cedarjet wrote:

And shockingly bad flying skills.

Really? The PIC left an "A" out of a manual entry in the FMC when the route wouldn't auto-load for him. He decided to engage the AP right after takeoff to reduce workload following an abnormally hectic pre-flight period and apparently having to monitor multiple MEL restrictions during takeoff (which sounds to me like a highly appropriate use of automation; I bet B's and AB's training departments probably would agree). The low level chaos due to the missing "A" and the airplane having VNAV engaged at a relatively low altitude took the crew a little while to sort out but they did.

I will grant, based on some of the rants in this thread, that many feel typing appears to have become accepted as a basic flying skill these days but still...

zeke wrote:

BoeingGuy wrote:

I don’t agree. The 777 Autothrottle has the auto-reengage function. The 737 does not. It’s just that the Asiana crew screwed up the mode selection so badly that Auto-reengage didn’t function. I’m sure you are aware the Autothrottle was changed to better protect against such crew errors. The 777 also has the AIRSPEED LOW caution. The Asiana crew was so asleep at the switch that they weren’t even monitoring their Airspeed. It wouldn’t surprise me if the PF didn’t have his hands on the Thrust Levels (like the EK 521 captain didn’t have his hands on the thrust levers during a balked landing).

At what point do you have to design an airplane to account for someone who doesn’t show basic flying skills?

I will just let the facts speak for themselves

"Boeing also has a "caution tone" for low airspeed on the 777, along with versions of the 747, according to an FAA panel studying the issue. But other planes have a verbal warning such as "airspeed low." Clifford said the 777 tone "has proven to not capture the pilot's attention in these low airspeed circumstances.""

"Airbus has a warning that says "speed speed speed" every five seconds between 2,000 and 100 feet when an aircraft goes too slow. The equipment is on the A320 family since 1995 and on all A330, 340, 350 and 380 aircraft.

Boeing puts a warning that says "airspeed low, airspeed low" on 737-600, -700, -800 and -900 planes, according to an FAA panel studying the issue. Boeing began putting the warning on 737s after a Turkish Airlines crash Feb. 25, 2009, near Amsterdam that killed nine and injured 117."

from https://www.usatoday.com/story/news/nat ... 4/2508863/

I assume you are familiar with EICAS alerting. The AIRSPEED LOW alert is on the 747-8, 777, 787, and KC-46. Not sure about the 747-400.

It is annunciated as an EICAS caution. This includes stimulating two senses. There is a four tone aural, the Master Caution Light illuminates, and the AIRSPEED LOW text shows on EICAS.

Boeing has lots of EICAS Cautions, as I assume does Airbus. I’ve never seen data that shows Caution alerts are insufficient to get the crew’s attention.

The trigger for AIRSPEED low is well above Stick Shaker, so you are not yet near stall at that point. The alert triggers at 70% up the Minimum Maneuver Speed Amber Band - or 30% from the top. A recent change on some models moves the trigger to top of Amber Band when Flaps are 25 or 30.

Also, as mentioned earlier the Autothrottle Auto-Reengage function was enhanced to function when in HOLD mode or other non-SPD modes. This covers conditions where the crew may have selected an incorrect mode.

BoeingGuy wrote:

I’ve never seen data that shows Caution alerts are insufficient to get the crew’s attention.

From the article "Clifford said the 777 tone "has proven to not capture the pilot's attention in these low airspeed circumstances.""

BoeingGuy wrote:

This covers conditions where the crew may have selected an incorrect mode.

If something is a wrong mode, it should be inhibited to prevent selection. Its like not letting you remove the keys form the car until its been put in park, every driver would know not to do that, yet hundreds of people every year did it.

zeke wrote:

BoeingGuy wrote:

I’ve never seen data that shows Caution alerts are insufficient to get the crew’s attention.

From the article "Clifford said the 777 tone "has proven to not capture the pilot's attention in these low airspeed circumstances.""

BoeingGuy wrote:

This covers conditions where the crew may have selected an incorrect mode.

If something is a wrong mode, it should be inhibited to prevent selection. Its like not letting you remove the keys form the car until its been put in park, every driver would know not to do that, yet hundreds of people every year did it.

