Ben Sandilands — Missing You!

Every morning UK time, I would switch on my computer to see as often as not an informative and thought-provoking post on the Australian Plane Talking Crikey website by aviation journalist Ben Sandilands. Sadly, he is no longer with us.

Whether it was the mystery of Malaysian Airlines MH370, space exploration, astronomy, or the increasingly cramped conditions for economy passengers even on long-haul flights, he had strong views, and more importantly was extremely knowledgeable regarding the highly technical aspects.

Some time ago I discovered we were both “soixante-huitards” (68ers), having been in Paris in May 1968 at the time of the student riots, he as a journalist.

In finalizing the second edition of Air Crashes and Miracle Landings  I asked for his advice only six weeks ago, and though very ill in hospital he gave unstinting help. He did mention he would continue his blog but would not keep at it when no longer able to do it justice. His last blogs were spot on, and his silence while with us only lasted a week or so.

He could be very outspoken and a breath of fresh air, outraging some. It is a pity that we cannot ask him whether, with hindsight, he thought his criticism of Qantas management was too extreme now the company seems to be doing so well financially.

We had similar views on many matters, and I will greatly miss him.

We have lost two great Australian aviation writers; the other being Macarthur Job just a few years ago.

Christopher Bartlett

Expanded Second Edition out shortly

We have been working on the second edition of Air Crashes and Miracle Landings and hope to have it “perfect” for publication in early January 2018.

To cover as many incidents as possible, some are described in detail while others are only allotted a couple of pages or so. Also, to include some we missed in the first edition and more recent ones we have added pages.

Other accounts have been partly rewritten with new facts, or in one or two cases to be harder-hitting at the risk of infuriating some.

As before, each chapter covers a specific type of incident in chronological order. While showing the evolution of accidents over time, this means those of the greatest interest may well be nested within the chapter rather than found at the beginning.

The glossary has been placed in the second part of our book Plane Clever: Booking Strategies and All about Flying.

Perth to London non-stop medical implications

A couple of weeks ago there was an article in the Daily Mail about a doctor complaining about the poor quality of the on-board medical kit at her disposal when asked to treat an sick passenger.

Incidentally, she recommended the flight divert, but the captain in consultation with the airline’s (KLM?) medical service declined. In fact, the passenger in question recovered and   proved well enough to continue the flight after the next stop and there was no need for a costly diversion!

An interesting article in Condé Nast Traveler tells what a great job some talented doctors have done on flights, including one who asked the captain to increase the cabin pressure to the equivalent of 2,000 feet for the benefit of a passenger who, suffering from a detached retina had had gas injected into his eyeball by an ophthalmologist to press it down and stick back on over time. The low pressure in the cabin was causing the artery in the back of his eye to be squeezed and making him blind. The captain had enough fuel to do that and the man was soon able to see from that eye again.

The Condé Nash article said that in the US there is one medical emergency per 604 flights, or 16 medical emergencies per one million passengers. As many of those flights would be short-haul and the aircraft about to arrive at its destination anyway, the average for really long-haul flights would surely be much higher.

The famous Singapore Airlines ultra-long-haul flight to New York about to be reinstated probably had mostly healthy working-age business passengers. However, Perth to London might have more elderly passengers, and one wonders whether there should be some form of triage to exclude passengers likely to have medical problems. Or require those of a certain age to have a medical certificate.

This poses the problem of accusations of discrimination, though most sensible passengers, aware of their condition, would avoid such flights themselves if at  risk of suffering an in-flight medical emergency. One solution would be to insist they have insurance to compensate the airline, leaving the triage to the insurance companies.

Qatar Airways and Pratt and Whitney–Instant Gratification or Future?

When the Boeing 747 first appeared on the scene, the new extra-large diameter engines would flex and often have to be replaced, sometimes by a spare engine carried under the wing in an extra pod.  All that is forgotten now, but reminds one of the problems the Pratt and Whitney is having with engine for the A321 NEO–publicized and levered by Qatar Airways.

According to Airbus the engine is meeting its fuel burn and other performance targets.  There are still problems, particularly where the engines are operated via the hot and gritty middle-east hubs. Also, Pratt is not able at the moment to produce the engines fast enough to meet promised deliveries.

The problem for Qatar Airways is that if they do play it safe by switching to another supplier with compensation for an immediate cash benefit in terms of compensation and advantageous pricing for favoring a new supplier, they may find in a few years time that they have a less efficient engine and lose out in the long run.

While manufacturers of airliners and engines may think twice in the future about using Qatar Airways as a lead customer for a new model, there may be an upside. Meeting the difficult conditions at the mid-east hubs, and satisfying Qatar Airways-like perfectionist demands for anything from cabin furnishings to engines may make makers improve their products early on and so all benefit long-term.

