BEA - Bureau d'Enquêtes et d'Analyses pour la sécurité de l'aviation civile - Zone Sud - Bâtiment 153 - 200, rue de Paris - Aéroport du Bourget - 93352 Le Bourget Cedex - FRANCE - Téléphone 33 1 49 92 72 00 - Télécopie 33 1 49 92 72 03BEA - Bureau d'Enquêtes et d'Analyses pour la sécurité de l'aviation civile - Zone Sud - Bâtiment 153 - 200, rue de Paris - Aéroport du Bourget - 93352 Le Bourget Cedex - FRANCE - Téléphone 33 1 49 92 72 00 - Télécopie 33 1 49 92 72 03


Flight AF 447 on 1st June 2009

A330-203, registered F-GZCP


Press Conference, 29 July 2011
Questions and Answers Session


Question: without the Pitot probe problem, would the accident have happened? After the autopilot stopped, were the pilots capable, humanly and operationally, of avoiding the accident?

Jean-Paul Troadec: As mentioned in the BEA's second interim report, we analysed a certain number of events which resulted in the loss of speed displays and which were simple incidents, so the answer is yes, the situation after the loss of speed displays was recoverable since recovery had been possible in dozens of events.

Question: You said in a previous report that the Trimmable Horizontal Stabiliser (THS) moved from 3 to 13 degrees, I would like to know:

  • If it stayed like that until the end of the flight [YES director's answer]

  • If the airplane stayed at 13° controlled by the computer alone or if the pilot could have manually reversed the situation

  • What was the command given to the pilot by the Flight Director and if the pilot's decision to pitch the airplane nose-up was consistent with the command given by the flight director

Jean-Paul Troadec: well, for that last question, that is one of the parameters which were not recorded and so we do not have a very clear vision of the commands given by the flight director. So we cannot answer this question today. We therefore do not know if the pilots followed them or not.

About the THS, in this alternate law, you still have the automatic trim function so the pilot does not directly manually control the automatic trim so it stayed in nose-up position at 13° following the pilot's inputs but nose-down inputs would have made it possible to reduce the angle of the THS.

Question: is what was done the opposite of what should have been done?

Jean-Paul Troadec: As I said earlier, this is a complex event, even if it lasted a relatively short time. As described earlier by Alain Bouillard, the first event which triggered all this was the autopilot disconnection, as a result of the loss of speed displays very probably following the icing of the probes by ice crystals. At that moment, the pilot should have applied the unreliable IAS procedure and in fact this procedure consists specifically of adopting a pitch attitude of 5°, whereas the pitch attitude that was adopted at that moment was greater.


Question: meaning that the airplane pitched nose-up?

Jean-Paul Troadec: no, that is why the airplane climbed, this was indicated earlier, the airplane gained considerable vertical speed, around 7,000 feet/minute.


Question: What was the right thing to do?

Jean-Paul Troadec: The pitch attitude was excessive.


Question: that is to say?

Jean-Paul Troadec: Well, it was 10 to 12 degrees; it should have been 5 degrees

Question: some analysts say that this problem with the THS, calculated by computers, would have prevented the airplane from being able to pitch down. Is it true that the airplane was too tilted?

Jean-Paul Troadec: no, I said that if the pilot had made a sufficiently pronounced nose-down input, the THS would have taken a less nose-up position.


Question: yes but when there is this change from 3 to 13 in such a short time, is the airplane controllable since it is the computer? Is it possible to go back?

Jean-Paul Troadec: the airplane is controllable via the computer.


Question: you say that towards the end, the Pitot probes were working. At that stage, could the pilot have done something and to prevent the disaster or did they take a wrong decision again?

Jean-Paul Troadec: The second important event was the stall that happened at an altitude of 38,000 feet. The airplane had climbed quite quickly. It was at a lower speed because of this climb. Before the stall, yes, of course, the situation was recoverable.

Question: you said that the Captain had not clearly defined the task-sharing when leaving the cockpit. Did he hand over command of the airplane to the pilot in the right-hand seat? Also, do you understand why the more experienced co-pilot in the left-hand seat did not take back control of the airplane?

Jean-Paul Troadec: I will let Alain Bouillard answer this question.

Alain Bouillard: certainly, on his departure, the Captain implicitly designated the pilot flying who was on the right as his relief.

It is always the co-pilot flying who relieves the Captain. At this stage, we are trying through the human factors working group to understand also what the relations could have been between a more experienced pilot and a younger pilot. This is the whole point of this working group.

Question: from the conversations that you have heard, did the Captain understand that the airplane was in a stall when he came back into the cockpit? Did he do anything to prevent that?

Alain Bouillard: just before entering the flight deck the Captain probably heard the stall warning, but the problem is that it stopped a few seconds after he entered the flight deck and that could have been a source of incomprehension, that is to say that it did not help him identify the airplane's situation at that moment.