First of all it’s clear you’ve always had a bias towards Boeing. Secondly, there are possible issues with that which you clearly don’t grasp. This isn’t putting a car in park. It’s an airplane.

What, are you going to say inhibit selection of FLCH below 1000 feet. What is the required integrity of the inhibit? What if the altitude is in error and you inhibit FLCH when it should be used? What if there is another operational use case for selecting a mode at a certain phase of flight, yet it was inhibited to cover a different condition? There could be unintended consequences if you start designing stuff like this.

That said, there are conditions where certain modes are inhibited.

Just because Clifford said it, what data supports it? I’ve seen no such data and I have exposure to that area.

Nice try continuing to bash Boeing, but your arguments don’t hold water.

BoeingGuy wrote:

Just because Clifford said it, what data supports it? I’ve seen no such data and I have exposure to that area.

Nice try continuing to bash Boeing, but your arguments don’t hold water.

I am not “Bob Clifford, a Chicago attorney specializing in aviation-disaster cases”, nor the author of the USA Today article. Please direct your comments to them if you feel what is published is inaccurate.

I just presented the facts, and provided the source.

hivue wrote:

cedarjet wrote:

And shockingly bad flying skills.

Really? The PIC left an "A" out of a manual entry in the FMC when the route wouldn't auto-load for him. He decided to engage the AP right after takeoff to reduce workload following an abnormally hectic pre-flight period and apparently having to monitor multiple MEL restrictions during takeoff (which sounds to me like a highly appropriate use of automation; I bet B's and AB's training departments probably would agree). The low level chaos due to the missing "A" and the airplane having VNAV engaged at a relatively low altitude took the crew a little while to sort out but they did.

I will grant, based on some of the rants in this thread, that many feel typing appears to have become accepted as a basic flying skill these days but still...

Sorry you don’t level off at 500 feet and fly level for 30 seconds without noticing. That’s just unbelievable that no one is monitoring flight instruments for that long only seconds after takeoff. Also once they realised there was a problem, one of them disconnected the automation, got it climbing, and immediately reengaged the autopilot which has already proved itself to not be trustworthy, and sure enough it immediately starts descending again back to 500 ft. This is the worst airmanship I’ve ever heard of in my life. “Took the crew a little while to sort out but they did”?! Dude, it’s not a stuck outflow valve in the cruise. 500 ft AGL is the height of a 35 floor building. Barrelling along at that height for 30 seconds with the thing shouting don’t sink and pull up and not even knowing it. Singapore Airlines (like Emirates recently) got super lucky, there wasn’t a block of flats, a crane, a TV tower in their way. Craziness.

hivue wrote:

cedarjet wrote:

And shockingly bad flying skills.

Really? The PIC left an "A" out of a manual entry in the FMC when the route wouldn't auto-load for him. He decided to engage the AP right after takeoff to reduce workload following an abnormally hectic pre-flight period and apparently having to monitor multiple MEL restrictions during takeoff (which sounds to me like a highly appropriate use of automation; I bet B's and AB's training departments probably would agree). The low level chaos due to the missing "A" and the airplane having VNAV engaged at a relatively low altitude took the crew a little while to sort out but they did.

I will grant, based on some of the rants in this thread, that many feel typing appears to have become accepted as a basic flying skill these days but still...

Why not just hand-fly for the first few minutes to stay sharp? High workload can be managed - that's what CRM and all hands on deck for takeoff/landing are for. I can't see how an 'abnormally hectic preflight period' is an excuse for not having/using stick and rudder skills. It still just sounds absurd to me that due to SOPs and what not at certain airlines that it's simply a given to start pushing buttons at VR+15, especially with crew who lack aforementioned 'reserve aeronautical knowledge'.

Chilling to read through the thread and watch Juan Browne's video. I wonder if there is any video of the plane flying over homes, etc?

zeke wrote:

BoeingGuy wrote:

Just because Clifford said it, what data supports it? I’ve seen no such data and I have exposure to that area.

Nice try continuing to bash Boeing, but your arguments don’t hold water.

I am not “Bob Clifford, a Chicago attorney specializing in aviation-disaster cases”, nor the author of the USA Today article. Please direct your comments to them if you feel what is published is inaccurate.

I just presented the facts, and provided the source.

