AC1804 Montego Bay Report Released

[Eric Janson]

Link to the report.

http://www.bst-tsb.gc.ca/eng/rapports-r ... 4f0065.pdf

Pretty clear unstabilised approach imho.

Very close to another Asiana 214 accident.

[pelmet]

Admittedly, I only read a general outline but I believe that the unstable approach is a red herring. The real cause of the hard landing was what happened after the autothrust was disconnected. No significant manual thrust inputs allowing the speed to decay while at the same time, the airspeed was not monitored resulting in a slow speed in the flare at idle thrust.

You have to monitor your airspeed on approach.

[Eric Janson]

Speed control is an important component of a stabilised approach.

As is engine thrust above idle.

As is the timely completion of checklists.

[pelmet]

Speed was the critical item and you are correct, an important part of a stabilized approach. I should have made it more clear that the earlier gyrations which led to the autothrust disconnect were not the main cause. If he had use the thrust levers once in manual, I suspect that we would have no report to read.

[crazyaviator]

Yep, let the automatics do the job UNTIL, heaven forbid, we MUST do a VOR-DME approach to save our life !!!!! remnants of asiana in SFO ,,,,,Its getting so stupid, you don't even need to be a pilot to understand this !!

[Rookie50]

Link to the report.

http://www.bst-tsb.gc.ca/eng/rapports-r ... 4f0065.pdf

Pretty clear unstabilised approach imho.

Very close to another Asiana 214 accident.

Absolutely, in reality, if not in people's perception, or AC 's PR dept.

Accidents are hard landings. Still no report on that one.

Near accident hard landings, are -- nothing.

Someone explain the substantive difference between this one and Asiana's, other than 125 feet of runway. I read on PPrune that at their aircraft attitude at TD, there is no way they could have seen the runway in front of them, therefore had no clear idea if they would hit it, or not.

125 feet? That's an extremely lucky outcome.

[justwork]

That was ugly to read. Way unstable. 3+G's? Yikes. As soon as the thrust went to climb power on the approach they should have just gone with it and tried again. That's my armchair quarterbacking for the day. It is however important to read reports like these, if you find yourself in an ugly approach situation like this remember the consequences and get the heck. Then try again.

[goingnowherefast]

It is however important to read reports like these, if you find yourself in an ugly approach situation like this remember the consequences and get the heck. Then try again.

Bingo! Learn from other's experiences, good and bad.

PM (or PNF) also has an important job to call deviations and/or call "unstable". Do not be afraid to voice your opinion if things aren't going to plan. Chances are the other guy is thinking the same, just needs the verbal hint to act.

[brooks]

AC might want to consider changing its stabilized approach criteria height to 1000ft AAE. It seems to work for WJ but the US controllers don't really like it.

[ahramin]

Having a 1000' hard stabilized approach criteria has been demonstrated to be completely useless. After hammering the crews about it with SOPs, verbal callouts, memos, and everything else proponents of stabilized approach criteria could come up with it still gets ignored 97% of the time. Complete failure and yet we still hear it being preached as if it's going to save us.

Imagine if V1 cuts had been dealt with the same way. We'd be crashing airliners every time an engine failed at low altitude. Instead we wrote an SOP for it and trained it over and over until we could get it right easily. It's obvious to me that landing accidents need to be dealt with the same way. Problem with crews not realizing they are in manual thrust? Regularly fail the autothrust in the sim on final without warning. Problem with pilots not going around when the landing goes squirrely? Unexpectedly give them a smooth transition to a 30 kt tailwind or crosswind in the last 100' and watch what happens. Not realistic failures and situations I know, but realistic training scenarios to teach pilot skills that are obviously lacking.

How the world pilot leadership has been able to keep their heads buried in the sand on this issue for so long I'm sure I don't know, but it's still the state of aviation today to looks at the complete failure of stabilized approach criteria and tear our hair trying to move the 3% figure for go arounds 1 percentage point further up to a whopping 4% as if that's the way of the future.

Sorry about the rant, it's just so sad. This crew was obviously way over their heads in a situation they had not been properly trained for. We give lip service to automation dependency and automation confusion but don't actually deal with it. I believe the fault for this accident lies directly on the Chief Pilot.

[Rookie50]

I believe the fault for this accident lies directly on the Chief Pilot.

