous over-runs.
The way I would do this is as such (there could be better solutions):
The pilots determine their safety margin for the actual runway finally used (in this case a pitiful 220 ft; see report and below).
When passing the theoretical touch-down point (i.e. ILS GS going offscale), if the pilot hasn't touched down within 1 sec (220 ft) then over-shoot anytime prior to reverse thrust initiation.
Keep in mind that the pilots seem to have had "I can do this landing"-itis.
From the Air France MANEX performance information, the predicted landing distance needed
for the landing in Toronto on a contaminated runway with zero wind and no thrust reverser
was 8780 feet. For Runway 24L, the extra margin was only 220 feet. This very small landing
distance margin was eaten up by the long flare during the landing. With a tailwind, there was
negative margin, which would mean an overrun. The pilots were not aware that the MANEX–
predicted landing distance when landing with a tailwind exceeded the length of Runway 24L.
In the absence of knowledge of the required landing distance under varying performance
conditions, crews will not be aware of rapidly developing overrun situations. Because of this,
there is a high potential that crews will make inadequate go/no-go decisions, thereby
increasing the risk of damage to persons, property, and the environment.
Therefore, the Board recommends that:
The Department of Transport and other civil aviation authorities require
crews to establish the margin of error between landing distance available
and landing distance required before conducting an approach into
deteriorating weather.
for the landing in Toronto on a contaminated runway with zero wind and no thrust reverser
was 8780 feet. For Runway 24L, the extra margin was only 220 feet. This very small landing
distance margin was eaten up by the long flare during the landing. With a tailwind, there was
negative margin, which would mean an overrun. The pilots were not aware that the MANEX–
predicted landing distance when landing with a tailwind exceeded the length of Runway 24L.
In the absence of knowledge of the required landing distance under varying performance
conditions, crews will not be aware of rapidly developing overrun situations. Because of this,
there is a high potential that crews will make inadequate go/no-go decisions, thereby
increasing the risk of damage to persons, property, and the environment.
Therefore, the Board recommends that:
The Department of Transport and other civil aviation authorities require
crews to establish the margin of error between landing distance available
and landing distance required before conducting an approach into
deteriorating weather.
Alph2Z, I agree with many of your points, but I suggest that careful consideration is required when searching for solutions so as not to fix the problems of one specific accident and in doing so leave open opportunities for future ones.
Visibility was a specific issue at YYZ, but the generic points (the YYZ crew did not know that the visibility had changed) are the limitations in the ground based detecting and reporting systems.
It could be argued that crews could interpret weather radar indications such that ‘red’ zones indicate reduced visibility – but reduced by how much, limiting or not limiting? The natural weaknesses of human performance do not guarantee the crew using radar in this way nor making such inference.
In routine operations wx radar detects rainfall rate which is used to infer convective weather (normal inference from training), but with these indications during the approach it is probably more important to consider the effects of heavy rain on the runway state – contaminated or not, rather than convection (although I recognise the concerns over the threat of windshear etc in the event of a GA). The state of the runway - braking effectiveness is another key item where crews have limited information; like visibility it could be inferred from radar (see refs @ #45). Then there is wind! See ref @#47.
Re the calculation of landing distance (YYZ), I think that you have overlooked the point that the crew may not have appreciated that the runway was contaminated (investigation hindsight), so from the crew’s perspective, their planned landing distance did have the required margin of safety.
Although the recommendation for a pre landing check of conditions is a positive step, the ‘how to’ remains open. Check the many references to working groups and meetings on aircraft performance on contaminated runways, the nearest workable solution so far is CRFI, but IIRC it does not correlate well with water contamination; then how often is the runway to be checked – at YYZ the required frequency might have been between each landing.
The points above build a case for crews to use radar in a wide ranging sense, making inferences and decisions at critical periods during the approach often associated with high workload. This can be done, but it would be more easily achieved with the aid of ‘trigger items’ cueing crews to think in this manner, still subject to all of the human limitations in such situations. This is not to excuse crews, but they do need help; knowledge, training, and technology.
I agree that the operational management have a role to play in there areas, so too the regulator, but there are also wider issues. There are ever increasing commercial pressures, public perception of safety (when you are already good any accident is seen as a negative trend), and complacency, particularly from the successes in normal operations where we may not question just how close we are to the ‘edge of safety’. I urge everyone (pilots, management, regulators) to read the ref @ #54 particularly re irrational decisions.
We should avoid criticism and implied blame when pointing out how others should be doing ‘something’; instead use meaningful critique of the facts from the mounting evidence of many overrun incidents and accidents to identify items for action. As asked previously re training, what do we train? Or more to the point what can we, the collective operational team or individually, do to improve safety.
I disagree with your point on the use of ILS GS re touch down point. If you recall the GS consists of a reflected beam along the approach path, thus at very low altitude when approaching the transmitter, the beam has a parabolic path. If you are able to view it at such time the GS starts increasing, but hopefully at that time the receiver would flag red.
The touchdown zone is marked on the runway, this and distance markers along the runway require good visibility and spare mental effort to appreciate them. In the YYZ accident both of these parameters were out of limit.