The closed loop / blinkered quality / safety culture, pre-MAX grounding, remains strong.

smartplane wrote:

zeke wrote:

BoeingGuy wrote:

Just because Clifford said it, what data supports it? I’ve seen no such data and I have exposure to that area.

Nice try continuing to bash Boeing, but your arguments don’t hold water.

I am not “Bob Clifford, a Chicago attorney specializing in aviation-disaster cases”, nor the author of the USA Today article. Please direct your comments to them if you feel what is published is inaccurate.

I just presented the facts, and provided the source.

The closed loop / blinkered quality / safety culture, pre-MAX grounding, remains strong.

I’m afraid I don’t quite see the inherent nexus between Boeing safety culture and crews relying on automation in phases of flight that don’t require its use.

Aaron747 wrote:

hivue wrote:

cedarjet wrote:

And shockingly bad flying skills.

Really? The PIC left an "A" out of a manual entry in the FMC when the route wouldn't auto-load for him. He decided to engage the AP right after takeoff to reduce workload following an abnormally hectic pre-flight period and apparently having to monitor multiple MEL restrictions during takeoff (which sounds to me like a highly appropriate use of automation; I bet B's and AB's training departments probably would agree). The low level chaos due to the missing "A" and the airplane having VNAV engaged at a relatively low altitude took the crew a little while to sort out but they did.

I will grant, based on some of the rants in this thread, that many feel typing appears to have become accepted as a basic flying skill these days but still...

Why not just hand-fly for the first few minutes to stay sharp? High workload can be managed - that's what CRM and all hands on deck for takeoff/landing are for. I can't see how an 'abnormally hectic preflight period' is an excuse for not having/using stick and rudder skills. It still just sounds absurd to me that due to SOPs and what not at certain airlines that it's simply a given to start pushing buttons at VR+15, especially with crew who lack aforementioned 'reserve aeronautical knowledge'.

Great idea, if the company allows it or, at least, doesn't discourage it.
I usually handfly till at least 10.000 ft, if circumstances permit (fatigue, other traffic, my colleague, weather, etc). The same during descent/approach, but that's partly because my company thinks it is a good thing to do, which is not the case at every company in the world.

Is there any more info available by now?
Did the crew try to climb but was unable to do so for some reason like wrong trim, loading or thrust setting or did they miss to climb blindly trusting the FD needles that did not move?
Will anybody not from Dubai investigate this?

smartplane wrote:

zeke wrote:

BoeingGuy wrote:

Just because Clifford said it, what data supports it? I’ve seen no such data and I have exposure to that area.

Nice try continuing to bash Boeing, but your arguments don’t hold water.

I am not “Bob Clifford, a Chicago attorney specializing in aviation-disaster cases”, nor the author of the USA Today article. Please direct your comments to them if you feel what is published is inaccurate.

I just presented the facts, and provided the source.

The closed loop / blinkered quality / safety culture, pre-MAX grounding, remains strong.

Don’t slam my commitment to safety at my job when you have no clue what you are talking about. It’s pretty offensive.

BoeingGuy wrote:

smartplane wrote:

BoeingGuy wrote:

I don’t agree. The 777 Autothrottle has the auto-reengage function. The 737 does not. It’s just that the Asiana crew screwed up the mode selection so badly that Auto-reengage didn’t function. I’m sure you are aware the Autothrottle was changed to better protect against such crew errors. The 777 also has the AIRSPEED LOW caution. The Asiana crew was so asleep at the switch that they weren’t even monitoring their Airspeed. It wouldn’t surprise me if the PF didn’t have his hands on the Thrust Levels (like the EK 521 captain didn’t have his hands on the thrust levers during a balked landing).

At what point do you have to design an airplane to account for someone who doesn’t show basic flying skills?

Surely Boeing has a duty to ensure more consistency between MAX, 777 and 787, especially where common / 'quick change' ratings are offered / promoted? When buying from Boeing, there is a disconnect between the 'owners' of the various model families, which from your comments, extends even deeper to design, development and standards.

Again. More arm chair quarterback. The 737 Max does in fact have an Airspeed Low alert and a host of other enhancements. It’s the customers who balk at making the 737 Max into a 787 for commonality reasons.