Accident? What accident? It sure isn't an accident to AC. The YHZ near disaster isn't even an "accident" to them by their comments.

Fault with chief pilot? How about upper management, along with everyone in Canada it seems who sees AC as so much safer than a "foreign" airline like Asiana.

Guess what, they aren't. Period.

This was inches from another Asiana. And look at the difference in commentary. Ho hum, nothing to see, move along.

I mean, until one can call "accidents", accidents, not hard landings or "anomolies" call a spade a spade, seems pointless to go further.

Heads in the sand.

Search AC's press releases. You will not find one release with the word "accident" in it.

[Cat Driver]

It gets worse if you read further.

The TSB also get on board with trying to misrepresent what it really was.....An accident causing major damage to the airplane and injuring passengers.

On Monday, the Transportation Safety Board of Canada issued a release describing Sunday's Air Canada Airbus A320 crash as a "collision with terrain" rather than as a "hard landing," the term used by officials with Halifax Stanfield International Airport and Air Canada.

The TSB is just another government agency that has no problem with lying to the public.

Before the airplane crashed onto terrain it hit antennas over a thousand feet before the runway.

[Old fella]

I believe the fault for this accident lies directly on the Chief Pilot.

Accident? What accident? It sure isn't an accident to AC. The YHZ near disaster isn't even an "accident" to them by their comments.

Fault with chief pilot? How about upper management, along with everyone in Canada it seems who sees AC as so much safer than a "foreign" airline like Asiana.

Guess what, they aren't. Period.

This was inches from another Asiana. And look at the difference in commentary. Ho hum, nothing to see, move along.

I mean, until one can call "accidents", accidents, not hard landings or "anomolies" call a spade a spade, seems pointless to go further.

Heads in the sand.

Search AC's press releases. You will not find one release with the word "accident" in it.

"The Air Canada Rouge LP, Airbus A319 (registration C-FZUG, serial number 697), operating as flight AC1804, departed Toronto Lester B. Pearson International Airport, Toronto, Ontario, under instrument flight rules for Montego Bay, Jamaica, with 131 passengers and
6 crew members on board. The flight crew was cleared for a non-precision approach to Runway 07 in visual meteorological conditions. The approach became unstable and, at 1429 Eastern Daylight Time, the aircraft touched down hard, exceeding the design criteria of the landing gear. There was no structural damage to the aircraft, and there were no injuries."

TSB as noted above in their summary indicated the aircraft touched down hard, so to me that seems a reasonable description.

[complexintentions]

Having a 1000' hard stabilized approach criteria has been demonstrated to be completely useless. After hammering the crews about it with SOPs, verbal callouts, memos, and everything else proponents of stabilized approach criteria could come up with it still gets ignored 97% of the time. Complete failure and yet we still hear it being preached as if it's going to save us.

Imagine if V1 cuts had been dealt with the same way. We'd be crashing airliners every time an engine failed at low altitude. Instead we wrote an SOP for it and trained it over and over until we could get it right easily. It's obvious to me that landing accidents need to be dealt with the same way. Problem with crews not realizing they are in manual thrust? Regularly fail the autothrust in the sim on final without warning. Problem with pilots not going around when the landing goes squirrely? Unexpectedly give them a smooth transition to a 30 kt tailwind or crosswind in the last 100' and watch what happens. Not realistic failures and situations I know, but realistic training scenarios to teach pilot skills that are obviously lacking.

How the world pilot leadership has been able to keep their heads buried in the sand on this issue for so long I'm sure I don't know, but it's still the state of aviation today to looks at the complete failure of stabilized approach criteria and tear our hair trying to move the 3% figure for go arounds 1 percentage point further up to a whopping 4% as if that's the way of the future.

Sorry about the rant, it's just so sad. This crew was obviously way over their heads in a situation they had not been properly trained for. We give lip service to automation dependency and automation confusion but don't actually deal with it. I believe the fault for this accident lies directly on the Chief Pilot.

Sorry, but maybe I don't understand your comments regarding the "complete failure of stable approach criteria" being "demonstrated to be completely useless". Come again?

It's been policy at several of my employers, heavily enforced, and reduced unstable approaches to nearly zero. If an unstable approach is continued to a landing, it's detected by QAR and results in an automatic trip to the office to explain. Depending on how extreme the deviation from the criteria, the actions taken could be anything from a simple safety briefing, to remedial training, to disciplinary action ranging from demotion to dismissal!

