Airplane Operational Issues

THE HISTORIC EVOLUTION IN SOPHISTICATED JET FLIGHT DECKS

This is a picture of an airliner cockpit, a British BAC 111. It was first put into service in 1965 and flew until the early 1980’s. How did the pilots manage all the switches, knobs and gauges? The fact is thirteen of them crashed, killing hundreds of passengers and crew, most caused by pilot error and disorientation, easy to imagine looking at the picture

BAC 111 Flight Deck provided by Dave Hurley

Truth is, flying the BAC 111 and other jets of that vintage was a constant struggle. We didn’t know any better because that was all we had, and it was better than its predecessors! There were lots of accidents which we can now trace back to pilots losing track of their situation. The instrument scans pilots had to perfect to establish airplane attitude and location were extraordinary, something completely unnatural and something to be proud of when you pulled it off. 

Fast forward to today …

Image provided by Max Kingsley-Jones

This is the flight deck of an Airbus A350, one of the most innovative airliners in service today.

The modern cockpit is beautiful. It’s well laid out and features bright colors that make it easy to find what you’re looking for. Once you learn how to use the touch controls (assuming you have an iPhone it’s much the same), the airplane will tell you just about anything you’d ever want to know about your flight with a few intuitive pushes here and there. The results pop up in bright colors, sometimes graphic, other times numbers, but designed to always be what you expected. 

All this is going on while the airplane is flying itself. Fly-by-wire systems are computers that keeps the airplane in the flight envelope while accepting pilot inputs to guide it throughout the flight. With intuitive side-stick controls the pilot can fly the plane more precisely with less effort than in any previous generation aircraft.

The results of automation have been hugely positive. In the last twelve years US air carrier operations flying billions of passenger miles have experienced one fatality. That was a Southwest Airlines uncontained engine failure, resulting in a passenger being fatally injured while the crew was managing to land the airplane safely.

Experts attribute this miracle of modern aviation safety to the fact that pilots have more information available to them than ever before in the history of flight. Cockpit automation systems, along with an operational environment that provides flight crews with flight plans that take every trip factor into account, and enroute resources to assist in managing any problem that could arise to a safe conclusion are the tools that deliver information to pilots that is vital to the safety of their flight. They reduce the guesswork in flying and the instances in which pilots must make life and death decisions to a fraction of what they were twenty years ago. 

Those instances of confusion that used to end in fatal accidents are history, for the most part, but pilots are now faced with a new challenge: recognizing when the automated systems are having problems and solving them while maintaining control of the airplane. Pilot confusion has haunted us throughout the history of flight, but what we’re seeing now is different. Where pilots used to struggle with position fixing, getting lost on complex approach patterns and ever present airplane system failures, the latest challenge is losing track of what the airplane automatic systems are doing and identifying when they are malfunctioning.

We have seen recent accidents around the world in which perfectly functioning airplanes flying in benign weather have flown in to the ground while the crew was trying to figure out what the plane was doing. To address this the operating community is re-prioritizing training to build pilot skills in diagnosing system issues and, equally important, maintaining control of the airplane, in a word, doing some of that pilot stuff.

USING SOME OF THAT PILOT STUFF

Here we address a couple of the fundamentals of flying that with modern flight decks spoiling us we have somewhat forgotten. A couple of stark examples are included to remind us of the consequences should we forget the basics.

If the plane is wandering away from its planned flight path, the first thing pilots should do is turn off the automated systems and fly the airplane, in proper CRM rules, one pilot flying the airplane while the other manages the malfunctions. But one of the root causes here is the very reliability of the modern jet. Pilots are so used to letting the airplane fly itself day in and day out, year after year, that their manual flying skills deteriorate to the point that taking over manually may contribute to the problem rather than solve it. 

In the face of a number of accidents worldwide the push now is to re-vitalize those flying skills. Pilots are now being advised to fly a certain number of legs manually, rebuilding their confidence in using them when the airplane is acting strangely.  

An especially tragic example: in 2009 an Air France Airbus A330 was enroute to Paris from Buenos Aires. The pilots lost control of the airplane, stalled it and fell into the Atlantic, killing all 228 passengers and crew. It took years to recover the flight and voice recorders but when they were found and the data played back it told a chilling story. The pitot tubes, probes that sense airspeed iced up and the flight data system warned the pilots that their airspeed and altitude indications were faulty. Then the fly-by-wire computers dropped off the line.

When their autopilot flashed a warning and dropped off, the crew, a new hire cruise pilot in the left seat and a copilot who had come up through the cruise pilot program couldn’t figure out what was happening. In almost complete silence between them one of the pilots pushed his side stick controller almost full nose down while the other pilot used full aft stick, all while the A330 plunged ever faster toward the water. It eventually struck the ocean surface in a 6,000 fpm descent.

The pilots’ performance failed in so many ways, no discussion or comparing notes on the problem ignoring the very basics of CRM, all while their lack of instrument skills prevented them from righting the plane and stabilizing its flight. 

In a positive contrast: a few weeks after this accident a Northwest Airlines A330 enroute from Minneapolis to Tokyo encountered the same pitot tube malfunction that started the chain of events on AF 447. When the Northwest pilots experienced the same autopilot warnings followed by it dropping offline, they simply took control of the airplane and hand flew it the rest of the way to Tokyo. Arriving at the gate, they wrote it up and left for their hotel.

The question is, at what point should we punch off the automation and take over the airplane? The answer is simple: whenever the airplane isn’t going where you pointed it. The problem we have is we’re spoiled with the new system. Flying them day after day on near-perfect profiles builds a sense of complacency and trust, to the point where if we see an anomaly we assume it is temporary and the airplane will correct it.

Or, if we’re new to this type of flying we assume the airplane knows more than we do, the old adage ‘what’s it doing now?’ the question for pilots new to the airplane, while ‘oh boy, it’s doing it again!’ the response from crews familiar with the system. Either way, we should never cede the safety of the flight to the airplane. We always have the option to fly it ourselves if there is any question about the airplane. Maintaining our flight currency to the point that we have the confidence to take over and safely take the plane the rest of the way is key to the flight’s ultimate safety and survival.

A NEW SET OF SKILLS

This requires not really a new set of skills but a re-sharpening of the old ones. First those of us who spend out lives watching the airplane fly need to re-sharpen our hand flying skills. We should be able, at any time, to take over the airplane and fully execute an instrument approach to minimums without a second thought. That requires we break with what has been current guidance for years, to let the autopilot fly, and instead, fly manual legs frequently enough to maintain our skills.

As well, we need to learn how to diagnose system problems so we can report the issues correctly to our maintenance partners. Were we able to reset things with a disengagement and reengagement or did it require resetting circuit breakers? Or maybe it requires changing the inputs around, Air Data 2 replacing Air Data 1, for example. The more analytics we can provide the maintenance technicians the better their success rate and fixing it. 

 What skills do we need to re-build or learn? A thorough review of system descriptions and diagrams in our training literature will serve as solid basis to then develop these techniques:

·      Interpreting CAS messages

·      Researching workarounds that we can safely implement in flight

·      Composing detailed notes from which we can report the problem in detail to our maintenance technician partners once we’re on the ground.

Those diagnostic skills will often help us get the airplane back on line so we can continue the trip instead of ending it prematurely, and do everything we can to make sure the maintenance technicians can correct the problem once we’re on the ground.

Please do offer your own thoughts on this subject. It’s large and complex, and we all have different views on it. Only by getting as many of those views out there can we hope to steer the thinking in the right direction.

So, please email me your views and they’ll get posted here!