The newer models do have a lot of commonality. From an avionics and flight controls perspective, the 777-9 is very similar to a 787. The 787 is getting some never upgrades developed for the 777-9. But you can’t just go back and make a 737 or 767 and turn it into a 787.

I’m guessing the new 87 changes will be incorporated into the IGW airplane?

Noshow wrote:

Is there any more info available by now?
Did the crew try to climb but was unable to do so for some reason like wrong trim, loading or thrust setting or did they miss to climb blindly trusting the FD needles that did not move?
Will anybody not from Dubai investigate this?

I just saw something on this event. It’s still under investigation as to exactly what happened and why. Boeing is investigating and I’m sure a fix will be implemented, if needed.

We can all argue on what level of automation is needed, on what conditions shouldn't be allowed by the aircraft, etc.
But unless the plane doesn't need a pilot at all (fully automated), then their needs to be pilots involved and monitoring, and performing the portions of flight that aren't automated.
When those pilots aren't paying attention, or are incompetent, then bad shit happens. And you can't blame the aircraft.

smartplane wrote:

zeke wrote:

BoeingGuy wrote:

Just because Clifford said it, what data supports it? I’ve seen no such data and I have exposure to that area.

Nice try continuing to bash Boeing, but your arguments don’t hold water.

I am not “Bob Clifford, a Chicago attorney specializing in aviation-disaster cases”, nor the author of the USA Today article. Please direct your comments to them if you feel what is published is inaccurate.

I just presented the facts, and provided the source.

The closed loop / blinkered quality / safety culture, pre-MAX grounding, remains strong.

So it’s Boeing and aircraft fault that the crew allowed this aircraft end up in this state.

bigb wrote:

smartplane wrote:

zeke wrote:

I am not “Bob Clifford, a Chicago attorney specializing in aviation-disaster cases”, nor the author of the USA Today article. Please direct your comments to them if you feel what is published is inaccurate.

I just presented the facts, and provided the source.

The closed loop / blinkered quality / safety culture, pre-MAX grounding, remains strong.

So it’s Boeing and aircraft fault that the crew allowed this aircraft end up in this state.

What angered me is his not so hidden suggestion that I follow the proverbial 737 Max poor safety culture in my work. That is offensive and definitely not correct.

Just because I may not agree with some Boeing bashing doesn’t mean I follow that supposed Pre-Grounding culture.

Revelation wrote:

Gremlinzzzz wrote:

Are they dealing with it or not? They are prepared and handling it. When was the last time you consistently saw a plane brought down by such? We really are arguing for the sake of it.

That's your perception. My perception is that you're continuously changing the argument to suite your narrative. In particular you wrote "With automation, he will rarely have to intervene unless you have a Qantas 72 edge of the envelope case scenario" and when I showed manual intervention for TCAS RA was not an edge of the envelope scenario you could have simply thanked me for the correction because you are now admitting I am right and manual intervention is not an edge of the envelope thing, but instead you try to change the argument to the accident rate.

Bottom line is you now tacitly admit intervention is not an "edge of the envelope" thing which totally destroys your earlier argument saying manual flying skills are not important.

None of the effects of the data-spike failure mode were observed until the failure mode was triggered, and the effects recurred continuously until the unit was shut down. In addition, the failure mode could not be replicated during subsequent testing using a wide variety of potential triggers, indicating that the failure mode could only be triggered by either a rare event, or under rare circumstances.
This behaviour indicated that the disruption to the CPU module’s processing was triggered by a single event rather than a series of events (that is, it was very unlikely that a separate event triggered each data spike or other processing problem). https://www.atsb.gov.au/media/3532398/ao2008070.pdf

Page 155.

Some of us actually read reports.

It is also fun reading you state that I stated that manual flying skills are not important. I stated that, and I believe repeated that automation was there to get rid of pilot error and humans behaving differently when the same scenario presents itself.

Has automation significantly reduced aviation accidents? Absolutely! Even a mad man would not argue otherwise because empirical evidence states so.

Do airlines want pilots showing off how great their manual skills are? No. Watch the automation,and if something does go wrong, step in.

Do the likes of Boeing/Airbus/Embraer dream of going back to pilots crashing equipment because we need more hand flying? We are never going back to those days, and that is a blessing.