Question: you mentioned a nose-down input at one moment, an attempt… Did this attempt stop with the reappearance of the stall signal? Were the pilots misled by this erratic warning?

Alain Bouillard: There are two things: we are looking at the operation of the stall warning in these conditions and we are also looking in this human factors group – that I will often mention – at what could have been the understanding, in this phase of the flight, of a stall warning that reappears when the angle of attack is decreased and which deactivates as soon as the stick is pulled again. Indeed, at that angle of attack, the speed parameters become invalid and the warning deactivates.


Question: But in your projection, is there really a direct relation between the inputs on the side stick and the triggering of the warning?

Alain Bouillard: yes, as soon as the pitch attitude decreases, the angle of attack decreases, the speeds become valid again. When the airplane pitches nose down, and when he pulls on the side stick again, the warning disappears due to the fact that the speeds become invalid; yes this is something that we are going to look at, with this human factors group, what could have been the influence on the pilot's understanding and actions.

Question: can you tell us if when the pilot hears the warning re-trigger, he stops his nose-up input?

Alain Bouillard: This is something that is ongoing now, the investigation is not completed. We are trying to understand each action on the side stick, the consequences, the triggering of warnings and the pilot's reactions.

Question: some media, particularly in France, have stuck with the piloting error formula. Can I ask you objectively if you, you also adhere to the idea of piloting error?

Alain Bouillard: if you wish to ask this question, it should be asked of the people whose duty it is to determine responsibility, as for us, we are trying to understand what happened through this accident, to understand behaviour, to understand the airplane's operation and to issue safety recommendations in order to improve flight safety.

Question: Well, how do you explain that none of the pilots formally identified the stall situation? Are there too many warnings; are these warnings not clearly comprehensible?

Alain Bouillard: I will repeat once more, we are going to create a human factors group which will have responsibility for looking at this problem of perception in very particular circumstances: in these circumstances at night, in clouds, in a stressful situation, in a situation of incomprehension, we are going to try to understand through this working group what the pilot's behaviour could be.

Question: so how can you explain that they did not identify this stall situation?

Alain Bouillard: Well we are waiting for the human factors group work to try to answer these questions.


Question: on the stall warning, do you have any notifications to make already?

Alain Bouillard: No, we know that it worked for 54 seconds, now we still need to understand under what conditions, a pilot, a human being does not notice a warning in stressful situations …

Question: the way in which it triggers, retriggers …?

Alain Bouillard: that will be dealt with in the systems group who are going to look at the stall warning operation in detail.

Question: concerning the Captain, when he arrived in the cockpit, could he understand, even physically, even if the warning de-activated, what was happening? Physically, can you feel a stall? Could he have analysed it otherwise than with the stall warning?

Alain Bouillard: The Captain arrived, the airplane was in a stall situation, he probably heard, as I told you, the warning before going into the flight deck, the warning stopped in the following few seconds and the Captain did not have this instant understanding of the situation due to these contradictions in the triggering of the warning, the warning de-activation and he was not able in a few seconds to make a full analysis of what was happening in the cockpit.

Question: from the conversations that you heard, can you describe the atmosphere in the cockpit? What happened between these pilots …?

Alain Bouillard: This is something to which I will again answer by referring to the human factors group which should help us to understand what atmosphere, what was the situation at that moment, what frame of mind each pilot was in at that moment; now, these are matters for specialists and so we are waiting for the answers.

Question: But from the conversations, you yourself must have an idea if things were alright, if they were tense between the two co-pilots, did they panic or …

Alain Bouillard: the two pilots were in a situation that they did not understand and now the aim of setting up this human factors group is to try, through their dialogue, listening, tone of voice, to understand what state they were in and why they did not understand or recognise the situation they were in.

Question: Did Air France agree to collaborate and did it give you all the elements of the crew members' professional files? Simulator training, assessments? Have you been able to consult all this and what is the result, please?

Alain Bouillard: Today we have access to all the pilots' career details; we have been able to establish their career path, their licences, their ratings, their checks, their training, with no problems. We are working with Air France; it is our technical advisor in this matter. It provides us with all the documents that we are entitled to expect in order to understand.

Question: Air France is convinced that the stall warning was difficult to follow, sometimes misleading. Has your investigation shown that the presence or absence of this warning corresponded to the airplane's real situation?

Alain Bouillard: Today, according to the parameters, yes. The airplane was always in a stall situation with a very high angle of attack. Remember, always above 40 degrees. And we are trying to understand what influence the de-activation, the reactivation, of this warning could have had on the crew's understanding or incomprehension of the situation.