Colombia Crash–Too little fuel puts others at risk too

Leaving aside the scandal of the captain of the British Aerospace 146 plane PLANNING to arrive at Medellin with virtually empty fuel tanks, it raises the interesting question of the danger of several aircraft declaring emergencies at the same time.

Fearful of the repercussions and the possibility of losing his licence (because the fuel emergency was his own fault), the captain doubtless did not declare one in timely fashion with the result that another aircraft with a perhaps a less pressing problem was sequenced ahead of him.

Though a flight plan with the maximum range of the aircraft being equal to the distance to be covered would never be submitted in normal circumstances, there are situations where pilots need at their discretion to load extra fuel above the legal requirement where they suspect weather conditions at their destinations are likely to be such that many aircraft will be stacked or diverted.

A few years back a number of Ryanair flights to Spain had to divert with limited fuel reserves in somewhat exceptional circumstance, with the company later giving pilots more leeway. The situation is somewhat like vaccination; if only a few parents refuse to have their children vaccinated they can “selfishly” benefit. Were too many to do so, it could well be disastrous for all, including themselves.

The captain of the Colombia aircraft’s failure to declare a fuel emergency early on meant another that had declared an emergency was sequenced ahead of them. In fact when the controller asked the Colombia captain how long he could hold he never answered.

If too many airlines were to save money by keeping reserves to a minimum even in adverse weather conditions–say just a higher than usual headwind–it could spell disaster. Having to declare a fuel emergency could put another aircraft at risk.

Colombia Crash–Reminiscent of Avianca 52 in New York?

Air traffic controllers can only work with the information they are given, and that said, their primary concern is to avoid aircraft colliding although TCAS  now should make that unlikely.

When Avianca Flight 52 crashed in a suburb of New York 1990 after failing to see the runway on its last-chance landing at JFK with only dregs in its tanks, the FAA maintained it was the failure of the copilot to use the word EMERGENCY instead of “We are running out of fuel.” (On that murky day many aircraft were on hold running out of fuel.)

It seems the pilot of the Avro in the Colombia crash did use the word “Emergency,” but perhaps only at the very last minute when it was too late. This might be why the controller had given precedence to another aircraft that had requested priority because of its own problems.

In both the Avianca and Colombia cases, the pilot “would have got away with cutting things so fine” had there not been last-minute hiccoughs making expected live-saving landing impossible.

Not wanting to instigate an official inquiry, the pilot might have not wanted to officially declare an emergency earlier, and have used vaguer words. Early declaration of an emergency would have given the controller a better chance of sequencing the two aircraft more appropriately.

Perhaps the electrical problems mentioned might have meant the pilot was not fully aware of just how little fuel he had left. That said he must have known he was operating the aircraft with a full load of passengers at the limit of its range.

At this stage, this is only speculation raising questions that need to be asked.

Artificial Intelligence could have prevented AF447 PF having to take full and disastrous manual control.

In the process of updating my “Classic” book Air Crashes and Miracle Landings (which I hope to complete before December) I have just rewritten the beginning of the piece on AF447 as follows only to find a very pertinent article in The Economist.

AF447 IF ONLY JUNIOR FIRST OFFICER HAD DONE NOTHING
(S. Atlantic 2009)

An unnecessary disaster with many elements and lessons

The first thing to do in a crisis “IS NOTHING.”

That is what instructors might say to airline pilots, adding, “IT GIVES YOU TIME.”

Had the junior first officer taking over Air France AF447 done just that when the autopilot gave him manual control high over the South Atlantic, everything would have been fine.

The Airbus A330 was flying perfectly at its cruise altitude of 35,000 ft and would have continued to do so had he left well alone.  Instead he pulled back sharply on his sidestick causing the aircraft to immediately climb 2,924 ft and stall in the thin air. With the aircraft mushing downwards—not quite like a brick as it had some forwards airspeed—he continued to pull back on his stick thus precluding any chance of recovery.

 

[Air France Flight 447]


The fascinating article in The Economist of September 17 entitled Flight Response covering such a situation, essentially said that if aircraft computers using artificial intelligence could continually learn from pilots’ actions what to do in difficult situations there would be no need for them to hand over total manual control to nonplussed pilots as in the case of AF447:

Haitham Baomar and his colleague Peter Bentley, artificial-intelligence (AI) experts at University College London (UCL) are developing just such an autopilot. one that uses a “machine learning” system to cope when the going gets tough, rather than ceding control to the crew.

 

According to Mr Baomar, present-day autopilots run on “hard coded” programs in which a limited number of situations activate well-defined, pre-written coping strategies—to maintain a certain speed or altitude. “They do not handle novelty well.”

