Announcement

Collapse
No announcement yet.

Human Error

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Human Error

    OK, so now I know that a minor mechanical failure followed by a whole series of crew mistakes led to the 1985 near-disaster of China Airlines 006. The captain orders the engineer to find a 3-engine altitude, but almost immediately interrupts that by saying to restart the flamed out engine. The engineer flips a reignite switch without shutting off a bleeder valve in order to provide all oxygen to the engine. The plane banks and the captain doesn't compensate with the rudder because autopilot is on and the captain assumes the rudder compensation has already happened (does rudder position show anywhere on the bank of instruments?). The whole crew looks at attitude meters and don't believe what they are saying and concludes they've all failed. And the plane plunges into clouds, disorienting the crew.

    The NTSB attributes the whole thing to the captain's jet lag.

    One hardly knows where to start, so I'll start here. If reigniting an engine is a 2-step process, why does a very experienced engineer forget 50 percent of it?

  • #2
    WhiteKnuckles,

    I think you have rather misinterpreted some of the prominent points from the NTSB report!

    I assume by your username, and your other questions, that you are a somewhat concerned flyer, and you want to know about these incidents, but unfortunately sometimes a little knowledge can be a bad thing.

    For one - the NTSB specifically REJECT the fact that jet lag was the cause, however they do investigate the possibility, and discuss other work being done on the subject.

    The major cause of the "upset" was the Captain's failure to maintain control of the aircraft. We know it was the Captain's responsibility due to the operators policy at the time. The triggering factor was the No. 4 engine hanging at 1.10 EPR, which caused a resultant yaw that was not managed. Control should have been easily managed... it was not, most likely due to the Captain's distraction of managing the failure and over concentration on the airspeed to the detriment of the other instrumentation.

    To answer your question, no, there is no rudder deflection as part of the crew's immediate instrumentation, however there does not need to be. There is a device that shows if the aircraft is in a "slip", ie not flying in a balanced way. This, and the fact there was a large aileron displacement accompanied by opposite bank, makes it obvious that rudder is required. The captain was also aware that the number 4 engine had hung and so would know he needs to apply left rudder.

    The decision to restart the No 4 engine prior to the aircraft being completely under control and everything stable is one that has most likely added to the confusion and distraction. Yes, the engineer did accomplish the checklist by recall, and he omitted a step. This is somewhat understandable given the unusual attitude they ended up in. Had he selected the bleed switch off, the engine may well have recovered from the hung situation, or it may have been unsuccessful until back in the relight envelope, for which they needed to descend anyway. It may have brought the engine back to normal faster, but the fact the engine was not responding really shouldn't have posed an immediate problem.

    You will be glad to hear that the industry has also learned a lot from these type of incidents, and we have a lot of training to understand how they occur.

    Things that would be done differently today include -

    1) The aircraft would be very closely monitored, and that role would be specifically designated to a crew member who will do next to nothing else. Unlike the China Airlines policy at the time, where the captain was to fly, it is entirely possible that after initial control is made by whoever is the Pilot Flying, the First Officer would be designated the role. This enables the captain to have free reign to do all the "miscellaneous" tasks while the First Officer does nothing except for maintain the safe flight of the aircraft.

    2) The engine would not have a restart attempted until the aircraft is completely under control, and the restart would be done by reference to the checklist.

    It is important to remember that this incident was 25 years ago, and we have learned a lot about the way to get the best out of ourselves in a crew situation in that time.

    They were extremely lucky to get away with it.

    Comment


    • #3
      Interesting. So they just said the captain and the engineer screwed up without attempting to answer why their ample training deserted them.

      How could it be that the captain wouldnt realize he couldn't get the plane out of its banking position without stepping on the rudder pedal? Surely that's pretty elementary. I can't remember if he even stepped on it after taking auto pilot off. If you think he assumed auto-pilot was making a rudder adjustment, surely as soon as he switched off auto-pilot, his foot should immediately hit the pedal.

      Also, what explains his scatter-brained way of saying to find a 3 engine altitude and then immediately saying to reignite the engine. What's the point of doing the second thing before changing altitude?

      Some of the changes mentioned certainly are common sense. But, face it, we don't have a new species reading the gauges. Still the same old homo sapiens. And they can fail in brand new ways.