Many of my points refer to the human aspects, which would seek alleviation for an "I can do this landing-itis” attitude (irrational decision). However there are many aspects of management, including SOPs, training, and increased margins for scheduled landing distance, all of which could reduce the risks of an overrun, primarily by reducing the opportunities for error.
There is also much more that can be done in mitigating activities to counter the inevitable ‘hole in the cheese’, by improving overrun areas and increasing landing distance requirements.
To avoid arguing to agree violently, I concede that we are seeking the same objective, but perhaps taking slightly different views.
Visibility was a specific issue at YYZ, but the generic points (the YYZ crew did not know that the visibility had changed) are the limitations in the ground based detecting and reporting systems.
It could be argued that crews could interpret weather radar indications such that ‘red’ zones indicate reduced visibility – but reduced by how much, limiting or not limiting? The natural weaknesses of human performance do not guarantee the crew using radar in this way nor making such inference.
In routine operations wx radar detects rainfall rate which is used to infer convective weather (normal inference from training), but with these indications during the approach it is probably more important to consider the effects of heavy rain on the runway state – contaminated or not, rather than convection (although I recognise the concerns over the threat of windshear etc in the event of a GA). The state of the runway - braking effectiveness is another key item where crews have limited information; like visibility it could be inferred from radar (see refs @ #45). Then there is wind! See ref @#47.
Re the calculation of landing distance (YYZ), I think that you have overlooked the point that the crew may not have appreciated that the runway was contaminated (investigation hindsight), so from the crew’s perspective, their planned landing distance did have the required margin of safety.
Although the recommendation for a pre landing check of conditions is a positive step, the ‘how to’ remains open. Check the many references to working groups and meetings on aircraft performance on contaminated runways, the nearest workable solution so far is CRFI, but IIRC it does not correlate well with water contamination; then how often is the runway to be checked – at YYZ the required frequency might have been between each landing.
The points above build a case for crews to use radar in a wide ranging sense, making inferences and decisions at critical periods during the approach often associated with high workload. This can be done, but it would be more easily achieved with the aid of ‘trigger items’ cueing crews to think in this manner, still subject to all of the human limitations in such situations. This is not to excuse crews, but they do need help; knowledge, training, and technology.
I agree that the operational management have a role to play in there areas, so too the regulator, but there are also wider issues. There are ever increasing commercial pressures, public perception of safety (when you are already good any accident is seen as a negative trend), and complacency, particularly from the successes in normal operations where we may not question just how close we are to the ‘edge of safety’. I urge everyone (pilots, management, regulators) to read the ref @ #54 particularly re irrational decisions.
We should avoid criticism and implied blame when pointing out how others should be doing ‘something’; instead use meaningful critique of the facts from the mounting evidence of many overrun incidents and accidents to identify items for action. As asked previously re training, what do we train? Or more to the point what can we, the collective operational team or individually, do to improve safety.
I disagree with your point on the use of ILS GS re touch down point. If you recall the GS consists of a reflected beam along the approach path, thus at very low altitude when approaching the transmitter, the beam has a parabolic path. If you are able to view it at such time the GS starts increasing, but hopefully at that time the receiver would flag red.
The touchdown zone is marked on the runway, this and distance markers along the runway require good visibility and spare mental effort to appreciate them. In the YYZ accident both of these parameters were out of limit.
Many of my points refer to the human aspects, which would seek alleviation for an "I can do this landing-itis” attitude (irrational decision). However there are many aspects of management, including SOPs, training, and increased margins for scheduled landing distance, all of which could reduce the risks of an overrun, primarily by reducing the opportunities for error.
There is also much more that can be done in mitigating activities to counter the inevitable ‘hole in the cheese’, by improving overrun areas and increasing landing distance requirements.
To avoid arguing to agree violently, I concede that we are seeking the same objective, but perhaps taking slightly different views.
Avoiding an overrun: what should be trained?
The TC report calls for more training, one of many possible initiatives to minimize risk or exposure to potentially hazardous situations. In a previous post, I asked “What do we train?”
To avoid responses being too focused on this specific accident - we require generic solutions which should help prevent any future events, - I have started a new thread in CRM and Safety. Avoiding an overrun: what should be trained?
http://www.pprune.org/forums/showthread.php?t=306748
To avoid responses being too focused on this specific accident - we require generic solutions which should help prevent any future events, - I have started a new thread in CRM and Safety. Avoiding an overrun: what should be trained?
http://www.pprune.org/forums/showthread.php?t=306748
Join Date: Sep 2007
Location: Toronto
Posts: 49
Likes: 0
Received 0 Likes
on
0 Posts
While I'm just a SLF, I was reading through the report and found out that the FO disengaged the autopilot at around 100 feet AGL. I'm wondering if that could have been a cause in the crash, since the weather was pretty bad in the last few seconds and therefore wouldn't you want to have a good "feel" of the situation before putting the plane down. It's kind of hard to explain, but an analogy would be taking a soccer player who didn't play for a while and then simply telling him to go play, without any warm ups.