 

The UCL team has written what it calls an Intelligent Autopilot System that uses ten separate Artificial Neural Networks (ANNs). Each is tasked with learning the best settings for different controls (the throttle, ailerons, elevators and so on) in a variety of different conditions. Hundreds of ANNs would probably be needed to cope with a real aircraft, says Dr Bentley. But ten is enough to check whether the idea is fundamentally a sound one.

 

To train the autopilot, its ten ANNs observe humans using a flight simulator. As the plane is flown—taking off, cruising, landing and coping with severe weather and aircraft faults that can strike at any point—the networks teach themselves how each specific element of powered flight relates to all the others. When the system is given a simulated aircraft of its own, it will thus know how to alter the plane’s controls to keep it flying as straight and level as possible, come what may.

 

Very interestingly the “intelligent” autopilot can apply lessons learnt on one model of aircraft to others in what is called “generalization.”

 

Then in situations such as in AF447 when there is no airspeed data the computer the system can look at other sources of information such as GPS data which stressed and overwhelmed pilots might not immediately consider.

 

Though such artificial intelligence systems could undoubtedly save lives getting them approved in the near future is unlikely as certification depend on verifying traditional instructions that can be easily replicated. However, the system might be applied to drones.

 

 

 

Did MH370 captain consider replicating Air NZ Mt Erebus crash?

Australian journalist Ben Sandilands interestingly cites the initial assessment of information retrieved from the home flight simulator of the MH370 captain by the Independent Group (IG) of scientists studying the available data and claims made about the disappearance of Malaysia Airlines 777.

Sandilands’ piece gives the key points and a graphic showing  a flight path in black recovered from the captain’s home simulator taking a 777 further due west than the one calculated from actual satellite MH370 handshakes.

“The the data points can be assembled to suggest a hypothetical flight toward McMurdo Sound’s graded ice runway in Antarctica, but with fuel exhaustion bringing it down in the southern Indian Ocean well before it could reach that point.”

The fact that the aircraft flies so far west before turning south suggests an intention to evade radar coverage. However, the fact that it flies westwards might also suggest the captain was envisaging a flight destined for somewhere much further away from KL than Beijing and consequently with sufficient fuel to take it all the way to McMurdo Sound.

If the intention were to land on a runway there, and for him (and eventually also the passengers) to escape, one cannot think of a more inhospitable or unsuitable place. So that seems somewhat unlikely.

On the other hand might not the captain been exploring a 1979 scenario described in Air Crashes and Miracle Landings:

Staff at Air New Zealand’s Flight Operations Division had a sinking feeling of culpability in their stomachs on learning their sightseeing DC-10 had flown straight into a mountain at the South Pole, for that was precisely what they had programmed it to do.

Without informing the pilots, they had made a ‘correction’ to the final waypoint the night before, moving it from the end of McMurdo Sound, where the track to it passed over water, to a point behind Mount Erebus where the track would pass right over the 12,450-ft high mountain.

 

A crash on a mountain at the South Pole would have been a dramatic finale. It would of course have necessitated a flight carrying sufficient fuel. Had changes to rosters meant Captain Zaharie Ahmad Shah had little chance of flying the longer routes?

This is only speculative, but might be worth exploring.

Changi Airport created against British advisers’ recommendation -Thames Estuary Airport?

Having had the pleasure of knowing Singapore’s late Howe Yoon Chong from even before I went to university, I did some research on his career — he always surprised me as on each of my visits to Singapore he seemed to have a new important role. What I did not know was that he was largely responsible for the creation of one of Singapore’s crown jewels, the new airport at Changi after British experts had recommended expanding the existing one at Seletar.

Similarly British experts are recommending the expansion of London’s Heathrow, dismissing outright the idea of a 24-hour airport in the Thames Estuary.

Though the recent collision between a drone and a British Airways A320 coming in to land there resulted in no great damage suggesting tiny drones are unlikely to actually bring down an airliner, there is an increasing likelihood of larger ones flying around controlled by idiots or terrorists. Shoulder launched surface to air missiles could be fired at airliners on their flight paths over London, not to mention the possibility of rounds being fired  from crude drainpipe mortars onto the airport itself.

Heathrow itself and its approach paths over habitation are impossible to defend and it is more than quite possible that an airliner coming in to land there will one day crash onto buildings and people. How much will that cost? Mortar rounds or rockets could bring operations there to a halt.

An isolated airport in the Thames Estuary with aircraft coming in over the sea would be much easier to protect provided development in the immediate area were prevented from the outset. It would then be possible to control who came in its vicinity.

Have the quantifiable construction and convenience costs led the experts to overlook impossible-to-predict costs of one or more disasters involving aircraft ploughing into London’s buildings and streets?