Question: I have two questions: following on from the question asked earlier and perhaps following on from your answer, were the pilots, by chance, too calm in the cockpit?

Alain Bouillard: no, when there was a speed inconsistency, there was professional work by pilots who were worried about being able to reach a higher flight level, who were concerned about the weather, who were concerned about their en route alternate aerodrome throughout the journey. These were quite professional conversations in the cockpit.

Question: Ok. And secondly, from what point was it no longer at all possible to take back control of the airplane? At what precise moment could they not do any more?

Alain Bouillard: it is very difficult to answer this question. The stall warning is a warning that indicates that the airplane is going to leave its flight envelope. Corrective actions must be taken very quickly to bring the airplane back within its flight envelope. The warning must be recognised and the situation must be recognised. Yet, visibly the pilots did not recognise it and we are trying to understand why this situation was not noticed by the crew.

Question: So we can say one minute ….

Alain Bouillard: between the disconnection and the continuous sounding for 54 seconds, there was less than one minute.

Question: I don't understand the problem of the image recorders very well in the recommendations? The CVRs – voice recorders – are fitted with a button, which, at the end of the flight is activated, which erases the CVR memory so what the pilots said during the flight cannot be used wildly as long as there was no incident. Why can't a similar image recorder be imagined. After landing, the pilots erase these images so no ill considered use is possible.

Alain Bouillard: It is up to ICAO to respond. This relates to the second recommendation that has been made: video-recorders in the cockpits with very strict rules for using these recordings.

Question: Were the pilots aware that they were in the process of crashing and when do you expect to publish the final report?

Jean-Paul Troadec: In the last minutes of the flight, yes besides they called out the altitude going through 10,000 feet, then they heard the EGPWS warning which indicated proximity to the surface, so, yes they were aware of the approach of the accident.

On the publication of the final report: I think that this conference has shown that there was a lot of work on very complicated subjects. When psychology, ergonomics, human factors are concerned, when you want to look at the operation of the airplane systems, more systemic subjects, that requires a lot of time. And in addition, given the consultation rules that we have in the context of this investigation, what I can tell you is that the final report will not be published this year, but in the course of the first half of next year.

Question: just for confirmation: up until now we do not know the reasons for the pilot flying making nose-up inputs, do we?

Jean-Paul Troadec: That is what Alain Bouillard said and repeated, yes. Those are actions that we do not understand.

Question: about the pilots' training, how many hours of training did each pilot have to cope with a stall situation at high altitude?

Alain Bouillard: we do not have the full details but they had done very little manual aircraft handling at high altitude. The Captain had done it in a training session a long time ago. In recent years – this will be checked – neither of the two co-pilots had done a training session on handling in alternate law or in direct law at high altitude.

Question: is this the first time that you are recommending aircraft handling training for commercial aircraft?

Alain Bouillard: There may already have been comments and working groups on this subject. This is something that is topical today in extremely modern airplanes so we issued this recommendation, but perhaps during the investigation other recommendations, as the director said, could be issued on this subject with further complete analyses. As for the BEA, I believe it is the first time that a recommendation of this genre has been formalised.

Jean-Paul Troadec: I would like to add that this preoccupation is shared by the whole aviation community. The FAA in particular has set up programmes on crew training in the use of modern aircraft and maybe, I know that there are also comments on the subject, on even initial training for pilots today which is still a little, how would you say, marred by past memories. This is to say that pilots must be trained today right from the start in the perspective of new equipment, depending of course on their way of operating but also on their reliability. These days there is no longer the same rate of failure on engines as when there were piston engines for example. So I think that all these comments come into a general framework that goes beyond the BEA's comments. We know that the FAA and EASA, Boeing and Airbus are all working on those subjects.

Question: you say….less than one minute between the triggering of the autopilot and the stall warning. What is this, is it a few seconds because the airplane was in normal speed, at a normal altitude and the triggering of the stall warning occurred almost instantly. It would have lost 60-80-90 knots in a few seconds?

Jean-Paul Troadec: when the autopilot disconnection occurred, the airplane was at 35,000 feet, it climbed very quickly to 37,500 feet so with a high rate of climb and it was in this cruise phase, almost stabilised but higher, that it did in fact lose around 80 knots since it was at about 180 knots at that moment, at the time of the stall warning.

Question: thank you very much. What is also linked is that we do not know the displays, as you told us, of the right side airspeed indicator. When drills were being done on 330 on take-off in particular, erroneous anemometric displays, often excessive speeds could be seen, it was programmed like that in the simulator on the right, which led the pilot to pitch his airplane nose-up since it had an excessive speed and he was prevented from doing it. That could perhaps lead to the thought that the speed was excessive in read-out on the pilot flying in the right seat.