Gremlinzzzz wrote:

Revelation wrote:

Gremlinzzzz wrote:

Do airlines want pilots showing off how great their manual skills are? No. Watch the automation,and if something does go wrong, step in.

Problem is, that if you only watch the automatics, in will lose the skills to adequately "step in" when something does go wrong.

There is a rise in incidents where pilots react wrongly to a minor problem because they don't understand what the aircraft is doing wrong and why. And they lost the skills to take over in a timely and suitable fashion.

I am glad my company encourages manual flying, as long as you pick the right place and time just to keep the basic skills up.

We regularly deselect the AP/AT during descent at FL whatever, followed by a manual approach or visual or on departure only switch them on when we get bored above FL 150 or do. I wouldn't do that at night on, for example, Kilimajaro after a busy flight with tunderstorms and spotty ATC, or a very busy airspace at JFK with gusty weather, but I am convinced that with a maximum of 2 sectors at the controle per week (semi long haul), regularly taking control from the computer increases safety and jeeps me from turning into a flightdirector slave.

Eikie wrote:

Gremlinzzzz wrote:

Revelation wrote:

Problem is, that if you only watch the automatics, in will lose the skills to adequately "step in" when something does go wrong.

There is a rise in incidents where pilots react wrongly to a minor problem because they don't understand what the aircraft is doing wrong and why. And they lost the skills to take over in a timely and suitable fashion.

I am glad my company encourages manual flying, as long as you pick the right place and time just to keep the basic skills up.

We regularly deselect the AP/AT during descent at FL whatever, followed by a manual approach or visual or on departure only switch them on when we get bored above FL 150 or do. I wouldn't do that at night on, for example, Kilimajaro after a busy flight with tunderstorms and spotty ATC, or a very busy airspace at JFK with gusty weather, but I am convinced that with a maximum of 2 sectors at the controle per week (semi long haul), regularly taking control from the computer increases safety and jeeps me from turning into a flightdirector slave.

Risk/Reward.

There risk in coming up with automation done well is less than the rewards gained from safety, therefore, automation of the cockpit is only increasing, not decreasing. We have gone through this propaganda before when automation was introduced, and it will continue showing up. Either way, the benefits outweigh the 'concerns' we keep reading about, and the results show this to be true.

Pilots were making vastly more errors even when they were doing more hand flying i.e. more skilled. So that alone kills the folly of that argument.

Gremlinzzzz wrote:

Pilots were making vastly more errors even when they were doing more hand flying i.e. more skilled. So that alone kills the folly of that argument.

Not so fast - you're talking about disparate bell curves running in parallel. Here, we're talking about the automation vs. hand flying skill degradation curve. The era you were referring to involves many more variables - primitive navigational aids, non-precision approaches in many parts of the world, less understanding of weather phenomena, lack of onboard traffic avoidance systems, primitive simulator training, less robust and/or no CRM training, overconfidence from military careers, whatever the case may be.

Aaron747 wrote:

Not so fast - you're talking about disparate bell curves running in parallel. Here, we're talking about the automation vs. hand flying skill degradation curve.

Fundamentally he is correct. Whenever hand flying either back then or today the amount of spare capacity any pilot has is less than if you are monitoring.

If you were to ask any pilot to fly an attitude, and a thrust setting, anyone can do that, they all know how to do it. The issue is not how to fly, the issue is knowing what they should fly, having the spare capacity to know what they want, once you know what you want, how to do it is simple.

Same applies with automation, once you know what you want, how to do it is easy.

zeke wrote:

Aaron747 wrote:

Not so fast - you're talking about disparate bell curves running in parallel. Here, we're talking about the automation vs. hand flying skill degradation curve.

Fundamentally he is correct. Whenever hand flying either back then or today the amount of spare capacity any pilot has is less than if you are monitoring.

If you were to ask any pilot to fly an attitude, and a thrust setting, anyone can do that, they all know how to do it. The issue is not how to fly, the issue is knowing what they should fly, having the spare capacity to know what they want, once you know what you want, how to do it is simple.

Same applies with automation, once you know what you want, how to do it is easy.