      Oh, and one more important thing. The German consultant on the program brings up the "scan" factor. According to him, it is what ALL the instruments are saying that matters, not some novel thing that one of them seems to say. He says the captain should have not made decisions before scanning all instruments and processing the cumulative meaning of them.

      Maybe the report rejected jetlag, but my impression is of a guy who isn't getting a coherent picture and is uttering orders impulsively. I've definitely seen that in sleep-deficient people. One thing I'm wondering: Is it because they were Chinese that no one was bringing things to the captain's attention that he was apparently missing? Or is that how cockpit crews always act, with the top officer's reading of the situation being the law.

      Comment


      • #4
        Originally posted by MCM View Post
        unfortunately sometimes a little knowledge can be a bad thing.
        How very true.

        Comment


        • #5
          WhiteKnuckles,

          I'm happy to have a discussion about this (or any other) accident, but I think its only fair to expect you to have actually read and understood what the NTSB actually wrote (not what some show portrayed) before we get into too much detail.

          In your first post, you stated that the NTSB just blamed it on jetlag, and when pointed out that this wasn't true, you've gone to the polar opposite of saying that they just blamed the crew without considering why!

          If you read the NTSB report, you will see that they did in fact search for reasons why the crew failed to respond appropriately. Jetlag was discussed, and they found that based on the nature of the event, it wasn't a prominent feature.

          You can disagree with that if you like, but don't go putting words in the NTSB's mouth.

          Also, what explains his scatter-brained way of saying to find a 3 engine altitude and then immediately saying to reignite the engine. What's the point of doing the second thing before changing altitude?
          The point of doing the second thing as the first, is that it may negate the need to work out the first thing. If you manage to fix the hung engine, then you don't need a 3 engine altitude. Whilst eventually it is something you need to know, it isn't a complete disaster to start a descent down and work it out from there.

          The captain was placed under pressure, and his first response was to calculate the three engine altitude. He then realised that this wasn't in fact a high priority, and asked the engineer to start the process to restart the engine. That was his decision on the day. Based on what we know now, it is more likely that a crew would wait until they were established in a drift down descent, obtained clearances, and everything is settled, before attempting a restart, however that is based on the lessons learned from this (and other) incidents.

          One thing I'm wondering: Is it because they were Chinese that no one was bringing things to the captain's attention that he was apparently missing? Or is that how cockpit crews always act, with the top officer's reading of the situation being the law.
          There has, in the past, been some evidence of a reluctance for junior crew to question the captain in some (often Asian) airlines, due to the more military style of operation. However, this can occur in all airlines. Again, it is something that we have learned, and now train to avoid. That said, I do not think that it is a major part of this accident. Again, if you read the report, you'll see that at certain times the other crew did bring things to the captain's attention, (like the First Officer mentioning the banking) however it wasn't enough to stop the incident.

          "Scanning" the instruments is just a normal part of flying, no special event. Had the captain had an adequate scan then he would have picked up the aircraft deviating from controlled flight earlier, and probably in time to prevent it, however we will never really know.

          There is no doubt he got fixated on one factor, and that is something that can easily happen to anyone. All we can do is train and educate, and give people the tools they need to manage non-normal situations.

          Comment


          • #6
            Try to approach this, er, discussion with an open mind. I can sense the hackles in the air already.

            If you've been a member of the cockpit crew, what do you think the result would be if the captain says "Reignite Enginer 4" and the engineer says, "We can fly with 3 engines, but first the rudder must compensate for the difference of thrust between the two sides. That will require stepping on the left pedal, sir".

            Will that guy be in career trouble? Personally, if it we me, I'd think "I'll save my career later, right now, the captain has to step on the effing pedal or we all DIE!"

            See, that's what I see is a big part of our problem here and on the plane. The issue is not "What is my rank?" or "Who has the most flying experience" or "Who has understood the report". Not any of that stuff. There's only one right answer I see. And it has been known to all the experts in the cockpit for a long time. When an engine flames out, the plane starts to turn until you use the rudder to compensate. Apparently the captain got fixated on the weak engine to the exclusion of the simple step of changing the rudder.

            Or does the report say the rudder was NOT a factor in the dive of the plane?