Jean-Paul Troadec: as we said, we do not know now what the display on the right pilot's side was. It so happens that a maintenance calculator was found among the parts that were stored in Toulouse and we are going to try and recover this information. So if we have that information, we will be able to answer your question.

Question: which could explain that they pitched nose-up…

Jean-Paul Troadec: we will see.

Question: there are a lot of airplanes which, when they stall, pitch nose-down including when they have stalled and their last command was to pitch up. This is apparently a phenomenon that many glider or flying school pilots are familiar with. Yet, here we can see that this phenomenon did not occur, that the airplane fell straight while remaining nose-up. Is this a stall mode that is normal or expected?

Jean-Paul Troadec: what was said, is that the airplane stalled from the pilot's nose-up input. So, mostly the pilot's inputs were to pitch nose-up, which maintained the stall.

Question: is the way the airplane stalled, that is to say straight while remaining nose-up and not nose-down, is it normal or typical or expected. Let's say that a lot of airplanes pitch nose-down when they stall.

Jean-Paul Troadec: it's a point where… Well, I cannot answer your question and say if it is normal or not. It is the situation of this airplane which is perfectly in accordance with the regulations and which was certified.

Question: you say that the loss of speed would have lasted less than one minute, and that the pilot would have had the time to recover the aircraft but in fact when you read the documents, you see that they recovered speed but lost it just afterwards, that in fact the recovery would have lasted only a few seconds and that, throughout the duration of the accident. So, could the fact that they did not succeed in recovering the airplane have been attributed to the fact that they had no speed displays following the problem with the probes or erroneous speed which led them to pitch the airplane nose-up?

Jean-Paul Troadec: at the time of the stall, the speed on the left has been recovered and on the ISIS and we do not know the value on the right. So the airplane was at 185 knots, it was in stabilised level flight and to recover the situation, a lighter hand would have been required and the airplane would have returned to a normal pitch attitude.

Question: You said, they say "we have no speed display" …

Jean-Paul Troadec:   yes, but that is well after that. If you like, there is the stall phase, which was illustrated by the end of phase 2. So at that moment, the situation was recoverable. In fact, we could have not reached the stall normally, and then afterwards you have phase 3 where there we are in a different stall with very high angle of attack values.

Question: talking about recurrent training because you seem to bring up this problem, are there standards today or does a pilot who has obtained his diploma or I don't know what have no obligation to keep informed of new technologies, to redo things, what is the norm, does it differ according to airlines?

Jean-Paul Troadec: there are of course rules, a regulation relating to recurrent training of crews. These are EASA rules. It is obvious that all pilots must have regular permanent training. The practical training depends of course on the airplane but can also depend on the airlines. The airlines have a certain flexibility to adapt their training programme according to the equipment and according to the routes they follow, obviously. So all this is fine, is regulated and in this case, the training and education situation of these pilots was in order.

Question: could the pilots' nose-up command have been given afterwards or not – the flight director's nose-up command whose parameters we do not know, as you said.

Jean-Paul Troadec: no, but there was no Flight Director probably at that moment.

Question: we are coming back again to this question, but it is to try and understand the nose-up command, you said that you did not understand this nose-up command, can you explain to us then the moves that should have been made or the flight handling procedures?

Alain Bouillard: it is very difficult to answer your question because there is a procedure which is to be performed from memory, which requires a certain number of actions: disconnecting the automatic systems and adopting a pitch attitude of 5 degrees. The pilot adopted a pitch attitude of 10 degrees and the whole question today, the big problem is to try and understand through his experience, through his training. We know that this pilot did not have, and nor did the copilot have, training on the speed loss, inconsistency at high altitude. We know that these exercises were done at low altitude where pitch attitude settings, the pitch attitudes are different since we are asked to set 10 or 15 degrees of pitch attitude depending on the airplane configuration. What we do know is that they had had this training a few months previously, and precisely, on a simulator, on take-off from Rio.

Question: but if there had not been these nose-up commands, could that have prevented the accident?

Alain Bouillard: Today in the studies and in the examination of the events, of the incidents which took place before, when you maintain either the pitch attitude or take a very moderate pitch attitude, the airplane continues to fly: it has the speed, it has the engines, it has the flight control surfaces and the airplane continues to fly.

Question: are there other references made to the VARIO in the CVR or is there just the 10,000 feet in descent?

Alain Bouillard: no, we have it in the CVR, it is something that was recorded but the pilots do not refer to it.

Question: is it normal that you do not have the recordings of the pilot in the right seat; does that happen in all flights?

Alain Bouillard: There are obligations to record two speed values. The choice was made from the Captain position and the safety instrument. We estimate that they are the two most important values. Today, we notice that we need this right side value too..

Question: so we will never know what the pilot in the right seat saw in front of him? The data he had in front of him?