If he is referring to statistical comparison, he is incorrect due to the change in variable sets. That's what I was getting at. Comparing rote accident rates between eras requires considering far more factors than simply amount of automation or hand flying.

zeke wrote:

Aaron747 wrote:

Not so fast - you're talking about disparate bell curves running in parallel. Here, we're talking about the automation vs. hand flying skill degradation curve.

Fundamentally he is correct. Whenever hand flying either back then or today the amount of spare capacity any pilot has is less than if you are monitoring.

If you were to ask any pilot to fly an attitude, and a thrust setting, anyone can do that, they all know how to do it. The issue is not how to fly, the issue is knowing what they should fly, having the spare capacity to know what they want, once you know what you want, how to do it is simple.

The only way to train that spare capacity needed to operate safely while flying manual, is by flying manual regularly. That mental capacity comes in handy when dealing with situations on board (routing issues, nnc, cabin issues,…)

Someone suggested there was an issue with the routing in the fmc right after takeoff which prompted the PF to engage the autopilot (as suggested by the company policy anyway).
2 things (ok I was not on board):
- what should you have to change at 400 ft on a normal departure that should require both pilot’s full attention to the FMC as such that they don’t monitor their airplane anymore? There should be a back up sept with conventional nav aids and if it’s an rnav issue, request vectors. If it’s an issue with the SID (wrong Sid loaded), sorry but why take off?
- levels of automation: going up a level doesn’t necessarily mean reducing workload… we know this since 1998 already, children of the magenta

Aaron747 wrote:

Gremlinzzzz wrote:

Pilots were making vastly more errors even when they were doing more hand flying i.e. more skilled. So that alone kills the folly of that argument.

Not so fast - you're talking about disparate bell curves running in parallel. Here, we're talking about the automation vs. hand flying skill degradation curve. The era you were referring to involves many more variables - primitive navigational aids, non-precision approaches in many parts of the world, less understanding of weather phenomena, lack of onboard traffic avoidance systems, primitive simulator training, less robust and/or no CRM training, overconfidence from military careers, whatever the case may be.

Automation took care of all that, and not only in the cockpit, but at airports too.

Automation kicked in, and the navigator, rulers protractors and all was booted out. More automation came in and the flight engineer was kicked out too. Safety went up despite the propaganda that was waged. Today, there are fewer accidents and all of that goes back to automation with fewer buttocks occupying the cockpit seats.

OEM's got through their biggest problem which was pilots crashing planes. Regulators got something they wanted, a safer work environment. Governments worldwide got fewer accidents and passengers a scenario where the expectation was to board a flight and to get home safe.

Drivers today are less 'skilled' than those that were driving manual with no driver assists, yet driving is less stressful.
Architects are less skilled than those who used to work with T-Squares; automation gets rid of a lot of mistakes in their work.

Automation makes work easier and no one gives a damn if the end result is better results. In aviation, all people care about is whether they get into a plane and get home safe. Automation done well is better than sky jockeys showing off their skills. That is not an argument we are having because their is empirical data to show it is the case.

Anything else?

Aaron747 wrote:

zeke wrote:

Aaron747 wrote:

Not so fast - you're talking about disparate bell curves running in parallel. Here, we're talking about the automation vs. hand flying skill degradation curve.

Fundamentally he is correct. Whenever hand flying either back then or today the amount of spare capacity any pilot has is less than if you are monitoring.

If you were to ask any pilot to fly an attitude, and a thrust setting, anyone can do that, they all know how to do it. The issue is not how to fly, the issue is knowing what they should fly, having the spare capacity to know what they want, once you know what you want, how to do it is simple.

Same applies with automation, once you know what you want, how to do it is easy.

If he is referring to statistical comparison, he is incorrect due to the change in variable sets. That's what I was getting at. Comparing rote accident rates between eras requires considering far more factors than simply amount of automation or hand flying.

Automation got rid of the other variables.

Do you know how data analysis or testing is done? You vary one thing and all the rest remain constant.

The only variable measured is flying without automation and flying with it. This is why it is always stated, and as a fact that automation made flying safer.

I am all ears as to how you are going to make considerations and measure more than one variable.

OOSFS wrote:

The only way to train that spare capacity needed to operate safely while flying manual, is by flying manual regularly. That mental capacity comes in handy when dealing with situations on board (routing issues, nnc, cabin issues,…)

I disagree, simulator time can do that.