On the other hand, missed approaches are completely non-punitive. In fact they're encouraged if a safe landing is in ANY doubt. If they're initiated above 2,000AGL, they're not even reportable. As a result of this kind of discipline, I can assure you most vehemently that stable approach criteria is NOT "ignored 97% of the time" anywhere I've ever been. So I don't quite understand your apparent strong opposition to stable approach criteria. I was a bit taken aback by this statement, but as I said perhaps I misunderstood?

And I have no idea how you relate V1 cut training, which is done to remove the need to make a decision and instead trains procedural memory (aka muscle memory), versus flying an approach where there is absolutely the time to assess and make decisions using conscious thought, evaluate the options, and then elect to take a course of action. I don't really see this as a valid analogy at all.

Of course as with any policy, if it's completely ignored, there is no point. It would seem in this flight the crew pretty much disregarded every approach SOP already in place, so I agree in that regard - no reason to change rules, if crews don't follow them at all. That's what I find worrisome about this particular incident, is how grossly out of the envelope it was. Not just a little high or a little fast, correcting - totally f-ed up and still pushing on. That's the "why?" that needs to be addressed.

As far as not being trained for - come on. Not trained for a VOR approach?? Mode confusion? Working at Air Canada? I mean at some point, a pilot has to take some responsibility for understanding how his machine operates and remember how to fly an approach other than a coupled ILS.

I can blame management for a lot of things, but I refuse to blame them for my own poor decisions or not knowing how to fly my aircraft.

[ahramin]

Having a 1000' hard stabilized approach criteria has been demonstrated to be completely useless. After hammering the crews about it with SOPs, verbal callouts, memos, and everything else proponents of stabilized approach criteria could come up with it still gets ignored 97% of the time. Complete failure and yet we still hear it being preached as if it's going to save us.

Imagine if V1 cuts had been dealt with the same way. We'd be crashing airliners every time an engine failed at low altitude. Instead we wrote an SOP for it and trained it over and over until we could get it right easily. It's obvious to me that landing accidents need to be dealt with the same way. Problem with crews not realizing they are in manual thrust? Regularly fail the autothrust in the sim on final without warning. Problem with pilots not going around when the landing goes squirrely? Unexpectedly give them a smooth transition to a 30 kt tailwind or crosswind in the last 100' and watch what happens. Not realistic failures and situations I know, but realistic training scenarios to teach pilot skills that are obviously lacking.

How the world pilot leadership has been able to keep their heads buried in the sand on this issue for so long I'm sure I don't know, but it's still the state of aviation today to looks at the complete failure of stabilized approach criteria and tear our hair trying to move the 3% figure for go arounds 1 percentage point further up to a whopping 4% as if that's the way of the future.

Sorry about the rant, it's just so sad. This crew was obviously way over their heads in a situation they had not been properly trained for. We give lip service to automation dependency and automation confusion but don't actually deal with it. I believe the fault for this accident lies directly on the Chief Pilot.

Sorry, but maybe I don't understand your comments regarding the "complete failure of stable approach criteria" being "demonstrated to be completely useless". Come again?

In 2013 at the Go-Around Safety Forum (supported by SKYbrary)[4][5], more recent studies and analyses were cited by presenters. These studies indicate that:
Between 3 and 4% of all approaches are reported/recorded as unstablised
Yet, only 3% of these result in a go-around being flown
In other words, 97% of unstabilised approaches continue to be flown to a landing contrary to airline Standard Operating Procedures (SOPs)

To reframe this statement a little, I see two possibilities here:

1. The professional pilots flying these aircraft are astoundingly unsafe pressing on with these dangerous unstable approaches 97% of the time and the high success rate is just luck over and over again, or
2. The professional pilots flying these aircraft know they haven't met the stabilized approach criteria, and their experience tells them that the stabilized approach criteria is not relevant in this case and they can safely land anyway.