Alain Bouillard: in the study of the systems functioning, we will perhaps be able to deduce the speed that was displayed on the right in particular via the calculators which were removed and in particular the maintenance recorder where the speed on the right side appears.

Question: at one moment did the Captain or the copilot try to take back the controls or did they give directives to the pilot who had them?

Alain Bouillard: in the last minute of flight or in the last two minutes of flight the pilot in the left seat took over the controls and said "I have the controls" and there was an instant pick up by the pilot on the right with stick priority.

Question: can you clarify what you said earlier, that is to say the Captain's implicit order to put the controls of the aircraft in the hands of the more experienced pilot. Is it normal that it ends up in the hands of the youngest and less experienced pilot?

Alain Bouillard: That is also something that we are looking at. It is the Captain's decision who, when he leaves the flight deck to take his rest, designates according to his own criteria, defines who will be the relief pilot during his absence.

Question: but was it he who designated the youngest one?

Alain Bouillard: Implicitly yes since it is the pilot still flying who ensures the Captain's relief.

Inaudible question …

Alain Bouillard: What?

Question: Perhaps it was not the youngest on the airplane?

Alain Bouillard: In that case, yes, it was the youngest on the airplane.

Question: excuse me but I believe you said implicitly, he designated the pilot with the most experience? Implicitly, because it is that….

The room – No no ….

Alain Bouillard: No, because the pilot flying was the one in the right hand seat, the relief Captain position must always be ensured by the pilot flying.

Question: and he was appointed pilot flying?

Alain Bouillard: Yes exactly. He was appointed pilot flying on take-off probably and because he was carrying out this duty, it was he who was relief pilot. If that is the question you are asking me, there was no clear announcement from the Captain when he left the flight deck saying "you are the relief pilot and it is you who will take my …." And on the other hand, he said to the second copilot that it was he who would take his place.

Question: is this in compliance with procedures?

Alain Bouillard: Yes, it is perfectly in compliance with procedures.

Question: that is to say that the youngest was designated … and that is normal? Or is it special?

Alain Bouillard: I would answer that it is a question of experience and that it is not always the most experienced who have the most skills, it is the Captain's choice, he made it according to his own criteria and, it was all anyway in compliance with the regulations.

Question: two other questions: do you intend to make available the complete conversations between the pilots? Second thing, very clearly, what explanation do you expect between now and the final report? What do you still hope for from what you have in the hangars, what can be used …

Alain Bouillard: Well, only the parts of the conversation recorder that are necessary to understanding the event will appear in the report. There are not the complete contents of the CVR. There is only what is necessary to understanding the event. On the other hand, there is still a lot of work on the calculators, there is still a lot of work on the analysis of the airplane systems operation, and there is an enormous amount of work left on the human factors element. This is a group that is going to take, that is going to be set up in the next few days. There is an extensive work programme for this group and then, there is everything concerning, as the director said earlier, all the systemic, the flight safety actions at Air France, the surveillance activities of the oversight authorities. All this will be recorded in the final report and, as the director said earlier, perhaps during the first half of 2012.

Question: and in the calculators, what do you expect to find exactly?

Alain Bouillard: Well, perhaps the speeds in the DAR1, the right side speeds, perhaps information which will enable us to know the position of the Flight Directors; there you are, so we have to look at the parameters that are recorded. In so far as these parameters are valid and available, as these are calculators which are not protected and which remained for nearly 24 months at the bottom of the ocean.

Question: is there an explanation in this right side speed on the command to pitch nose-up which seems incomprehensible to everyone?

Alain Bouillard: It is difficult to answer now. I am going to give a non-committal reply, perhaps yes, perhaps no. Right now I cannot speculate on the right side speed displays.

Question: would you mind just repeating please: you spoke of a training course that took place at low altitude, where a pitch attitude of 10 or 15 % would have been more suitable. What training course did he have and when?

Alain Bouillard: both co-pilots took it in 2008 and 2009. It was in a regulatory training course and it was 10 degrees, not 10%, it was 10 degrees pitch attitude and 15 degrees pitch attitude. They did this training during regulatory training sessions which are dedicated to crew training.

Question: for which ….

Alain Bouillard: for speed inconsistency.

Question: please, during the last press conference, you specifically said that everything started with the icing of the Pitot tubes. Now we learn that the pilots were not trained sufficiently to face these situations. Do observations have to be made again or can a conclusion be reached on these facts?

Alain Bouillard: the pilots were trained to face speed inconsistency at low height. They did not have this training at high altitude. There you have it now, the statement we made.

Question: can we reach a conclusion on these two events, clearly everything was triggered, as you said the last time, by the icing of the Pitot tubes.

Alain Bouillard: that's true, that has now been confirmed.