OOSFS wrote:

Someone suggested there was an issue with the routing in the fmc right after takeoff which prompted the PF to engage the autopilot (as suggested by the company policy anyway).
2 things (ok I was not on board):
- what should you have to change at 400 ft on a normal departure that should require both pilot’s full attention to the FMC as such that they don’t monitor their airplane anymore? There should be a back up sept with conventional nav aids and if it’s an rnav issue, request vectors. If it’s an issue with the SID (wrong Sid loaded), sorry but why take off?
- levels of automation: going up a level doesn’t necessarily mean reducing workload… we know this since 1998 already, children of the magenta

This is a cockpit discipline issue you are describing, not an automation issue, and not a hand flying issue. Airbus calls this their golden rules. https://www.smartcockpit.com/docs/Golden_Rules.pdf

Should always have one person heads up minding the shop at all times, and there is never a reason for any pilot to dive into the FMC immediately after rotation.

Gremlinzzzz wrote:

Aaron747 wrote:

zeke wrote:

Fundamentally he is correct. Whenever hand flying either back then or today the amount of spare capacity any pilot has is less than if you are monitoring.

If you were to ask any pilot to fly an attitude, and a thrust setting, anyone can do that, they all know how to do it. The issue is not how to fly, the issue is knowing what they should fly, having the spare capacity to know what they want, once you know what you want, how to do it is simple.

Same applies with automation, once you know what you want, how to do it is easy.

If he is referring to statistical comparison, he is incorrect due to the change in variable sets. That's what I was getting at. Comparing rote accident rates between eras requires considering far more factors than simply amount of automation or hand flying.

Automation got rid of the other variables.

Do you know how data analysis or testing is done? You vary one thing and all the rest remain constant.

The only variable measured is flying without automation and flying with it. This is why it is always stated, and as a fact that automation made flying safer.

I am all ears as to how you are going to make considerations and measure more than one variable.

Absolutely aware - data analysis is a major part of my work. What you are doing is varying the scenario, then comparing results, and that’s simply not logical. The accident rates due to pilot error in the 1960s encompass the complete picture of failure modes from that time, and the same is true of the modern era where safety is dramatically improved.

Your argument is also smeared by using emotionally-loaded descriptors like ‘sky jockeys’ - usually such language indicates some degree of animus and reduces credibility. Would you use such phrasing in a report?

Aaron747 wrote:

Gremlinzzzz wrote:

Aaron747 wrote:

If he is referring to statistical comparison, he is incorrect due to the change in variable sets. That's what I was getting at. Comparing rote accident rates between eras requires considering far more factors than simply amount of automation or hand flying.

Automation got rid of the other variables.

Do you know how data analysis or testing is done? You vary one thing and all the rest remain constant.

The only variable measured is flying without automation and flying with it. This is why it is always stated, and as a fact that automation made flying safer.

I am all ears as to how you are going to make considerations and measure more than one variable.

Absolutely aware - data analysis is a major part of my work. What you are doing is varying the scenario, then comparing results, and that’s simply not logical. The accident rates due to pilot error in the 1960s encompass the complete picture of failure modes from that time, and the same is true of the modern era where safety is dramatically improved.

Your argument is also smeared by using emotionally-loaded descriptors like ‘sky jockeys’ - usually such language indicates some degree of animus and reduces credibility. Would you use such phrasing in a report?

I've stopped debating the poster in question, I think we're gonna agree to disagree. But for anyone viewing this thread on the sidelines, I'll once again point out that most, if not all, the major airlines I know of and have friends at encourages hand flying. It's in their official pubs!!! Now that's not to say they say to brush up on your proficiency on a difficult arrival with super strong crosswinds, but most pilots I know will hand fly departures and approaches when it's not too busy.

I actually agree with this poster in the overall premise—automation is great. It saves lives, improves efficiency, is and should be encouraged, and will only increase (until us pilots are merely monitors or completely obsolete).

But I doubt the poster is a pilot. I doubt this poster knows what it's like to fly the line day after day and encounter the plane and all the great automation failing daily. In small ways, but it's there. It can often be misprogramming of the FMS or even an anomaly (known, predictable deficiencies in the automation) that requires pilot input. Sometimes, the automation can kick off at the worst time, the startle factor is real. When that happens, you need a sharp pilot... The automation is completely out to lunch and completely worthless.