            Comment


            • #7
              WhiteKnuckles,

              What exactly is your contention? I think in my first post I gave a pretty good summary of the NTSB report, and the lessons learned.

              From the NTSB

              The National Transportation Safety Board determines that the probable cause of
              this accident was the captain's preoccupation with an inflight malfunction and his failure
              to monitor properly the airplane's flight instruments which resulted in his losing control of
              the airplane.
              Pretty good summary if you ask me.

              If you've been a member of the cockpit crew, what do you think the result would be if the captain says "Reignite Enginer 4" and the engineer says, "We can fly with 3 engines, but first the rudder must compensate for the difference of thrust between the two sides. That will require stepping on the left pedal, sir".

              Will that guy be in career trouble? Personally, if it we me, I'd think "I'll save my career later, right now, the captain has to step on the effing pedal or we all DIE!"
              Of course I would say it. If any of the cockpit crew on that flight had noticed that there was no (or insufficient) rudder, then they would have said so too I would imagine. The engineer was preoccupied with relighting the engine, the F/O was preoccupied with the radio and obtaining clearances (although he did mention at one point that the aircraft was banking) and the captain became fixated on one element of flying to the detriment of the others.

              As I said in my last post, while sometimes the captain can be so overpowering that others are afraid to add their input, I can't see that it has happened here... I think they all genuinely missed it, mostly because initialy the autopilot does a pretty good job of hiding it without the use of rudder while they became distracted with other, less important, tasks.

              Or does the report say the rudder was NOT a factor in the dive of the plane?
              Thats a pretty easy one to find out for yourself don't you think?

              The hackles are not up, but I'm sure you can see it is near pointless discussing what a report said when one party hasn't even read it. It is available on the internet, and isn't particulally arduous reading.

              Comment


              • #8
                I really hate this board software. You compose a response, and by the time it is ready to be submitted, the board has logged you out and everything is lost.
                So, let me shorten it this time.
                1. What the NTSB report says is "it is premature at this time" to conclude circadian rhythms are involved, not because they weren't, but because a study on the subject is underway. By now, I'm guessing that study is long finished. And?
                2. Two crews successfully dealt with the underpower problem of Engine 4. In this case, the captain and the flight engineer both abandoned their training, making the situation worse. You tell me that "we've found better ways". So you've found a foolproof way to prevent experienced crews from just forgetting their training? Apparently the flight officer was just an interested bystander as his plane headed for oblivion. Doesnt he have a job to do?
                3. The report says 747 crews are much more monitors than active flyers. That's inherent to modern planes, right? If you talk about computer monitors, I may be the most experienced person here, since that's been a big part of my work since 1973. And I know intimately what automating my job does. I've been dragged kicking and screaming through phase after phase of making me a passive pair of eyes. I've got a lot of sympathy for these guys who once told the plane everything it would do and now watch LEDs and gauges. But I'm a lot less passive than this captain at seizing control when the situation warrants.


                "Isn't particularly arduous reading". What a joke. The conclusions aren't, but trying to read everything leading up to them is something else again.

                Comment


                • #9
                  Pressure Simulation

                  Another thing this near-disaster says to me is that "flight simulation" is probably a necessary but insufficient training mode for air emergencies. What they really need is to "simulate" the probability of killing yourself and everyone else on board. How can they know how each crew member will handle possible disaster when they give them training circumstances where the student's brain knows nobody's gonna die if they make a mistake? Bobby Kennedy wrote after the Cuban missile crisis that the security council learned that the human brain, faced with catastrophe, ceases to be flexible and open. The perception of alternatives clamps down very hard. There's a tendency to think in magic bullet terms. Now this was decades BEFORE CA006. I'm surprised the NTSB didn't already have research findings on such matters. I guess they got around to cybernetics a lot later. I would agree that the scope for discussion is limited to a much narrower field when seconds count. But despite the human perception, it is not GONE. I've actually seen people temporize in the middle of a crisis in a banking environment where loss of customers was the potential cost. But it does seem to me that silence can be very costly. The NTSB puts that under the umbrella of "coordination", but we who dislike bureaucratese just say "open your mouth!"

                  Comment


                  • #10
                    I can honestly say that having read your post 3 times, I have no idea what you are actually trying to say. Are you trying to just put this accident into the "the other crew didn't speak up" basket?