Question: Today we learn that the pilots were not sufficiently trained or not at all trained to face this situation at high altitude. There are two observations. Is it not possible to reach the conclusion on these two events, that there was no link between the two?

Alain Bouillard: that is to say that now we are making the link between what happened: between the fact that the Pitots caused a speed inconsistency and the reactions of the pilot, which appear excessive: adopting pitch attitudes of 10 degrees at 35,000 feet. Now that is the whole problem that we are trying to understand. They were not trained, but this 10 degrees input at high altitude seems very excessive to us and today we are trying to understand why.

Question: But, the three pilots who were there - with the experience that they had - how can they take such decisions – when most of the pilots in a flying club in a stall pitch the airplane nose-down – how can they take the completely opposite decision …?

Alain Bouillard: this is what we have been trying to explain to you for a while. We do not know, we are trying to establish the information that the crew had at the time of the autopilot disconnection: what could the role of the Flight Directors have been, what perception of the situation could they have had, what could their spatial disorientation have been, what were the phenomena that could have contributed to this pitch attitude. Today this is something that appears contrary to logic. We are looking for rational explanations, because they are things which do not appear to be rational in this situation. We have no explanation to give today.

Question: Sir… Sir can we suppose that most pilots today in any airline, not just Air France, are not trained to handle an airplane of this type manually at high altitude. Can we deduce that? And secondly, can we also deduce that there is human error, but that this is not yet proven?

Alain Bouillard: this situation is a very particular situation. We saw, and we told you earlier, that through the handling of a certain number of incidents that occurred with speed inconsistency, the pilots reacted and had reactions in accordance with the procedures to be applied. Now, since the accident, numerous improvements have been introduced, particularly

in the probes, in pilot training, in approach procedures and stall recovery. For the past few years, few months, the pilots are trained, we retrain pilots at high altitude. The recommendation will help improve this situation.

Question: but before no?

Alain Bouillard: before … Now, what remains to be determined is if it is specific to this flight as you know that each accident is nevertheless a very particular combination of circumstances and it is up to the investigation now to determine if this is a generality, which I do not believe, or if it is something, not a failing on the part of the pilot but something that occurred, an incomprehension of the situation. Today, we understand how the accident happened. Now the question remains; why did it happen?

Inaudible speech in the room

Question: … A human failing is possible, it has not been proved yet.

Alain Bouillard: Today, the investigation that is going to say. If it is, as you call it, a human failing or if it is something that was presented to the pilot that they followed, if there are external elements or other things which could have contributed to having an erroneous representation of the situation.

Question: you cannot be sure of it? That there was human failing?

Alain Bouillard: no not at all. Anyway, our job is not to determine blame, it is to understand and to explain why an accident happened and to propose corrective actions to improve safety.

Question: do you know the causes of the accident?

Alain Bouillard: no, not yet. We have not completed the investigation, madam.

Question: have you analysed other technical anomalies that could have led to the pilots' confusion? We understand that they were in a very difficult situation. Is this normal, this brings us back to the question about the stall warning that sounded for 54 seconds and apparently they did not manage to analyse it. How do you analyse it now? How is it that they did not understand the situation? Are there too many indicators lit up, too many warnings, a technical anomaly?

Alain Bouillard: I will repeat for the nth time that we have set up a human factors group to try and understand this non-perception of the stall warning.

Question: but just on the technical side? On the technical side, are there too many warnings lit up, a problem of sending technical signals? Too complicated?

Alain Bouillard: there was a stall warning, which is the priority warning in the cockpit when it triggers. Now all that we can say, the airplane at least in the longitudinal axis always responded to the crew's requests.

Question: so the description of anomalies, it was normal. This is the typical procedure of the description of anomalies to the pilot; warnings that are triggered…

Alain Bouillard: the stall warning yes, now we have specifically identified this warning which, when it is activated, may I remind you that it is the priority warning if there are other warnings, and covers a specific number of warnings. But in the systems' operations, we have not brought to light any failures.

Question: Wasn't the training on speed inconsistency at high altitude mandatory?

Alain Bouillard: We must look at that, it is currently being examined, and training courses on warnings, on speed inconsistency were in the training programmes. It was not particularly, specifically stated that it should take place at high or low altitude, but in the case of both pilots, they had this training, but at low height.

Jean-Paul Troadec: well, I will try to conclude this conference. First of all you have noticed the BEA's prudence. We are wary of hasty conclusions. The investigation is ongoing, it is far from being completed, there are very complex subjects to analyse so let's wait for the publication of the final report to truly understand what happened. In the meantime, you are invited to read the report which is going to be published in a little over an hour on the BEA website. This report is 120 pages. It is crammed with graphs, images, so it is very technical. I hope that it will provide further information on what we have tried to summarize in such a short time. Thank you.