Lets stipulate that stabilized approach criteria is actually a good measure and that the successful landings are just luck. There are approximately 100 000 scheduled airline landings a day. If 3% of those are unstable that's 3000 landings a day that do not meet stabilized approach criteria. If stabilized approach criteria is wrong 99.9% of the time but saves an accident 0.1% of the time then those unstable approached which are continued should lead to 3 accidents a day. But nowhere close to 0.1% of unstable approaches lead to an incident or accident. The best data I could find on short notice (Skybrary again, not very good) showed 110 incidents or accidents in the last 20 years in which unstabilized approaches were a factor. If correct (not robust data I know, but it illustrates the point), that means 99.9995% of unstabilized approaches are continued to a safe landing. So when I say that stabilized approach criteria has failed, that's fact as 97% of the time it gets ignored. I speculate that it's failed because it's garbage, but even if I'm wrong and it's a very good way of preventing accidents, the fact is it hasn't worked. Instead of seeing that fact, the safety industry keeps trying to make very marginal gains in how many unstabilized approaches are discontinued. Why don't they change their minds? Perhaps because of how much effort they've already invested and how embarrassing they perceive it would be to say "we were wrong, this didn't work".

It's been policy at several of my employers, heavily enforced, and reduced unstable approaches to nearly zero. If an unstable approach is continued to a landing, it's detected by QAR and results in an automatic trip to the office to explain. Depending on how extreme the deviation from the criteria, the actions taken could be anything from a simple safety briefing, to remedial training, to disciplinary action ranging from demotion to dismissal!

On the other hand, missed approaches are completely non-punitive. In fact they're encouraged if a safe landing is in ANY doubt. If they're initiated above 2,000AGL, they're not even reportable. As a result of this kind of discipline, I can assure you most vehemently that stable approach criteria is NOT "ignored 97% of the time" anywhere I've ever been. So I don't quite understand your apparent strong opposition to stable approach criteria. I was a bit taken aback by this statement, but as I said perhaps I misunderstood?

That's an interesting story but I can't really quantify it. You are working in an interesting part of the world where it is possible to fire a pilot for making a safe landing after an unstabilized approach. That's not going to be the case in North America for the foreseeable future. If I may ask, how many landings did you do in 2016 and how many of them were preceded by go-arounds for unstable approaches? Are you certain this punitive and non-punitive policy applies to everyone? What about the chief pilot's brother? In any case, it's very hard to know what to make of your statements as I'm sure your airline does not publish it's actual unstabilized approach rates to any safety database so we have no way of knowing what the actual numbers are. I believe Emirates has the exact policy you mentioned above and didn't they have a rather spectacular landing accident last year? I'm open to admitting that I'm wrong and that stabilized approach criteria can be made to work but I'd have to see the data for the fleet, and the data for that accident. Looking at airlines that do publish their data, it's a completely different story to what you say is happening at your airline.

And I have no idea how you relate V1 cut training, which is done to remove the need to make a decision and instead trains procedural memory (aka muscle memory), versus flying an approach where there is absolutely the time to assess and make decisions using conscious thought, evaluate the options, and then elect to take a course of action. I don't really see this as a valid analogy at all.

http://www.tsb.gc.ca/eng/rapports-repor ... 5h0002.pdf
http://www.skybrary.aero/index.php/A332 ... esia,_2013

I noted previously that possibly 99.9995% of unstabilized approaches are continued to a safe landing. Above are a couple examples of the type of unstabilized approach that more frequently results in an accident. You'll notice that where there was time to assess and make decisions using conscious thought, all stabilized approach criteria were met. When the approach later became unstabilized there was no time to think and the crew did their best to land the aircraft under very difficult conditions. I suspect that if we train crews to react instinctively to things going sideways close to the ground by performing a go-around, we will have a much better chance of reducing landing accidents. That's where the comparison to V1 cuts comes from. A go-around at low altitude or on the ground when things become unstable should not require conscious though, an evaluation of the options, and then a course of action. The course of action should be procedural memory for a go around after as little evaluation as possible to determine that this landing is not going as well as they normally do. This is brutally hard to do as the normal muscle memory is to pull off a bit of fancy footwork and do a landing despite the difficult conditions.

Of course as with any policy, if it's completely ignored, there is no point. It would seem in this flight the crew pretty much disregarded every approach SOP already in place, so I agree in that regard - no reason to change rules, if crews don't follow them at all. That's what I find worrisome about this particular incident, is how grossly out of the envelope it was. Not just a little high or a little fast, correcting - totally f-ed up and still pushing on. That's the "why?" that needs to be addressed.

As far as not being trained for - come on. Not trained for a VOR approach?? Mode confusion? Working at Air Canada? I mean at some point, a pilot has to take some responsibility for understanding how his machine operates and remember how to fly an approach other than a coupled ILS.