If all airline pilots were private pilots with 100 hours and trained to hit buttons only, I think our favorite poster would see how much manual flying is required to keep the airliners of today in the sky. I think this poster is thinking our tech is a lot further along than it actually is... In the mid future, I think he/she will be mostly correct. We ain't there yet.

Ah, I see the Gremlin is back to revive a months’ old topic.

In fairness, the Gremlin is correct that automation is very effective, and further advances will only improve flight safety.

Aside from that, the Gremlin only knows what his friend that is a station manager tells him, or what they read in accident reports. Gremlin has zero first hand knowledge of actual line operations, nor do they have knowledge of real world FOQA/ASAP/flight safety data.

Let’s not feed the trolls.

NW747-400 wrote:

Ah, I see the Gremlin is back to revive a months’ old topic.

In fairness, the Gremlin is correct that automation is very effective, and further advances will only improve flight safety.

Aside from that, the Gremlin only knows what his friend that is a station manager tells him, or what they read in accident reports. Gremlin has zero first hand knowledge of actual line operations, nor do they have knowledge of real world FOQA/ASAP/flight safety data.

Let’s not feed the trolls.

Yes, that was the point I was trying to make - that automation is certainly an effective tool, but even its use requires careful evaluation of human factors. When someone or something fouls up, what's important is understanding why. I'm not that interested in a discussion that inevitably leads to the conclusion that pilots are unnecessary because on any given day they may display incompetence if automation isn't there to save them. The reverse can just as easily be true.

Aaron747 wrote:

NW747-400 wrote:

Ah, I see the Gremlin is back to revive a months’ old topic.

In fairness, the Gremlin is correct that automation is very effective, and further advances will only improve flight safety.

Aside from that, the Gremlin only knows what his friend that is a station manager tells him, or what they read in accident reports. Gremlin has zero first hand knowledge of actual line operations, nor do they have knowledge of real world FOQA/ASAP/flight safety data.

Let’s not feed the trolls.

Yes, that was the point I was trying to make - that automation is certainly an effective tool, but even its use requires careful evaluation of human factors. When someone or something fouls up, what's important is understanding why. I'm not that interested in a discussion that inevitably leads to the conclusion that pilots are unnecessary because on any given day they may display incompetence if automation isn't there to save them. The reverse can just as easily be true.

I think the problem with automation and handflying (which to be honest is also already supported by a lot of automation in one form or another) is the communication between the human and the machine. While we do not have an engineer on board anymore, the engineer is replaced by the computer. So a lot of the training and focus during development and enhancement of the automation has to go into how the machine tells the human what is going on and especially what is going wrong and the human has to be able to understand that and act accordingly.

On the EK incident we have a chain of errors and missunderstandings: Human sets up computer wrong --> Computer strictly follows program not knowing that in the present situation that program can cause an accident --> Computer is also not telling the human what is going on --> human does not know what is going on nor what to change to get out of the situation --> computer flies aircraft dangerously low

This chain has to be improved so that the human machine symbiosis can become safer.

GCAA initial report is now available. The GCAA describes the incident as ‘ serious incident’

https://www.gcaa.gov.ae/en/ePublication ... Report.pdf

xwb777 wrote:

GCAA initial report is now available. The GCAA describes the incident as ‘ serious incident’

https://www.gcaa.gov.ae/en/ePublication ... Report.pdf

Boeing knows what happened, but not why. Somehow the pitch mode got inadvertently changed to Altitude Hold (ALT) instead of TO/GA prior to takeoff. Instead of being parked at 8 degrees as normal, the Flight Director was at 0 degrees pitch. The crew missed this and missed the incorrect mode shown on the PFD. After rotation, the Flight Director started giving guidance back down to runway altitude. The Captain started to follow it.

To my understanding the GPWS “DON’T SINK” alert occurred cluing the Captain that something was wrong. She pushed TO/GA and the Flight Director resumed proper climb guidance.

This scenario has happened a few times previously and those pilots recognized the erroneous Flight Director guidance and didn’t follow it, and continued to climb.