1 Digital Access Recorder –maintenance recorder





Accident in the Sinai Peninsula (Egypt) to the Airbus A321-231 registered EI-ETJ. Information on 2 November 2015
Accident on 5 September 2015 in Senegal to the HS 125 700A registered 6V-AIM. Information on 10 September 2015
Accident in Indonesia to the ATR 42-300, registered PK-YRN, on Sunday 16 August 2015. Information on 18 August 2015
Accident to the Airbus A320 registered HL 7762, flight OZ 162, on 14 April 2015. Information on 15 April 2015
Accident to the Airbus A320-211 registered D-AIPX, flight 4U9525, on 24 March 2015. Preliminary Report on 6 May 2015
Accident to the Airbus A320 registered C-FTJP, flight AC624, on 29 March 2015, Halifax Airport (Canada). Information on 30 March 2015
Accident in Argentina, between two Airbus Helicopters AS 350's on Monday 9 March 2015. Information on 10 March 2015
ATR 72-600 registered B-22816, accident on 4 February 2015 in Taipei (Taiwan). Information on 4 February 2015
Airbus A320-200 registered PK-AXC, loss of contact. Information on 1st January december 2015
Accident to the McDonnell Douglas MD-83, registered EC-LTV, on 24 July 2014 in the region of Gossi (Mali). Safety recommendations issued to the FAA and the EASA on 10 July, 2015 by the President of the Mali Commission of Inquiry and the BEA.
Flight GE 222 on 23 JULY 2014 - ATR 72-212A (ATR 72-500). Information on 24 July 2014.
Boeing 777-200 - Flight MH 370 - Malaysia Airlines. Information on 24 March 2014.
Accident on 4 March 2013 to the Hawker Beechcraft Premier 1A model 390, registered VP-CAZ, near Annemasse aerodrome (74). Information on 6 March 2014.
Accident to a Socata TBM 700, registered N115KC, at Mouffy (89) on 19 November 2013. Information on 20 November 2013.
Fokker 27, registered I-MLVT and operated by Miniliner, 25 October 2013, Paris Charles de Gaulle Airport, France. Information on 8 November 2013.
Accident to the accident to the ATR 72-600 in the vicinity of Pakse airport (Champasak province), Flight QV 301. Information on 21 October 2013.
Aeroplane State Awareness during Go-Around. Publication on 16 August 2013.
Accident to the A300-600, registered N155UP, Flight UP1354, Birmingham USA Alabama, 14 August 2013. Information on 14 August 2013.
Accident to the TBM 850, registered N850GC, Vertaizon (63), 8 August 2013. Information on 11 August 2013.
Accident to the Gulfstream G IV registered N823GA on 13 July 2012 in Le Castellet Airport (83). Information on 18 July 2013.