I can blame management for a lot of things, but I refuse to blame them for my own poor decisions or not knowing how to fly my aircraft.

I agree with your sentiment ... actually I should say I agree with your sentiments (all of them, I like go arounds and do them at every opportunity i.e. unstabilized approaches, 2 last year), and I don't think anyone who has this much trouble flying their airplane has any business being a professional pilot. But that sentiment is not going to prevent the next accident. The job description for our chief pilots is full of stuff about ensure pilot compliance with SOPs, pilot compliance with company standards, pilot compliance with CARs. And if that is what the purpose of a chief pilot is then yes, you can wash your hands of this crew and say they messed up, it's their fault.

I always saw my primary job as chief pilot as making sure we never had an accident. And it's very hard to say it's my job to make sure we don't have an accident and then blame the crew for the accident. You can say they should have known better, or you can make sure they know better.

[JigglyBus]

Ahramin, I don't agree with your math. Well I do and I don't. I think you might be missing the point.

You stated that "99.9995% of unstabilized approaches are continued to a safe landing"... now I realize you were just ballparking that, but it's misleading.

Sure, a verrrry small percentage of unstable approaches end in an accident, but the key is in the comparison. 99.99999999991% of stable approaches end in an accident.

So, when you compare 0.0005% to 0.000000001% it's terrifically dangerous.

To simplify......

Often it's said that 3% of approaches are unstable.

It's also stated an unstable approach is the causal factor in 60% of landing accidents.

If unstability makes no difference, it should be ~3%.

3% of flights cause 60% of the accidents? That means you are 20 times, or 2000% more likely to crash during landing, if you are unstable.

I realize numbers can be twisted to make whatever point you want, but that one seems pretty strong.

[ahramin]

It's also stated an unstable approach is the causal factor in 60% of landing accidents.

A causal factor, not the causal factor. Big difference there but I won't argue that unstable approaches have a higher chance of ending up in an accident. You are correct that my math with incomplete data is not compelling. I can try to convince you that there are much better ways of reducing landing accidents than a top down, arbitrary procedure to follow but I can't prove it. I read an article in FSF's magazine a couple years ago on unstabilized approaches leading to accidents which had an interesting figure tucked away amongst all the other data. Wish I had kept it as I can't find it now but they had a chart showing accidents over a given period and one column was whether or not the approach met stabilized approach criteria at 1000'. By my rough count about half the accidents flights listed in that article met the criteria at that gate.

However the point I'm making is not that unstable approaches are safe, but that stable approach criteria as designed by FSF and implemented by various airlines contributes little if anything to safety. 97% of the time it gets ignored. Again I suspect that it gets ignored because it isn't very useful and experienced pilots see that and ignore it, but even if I'm wrong I don't see how an SOP that gets ignored 97% of the time can be called anything other than a complete failure. I certainly haven't seen anyone presenting any data that suggests stabilized approach criteria has made a difference in the landing accident rate.

I believe a better approach to dealing with approach and landing accidents is to use simulators to put pilots in situations where the landing is not going well and get them used to doing a go-around when that happens.

[complexintentions]

First of all, I'm not at Emirates. The accident in DXB is a discussion I'm more than willing to have, I do have pretty thorough intel on what happened. But it has absolutely nothing to do with stabilization approach criteria. Mode confusion or lack of system knowledge, perhaps. Fatigue, definitely. But that accident wasn't caused by continuing an unstable approach to landing, so is completely irrelevant to your argument. If you're going to go around, of course you have to it correctly. My argument for stable approaches kinda assumes that, ok?

But the airline I'm working for now - also not in North America - has pretty much identical policies regarding stable approaches. And they, like Emirates, publish detailed QAR statistics every single month showing, literally every parameter measured (too high, too fast, too low, EVERYTHING) broken down by fleet, phase of flight, the works. So yeah, it's pretty easy to see exactly what's going on. And unstable approaches continued to landing are virtually non-existent. The stats are not publicly published, that is correct. They are absolute available to all line pilots, they are sent as an email attachment every month.

I didn't say someone would get instantly fired for landing off an unstable approach. That would be an extreme case, like Montego Bay. The guys in the office who analyze the QAR data are pilots, and are mostly curious to understand the "why" when an approach was continued. Generally it would be a simple safety briefing. As I said, you have to have a non-punitive culture.