                    Are you trying to say the NTSB didn't do their job properly? That their conclusions are wrong? I just don't understand.

                    People have known about the limitations of simulators for simulating real life situations for decades as you correctly point out. The NTSB are fully aware of that fact. I have been involved in real life engine failures, and yes you are surprised and the blinkers do come on. The ability to "think outside the square" diminishes, and you have little free brainspace for other than basic thought. That is exactly why simulator training is like it is... to practice the correct response so that when the event happens, the first response is the correct one, enabling time for the situation to calm down and well thought out decisions can be made. The more experience you have, the less the initial surprise and the better the decisions you make are.

                    Comment


                    • #11
                      The more experience you have, the less the initial surprise and the better the decisions you make are.
                      Provided you survive them. Trouble is that a lot of these surprises happen for the first and last time. I'm not faulting NTSB. They, after all, are the ones that brought up jet lag and circadian rhythms. You evidently dismissed these things much more quickly than they did. It is also clear that they pointed exactly where the faults were. The crew did not do the responses in the right order and therefore made a minor situation a near disaster. And what's more, it wasn't lack of either training or experience. It was simply failure to EXECUTE what they knew. Is there any evidence that the captain tried to let autopilot fix the attitude problem because he didn't KNOW the function of the rudder pedal? Wouldnt that kind of ignorance keep him from being captain? Well, I'm not beating the dead horse. I'm just saying that I see one of two things: Lack of a team function, which may be normal or abnormal. The NTSB report seemed to think it was abnormal. I think they made an analogy to that everglades crash where the whole crew was fixated on one gauge, and unaware the plane was plunging.

                      So what is most likely to fix this situation? Refresher courses in "flying as a team"? Annual written tests to make sure original training on a plane are still as fresh as the end of the first training? Pressure training? You can't duplicate a for-real disaster situation, but you can create intense pressure of some kind and then ask the crew to perform a team process in a brief period of time. Maybe they need documentation that reminds them of what the category of each outage is.

                      One thing I wonder about. If there's a correct sequence to the bleeder valve control and the ignition switch, why can that be done wrong? In my car, I have to have two or three things in correct setting to get my car to start. Since I drive alone, the stakes are pretty low compared to a jumbo jet.

                      One other thing I wondered tonight. My local major league team just sent a pitcher to the minors for a brush up. Should this captain have gone to a plane with less capacity for a while?

                      Comment


                      • #12
                        I'm not faulting NTSB. They, after all, are the ones that brought up jet lag and circadian rhythms. You evidently dismissed these things much more quickly than they did.
                        Where did you get that from? I did not quickly dismiss anything... however I have looked at this accident and have no reason to question the NTSB finding. Fatigue clearly affects the way you can react, and whilst it wouldn't have helped the situation, it wasn't the causing factor. I hardly think the fact that I have come to the same conclusion as the NTSB automatically means I have dismissed it.

                        I think you are trying to re-invent the wheel somewhat here.

                        This incident occurred 25 years ago. We DO CRM and Human Factors training now.

                        So what is most likely to fix this situation? Refresher courses in "flying as a team"? Annual written tests to make sure original training on a plane are still as fresh as the end of the first training? Pressure training? You can't duplicate a for-real disaster situation, but you can create intense pressure of some kind and then ask the crew to perform a team process in a brief period of time. Maybe they need documentation that reminds them of what the category of each outage is.
                        We do refresher courses in "flying as a team"... Crew Resource Management training.

                        We do written and oral as well as simulator training to ensure that we are upto standard (and to a higher standard than after initial training).

                        We do Line Oriented Flight Training... simulator sessions that set an uknown scenario and force the crew to use their team skills to come to a safe and satisfactory outcome, and then we debrief and review what went well, what didn't, and how we can improve.

                        The flight engineer forgot the Bleed Air valve... people are fallable. We learned our lessons... we now make sure that things like relights are done in a very controlled and planned manner, unlike what occurred in this case.

                        You aren't saying anything new... we know that the team didn't operate as well as it should have, and that meant that they didn't catch the errors that were made. The captain made a mistake... the rest of the crew didn't catch it. The engineer made a mistake... the rest of the crew didn't catch it. We have learned a lot from this style of mistake.