Acccident on 13 July 2012 to the Gulfstream IV registered N823GA and operated by Universal Jet Aviation (UJT) on Le Castellet aerodrome (France). Final report published on 23 October 2015
Accident to the Airbus A321 registered SX-BHS operated by Hermes Airlines and chartered by Air Méditerranée on 29 March 2013 at Lyon Saint-Exupéry Airport (France). Final report published in English on 14 August 2014
Accident on July 28 2013 to the Boeing 777-300 registered F-GSQA at Paris Charles-de-Gaulle Airport (France).
Accident on August 15 2011 to the Bombardier CRJ700 Canadair Regional Jet Series 700 registered F-GRZH at Lyon Saint Exupéry Airport (France).
Accident on June 11 2014 to the Piper J3C registered N61903 at Rolleville (France).
Accident on 24 March 2015 at Prads-Haute-Bléone (Alpes-de-Haute-Provence, France) to the Airbus A320-211 registered D-AIPX operated by Germanwings
Incident to the Airbus A320 registered TS-IMC and operated by tunisair, on 12 March 2013 at Paris Orly Airport (France). Report published in English on 24 September 2014
Accident to the McDonnell Douglas MD-83, registered EC-LTV, on 24 July 2014 in the region of Gossi (Mali). Interim Report published in English on 20 September 2014
Accident on 31 July 2013 to the Cirrus SR-20 registered F-HCPT in Poncins (France), final report published in English on 24 July 2014
Serious incident on 27 February 2012 to the Airbus A330-200 registered F-GZCG and operated by Air France in cruise at FL360 over Tanzania, final report published in English on 24 July 2014
Accident on 25 April 2010 to the Beechcraft 200 registered F-OIAN and operated by Air Alizé at Nadi (Fiji), final report published in English on 24 July 2014
serious incident on 10 May 2012 on Paris le Bourget airport (France) to the Beechcraft 200 King Air registered EC-KNP and operated by Air Taxi and Charter International, final report published in English on 1st July 2014
Accident on 19 August 2012 in Feings (France) to the Cameron Balloons Z-750 registered F-HDJH operated by Air Magic, SARL Flying Circus, published in English on 27 June 2014
Accident on 16 July 2013 in Authon-la-Plaine (France) to Lindstrand 180 A balloon registered F-GSAE operated by Aerfun Montgolfières, published in English on 27 June 2014
Serious incident on 29 June 2010 near Basel-Mulhouse airport (France) between the Airbus A319 registered HB-JZQ operated by easyJet Switzerland and the Airbus A319 registered F-GRHA operated by Air France, published in English on 12 May 2014
Accident on 4 March 2013 just after takeoff from Annemasse (France) to the Beechcraft Premier 1A registered VP-CAZ, published in English on 2 May 2014
Study, events Associated with an engine malfunction, Thielert TAE 125 engines, published in English on 17 April 2014
Accident on 16 October 2012 to the Bombardier CRJ-700 registered F-GRZE and operated by Brit Air at Lorient Lann Bihoué Aerodrome (France), published in English on 18 March 2014
Serious Incident on 16 November 2011 to the Boeing 777-200 registered F-GSPP at Paris Charles de Gaulle Airport (95)
Accident on 24 March 2012 on approach to Tunis Carthage Airport (Tunisia) to the Airbus A319 registered F-GRHU and operated by Air France, published in English on 20 december 2013.
Serious incident on 3 April 2012 on approach to Tel Aviv Ben Gurion Airport (Israel) to the Airbus A320-214 registered F-HEPE and operated by Air France, published in English on 20 december 2013.
Serious incident on 14 september 2010 in Wuxi (China) to the Airbus A319 registered B-6054 and operated by Sichuan Airlines. Civil Aviation Administration of China final report, published on 29 november 2013.
Serious incident on 13 March 2012 to the Airbus A340-300 registered F-GLZU and operated by Air France, on approach to Paris Charles de Gaulle Airport (France) Report published in English on 12 September 2013.
Incident on 25 June 2011 to the ATR 72 registered F-OIQU and operated by Air Tahiti, on initial climb, after take-off from Papeete airport (French Polynesia). Report published in English on 23 July 2013.
Accident on 10 June 2008 to the Airbus 310 registered ST-ATN, Flight Sudan Airways 109 at Khartoum Airport, Sudan. Sudanese Air Accident Investigation Central Directorate report.
Serious incident on 16 January 2010 to aircraft EP-IBB at Stockholm/Arlanda Airport, Stockholm county, Sweden. Statens haverikommission, Swedish Accident Investigation Authority report.
Incident on 17 June 2012 to the BAe146 registered EI-RJW at Basel-Mulhouse-Freiburg Airport (France). Final Report published in English on 25 October 2012.
Incident on 18 November 2011 to the ATR 72-212A registered F-OIQO at Moorea aerodrome (French Polynesia). Final Report published in English on 18 October 2012.
Accident on 17 june 2011 to the Piper PA-39 "Twin Comanche" registered G-AYZE at Peille (France). Final Report published in English on 15 October 2012.
Accident on 28 October 2010 off Adélie Land (Antarctica) to the AS 350 B3 Squirrel registered F-GJFJ operated by SAF HELICOPTERES. Final Report published in English on 24 September 2012.
Serious incident on 15 April 2010 south-east of Brive-la-Roche (19) to the Hawker Beechcraft Corporation Beech B200GT registered F-HSFA operated by SEFA. Final Report published in English on 20 August 2012.
Accident on July 12, 2010 to the Bombardier CRJ700 registered F-GRZN at Bilbao Airport (Spain). Final Report published in English on 25 July 2012.
Accident on 4 April 2011 to the Robin DR400-120 registered F-GABB at Le Touquet Paris-Plage aerodrome (62). Final Report published in English on 24 July 2012.
Accident on 25 July 2010 to the Schweizer 269 C helicopter registered F-GJGQ at Coullons (45). Final Report published in English on 24 July 2012.
Accident on 24 May 2012 to the aircraf Embraer ERJ 145 MP registered F-GUBF at Ljubljana aerodrome (Slovenia), runway 31. Final Report published in English on 24 July 2012.
Serious incident on 17 January 2011 to the Boeing 777 registered F-GSPM in cruise, over the Atlantic ocean. Final Report published in English on 16 July 2012.
Serious incident on 16 August 2008 on take-off from Paris Charles de Gaulle Airport (95) to the Boeing 737-800 registered SU-BPZ operated by AMC Airlines. Final Report published in English on 29 August 2011.
Accident on 1st June 2009 between Rio and Paris to the Airbus A330-203 registered F-GZCP, flight AF 447. Final Report, published on 5 July 2012.

Reporting an occurrence

Contact the BEA



Annual Report