“At a time when accidents are extremely rare, each one becomes a one-off event, unlikely to be repeated in detail. Next time it will be some other airline… and some other failure – but it will almost certainly involve automation and will perplex us when it occurs.”
– William Langewische, Vanity Fair, October 2014
The airplane was high before starting down. Very high.
I advised the pilot they were thirty-nine miles from the airport.
After a short pause, he keyed his mic – “The computer isn’t telling us to start down yet.”
Decision-making aside, flying is, technically speaking, not that difficult. Moreover, many aircraft are capable of flying themselves. Automation is part of every new passenger-carrying aircraft. The autopilot is engaged shortly after takeoff and remains on until only a few hundred feet before touch down.
How we got here
The autopilot was meant to make flying safer. It has.
Airline accidents have become increasingly rare. Many notable disasters are rooted in an insidious threat that has been driven home through numerous safety studies – pilot error.
If the humans are to blame, the remedy seems simple enough – change the pilot’s role from steady hand at the controls, to systems manager.
The modern autopilot flies with more precision than a human can and affords the pilot some margin to monitor the flight’s progress, one-step removed from the manual flying. The pilot can sip his coffee and scan the panels in the glass cockpit, ensuring the systems are operating normally. He enters data and turns dials, but only touches the controls ‘in the unlikely event of an emergency.’
Airlines have traditionally hired based on experience, quantified by flight hours. That said, monitoring an airplane as it self-corrects on autopilot is not the same quality of experience as assessing the landing of a float plane at a high-altitude mountain lake in gusty conditions or dropping water on a forest fire. Therefore, to solidify their skill-set, pilots in the past had to cut their teeth ‘up north’ before attaining the requisite number of hours for a coveted airline job.
Today, due to a surge in pilot hiring, most newly-minted commercial pilots don’t need to spend much time flying in the backcountry before being hired by a regional carrier. So with the rise of automation, and the systems knowledge that modern airliners require, it seems reasonable to think that the old-fashioned stick and rudder skills might not have as much a place in a contemporary pilot training curriculum as they once did.
A look at the most recent high-profile aviation incidents tells a different tale.
Theory of Flight
Stall recovery is one of the foundational elements of learning to fly.
The angle between the relative airflow and the wing chord is called the angle of attack. As the angle of attack increases, lift increases, until the critical angle of attack is reached (around 15 degrees). At this point, the air flowing over the top of the wing is unable to adhere to the cambered surface and lift suddenly starts to break down — the aircraft stalls.
The airplane loses altitude and recovery becomes counter-intuitive. Pulling back on the control column no longer causes the aircraft to gain altitude – it deepens the stall. A pilot must fight his instincts and push forward on the control column to reduce the angle of attack, thereby breaking the stall.
In 2009, the crew of Air France 447 failed to take this critical action. Doing so would have prevented the hulking Airbus from descending into the Atlantic Ocean.
The more recent cases of Lion Air 610 and Ethiopian Airlines 302 are potentially even more troubling. A system meant to keep the airplane from stalling fought the pilots into the ground.
The size and location of the engines on the 737 MAX had a potentially troublesome effect the aircraft’s centre of gravity. Some believed the forward position of the massive turbines increased the risk of a stall during takeoff.
A technology solution was developed and termed MCAS – the Maneuvering Characteristics Augmentation System. It was a program designed to push the nose forward, thus, enacting the stall recovery protocol if a pilot increased the angle of attack too much.
It was supposed to work like traction control does in your car – a safety feature that you don’t think about, but is there if you need it.
In the most recent incidents, the MCAS may have taken over because the aircraft believed it was about to stall. The stories have been well told by now. The pilots were left wrestling with an airplane at low altitude that was pitching below the horizon.
Boeing had discovered this problem and, in a bulletin, had documented the fix. The pilot could have disabled automatic control of the horizontal stabilizers, but it is now apparent that some pilots didn’t know how.
What this means for pilots
Automation has made flying safer. However, it has made unusual situations even more unpredictable.
Though it’s becoming more and more unlikely that pilots will have to deal with an in-flight emergency, when one does happen, the response will most likely involve troubleshooting an automated system.
Major aircraft manufacturers have simulated the force control feedback on their aircraft for years. The software gives the pilot the sensation that she can ‘feel the airplane,’ as she moves the control column, but it’s an illusion.
The stick and rudder days, for better or worse, are behind us. The job requirements of a pilot today may be changing, but it’s important to remember that when things go wrong, the last line of defence is still the pilot’s hands and feet, and her ability to command the airplane.