                        What would you have the industry do now? We have already enacted training to try and reduce this style of error, and while it may never be completely avoidable, it has certainly been reduced.

                        One other thing I wondered tonight. My local major league team just sent a pitcher to the minors for a brush up. Should this captain have gone to a plane with less capacity for a while?
                        This was an aircraft he had experience on, and just because an aircraft is smaller does not mean it is easier to fly. It is always difficult to know what to do with a crew after an incident like this, given ultimately he is possibly one of only a few pilots ever to sucessfully recover a 747 from such an unusual attitude.

                        Comment


                        • #13
                          Again the 25 years ago. Tell me how the human organism is different now than in 1985.

                          I looked for something a bit more recent. Here is another crash that didn't need to happen:

                          Flight operations can engender fatigue, which can affect flight crew performance, vigilance, and mood. The National Transportation Safety Board (NTSB) requested the NASA Fatigue Countermeasures Program to analyze crew fatigue factors in an aviation accident that occurred at Guantanamo Bay, Cuba. There are specific fatigue factors that can be considered in such investigations: cumulative sleep loss, continuous hours of wakefulness prior to the incident or accident, and the time of day at which the accident occurred. Data from the NTSB Human Performance Investigator's Factual Report, the Operations Group Chairman's Factual Report, and the Flight 808 Crew Statements were analyzed, using conservative estimates and averages to reconcile discrepancies among the sources. Analysis of these data determined the following: the entire crew displayed cumulative sleep loss, operated during an extended period of continuous wakefulness, and obtained sleep at times in opposition to the circadian disposition for sleep, and that the accident occurred in the afternoon window of physiological sleepiness. In addition to these findings, evidence that fatigue affected performance was suggested by the cockpit voice recorder (CVR) transcript as well as in the captain's testimony. Examples from the CVR showed degraded decision-making skills, fixation, and slowed responses, all of which can be affected by fatigue; also, the captain testified to feeling "lethargic and indifferent" just prior to the accident. Therefore, the sleep/wake history data supports the hypothesis that fatigue was a factor that affected crewmembers' performance. Furthermore, the examples from the CVR and the captain's testimony support the hypothesis that the fatigue had an impact on specific actions involved in the occurrence of the accident.
                          That's your cue to tell me you read this three times and didn't see what it was about.

                          This is 8 years later, and it sure looks to me that there is something in common with the Flight 006 scenario.

                          Comment


                          • #14
                            The human organism is no different.

                            What is different is that we have far more experience in understanding our own failings, and are getting far better at not allowing these sort of accidents to occur.

                            Are we perfect? No. It is an ongoing process. We'll never eliminate mistakes. But we sure has heck know our own limitations and the advantages of good CRM a lot better than we did when these incidents were occurring. There are elements of Human Factors in most accidents, and we find out what we find out and learn from it. There is not much more that can be done.

                            Yes, fatigue is an issue, and it is a constant battle between operational efficiency and fatigue management.

                            Again, I'll ask... do you have a point you are trying to make? You ask questions and then immediately attempt to disprove the answer.

                            Or are you just trying to say that you saw a show about an aircraft incident and you have decided the cause, so we should all congratulate you?

                            Comment


                            • #15
                              WhiteKnuckles, what part of 'Human Error' don't you understand? What are you trying top prove by pursuing this line of questioning? Every day all around the planet, humans make errors. I watched my 4yo son spill his drink at breakfast. Could it have been prevented? Probably. Will it happen again? Undoubtedly.

                              The odds of losing your life in an air crash are much lower that the odds of your death on the road. Risk is an acceptable part of life - it has to be. Nothing will be perfect, but you can bet as MCM has pointed out that when things are learned from air crashes, equipment and or procedures where possible are modified to try and counter the risk of the problem occuring again. But as long as humans are in the loop, there is always the risk it will occur again.

                              If you want to minimise the risk, avoid flying in most of Africa, or on local carriers in places like Indonesia (IMHO). Other than that I trust the bloke in the pointy end, the person he talks to at ATC, the mechanic who put the bolts into the wing, the construction dude in Renton/marsellies, the designer and his computer.

                              Comment

                              Working...
                              X