You can claim that 99.9999% of unstable approaches continue to a safe landing. But that's a backwards way of approaching safety - the accidents that HAVE happened, like AC or Asiana, would have absolutely been prevented if at any of the MULTIPLE gates defined in typical stable approach criteria (1,000 feet, 500 feet, speed, configuration, thrust setting, descent rate!!) they had just pushed the TOGA button. Of course I can't prove a negative. So who knows how many landing incidents/accidents have been prevented by crews going around when they couldn't meet the criteria? Isn't that the job though? Risk management? I'm sorry, but 110 accidents/incidents from unstable approaches somehow supports your statement that stable approaches are a "complete failure"...?

I have to disagree once again with your analogy to the V1 cut. This is an near-instantaneous event, and requires a very quick, reflexive reaction. Most unstable approaches develop over a much longer period of time comparatively, triggered by anything from poor planning (on either ATC or the pilots' part), improper energy management, misjudgment of winds and distance to run, a late change in landing sequence, and so on. There are multiple points over quite a long period of time to make a decision to try again. Sure, in the rarest cases it could go sideways very late and close to the ground, a sudden wind gust perhaps. But I categorically disagree that it's "brutally hard" to press TOGA and say "Go Around, Flaps 20".

However the point I'm making is not that unstable approaches are safe, but that stable approach criteria as designed by FSF and implemented by various airlines contributes little if anything to safety. 97% of the time it gets ignored. Again I suspect that it gets ignored because it isn't very useful and experienced pilots see that and ignore it, but even if I'm wrong I don't see how an SOP that gets ignored 97% of the time can be called anything other than a complete failure.

Wrong. Or at best, misleading. 97% of the time an unstabilized approach is continued to landing, and the criteria on unstable approaches is ignored (which is probably why they became unstable, hmm?!!). But the vast majority (according to your stats, 96-97%) of APPROACHES are flown to stabilized criteria. The criteria is NOT being ignored in the vast majority of approaches. The bottom line is: the fact that the majority of unstable approaches continued to landing didn't result in accidents, is not an argument against flying a stable approach, which is all the criteria is designed to mandate!

A policy or SOP is only useful if it's followed. Your statement ("I suspect it gets ignored because it is isn't very useful and experienced pilots" etc) smacks of the sort of ego that gets people into trouble, thinking they know better than their SOP's. Those experienced AC pilots probably wish they had followed theirs. It's fine to say improve sim training for events close to the ground on landing. But the seeds for this accident, and most landing accidents, were sown long before they got anywhere close to the ground. In this case it goes right back to not even knowing what approaches were available. Yes, we all know we can "save it" even with high descent rates and fast and so on. But the job isn't to see how awesome our skills are - it's to use those skills to keep the machine within conservative parameters.

I'm really not certain what you're arguing for. Scrapping stabilized approach criteria? Or is your entire post just an elaborate way of saying you consider a policy a "failure" because people routinely disregard it when it's needed most?

[ahramin]

What I'm arguing for is scrapping the idea that stabilized approach criteria SOPs are the solution to these sorts of accidents.

I'm all for stabilized approaches as a goal and stabilized approach criteria SOPs as a target to help that goal. Actually I don't care if it's in the SOP or not, it's still best practice, and if they removed it I'd still fly the same way and continue to do a go-around whenever I'm unstable.

But if a chief pilot writes in stabilized approach criteria into their SOP and thinks they've greatly reduced the chances of a landing accident caused by an unstable approach, they're dreaming. The incident above is an example of what we will continue to see until we realize this and start coming up with more effective ways of reducing them.

[mbav8r]

ahramin,
You are contradicting yourself with each post, it's impossible to know to what extent stabilized approach criteria has prevented possible incidents or accidents, clearly this one could have been prevented.
Without setting the goal you're leaving it up to individual pilots to set their own goals which may not be the same for both pilots on a given flight, what then. How about a policy? Also, if you as a pilot choose to ignore a set company limitation, in my company it's called, "intentional noncompliance" and even an SMS cannot save you from this.
SOPs are there so you know what to expect from the guy you're working with even if you've never met, also in an effort to reduce risks and prevent accidents, they are clearly not working because pilots who choose to ignore SOPs are not all crashing airplanes, maybe we should just get rid of SOPs