Would You Get on a Plane With No Pilots?

Or how about one with only one pilot? Sooner or later, you may not have the choice. In a recent article in the Wall Street Journal, (subscription required) it was revealed that NASA has commissioned a study to be conducted by Rockwell Collins to explore the feasibility of single pilot airliners:

Facing potential shortages of airline pilots and dramatic advances in automation, industry and government researchers have begun the most serious look yet at the idea of enabling jetliners to be flown by a single pilot. 

All large commercial jets for passenger and cargo service world-wide now fly with at least two pilots in the cockpit. A new study by the National Aeronautics and Space Administration and Rockwell Collins Inc. will focus on the provocative idea that co-pilots could remain on the ground, remotely assisting solo aviators on the flight deck during the busiest parts of flights, said John Borghese, Rockwell’s vice president of its Advanced Technology Center.

Actually, it wouldn't be quite fair to call the remaining human being on a single place airliner a pilot at all, for his required skill set would almost certainly not include the "stick and rudder" skills of today's pilots. The remaining person would be a systems manager, overseeing the computers which would do the actual flying.

There are actually very good reasons why you would not want a very bored and very rusty pilot just sitting up in the cockpit waiting for something to go wrong so he could grab the controls to save the day.

For starters, humans are uniquely unqualified and unsuited to sit around to watch and monitor machines. Most humans have an attention span of perhaps 20 minutes before the mind starts to wander. This type of arrangement is quite nearly the reverse of the ideal human-machine interface.

If you'd like to try this out for yourself, simply sit in the laundry room and watch your clothes washer closely to make sure it doesn't skip a cycle. It's not likely to happen, but if it does and you miss it, you and all your passengers die. No falling asleep! (Washers are probably slightly more reliable than current aircraft automation, but the analogy holds.)

Secondly, piloting, or "stick and rudder" skills take years to acquire and need to be maintained with routine practice. Neither of these conditions will be available in the cockpits of the future. Heck, they are hardly even available today! We are currently coasting on a slowly draining reservoir of pilot skills attained in the years before automation became pervasive.

Children of the Magenta Line

Pilots entering the profession today get rudimentary training before graduating to their first commuter airline job which will be in a glass cockpit with automation. These "children of the magenta line" (a reference to the electronic magenta line on the course display) will never develop the piloting skills their forbears recognized as their stock and trade.

This concept of in-flight computer system operators was demonstrated succinctly last year by the crash in San Francisco of an Asiana 777 on a clear and calm day. The "pilots" aboard that aircraft had many thousands of hours of flight time safely operating jumbo aircraft across the ocean with many hundreds of passengers. Unfortunately, they didn't know how to actually "fly" the plane when they needed to and they'd been doing it that way for years.

The big mistake made by the managers at Asiana was of getting ahead of the current state of the art. The automation deployed on the current generation of commercial aircraft is good, but it was never designed to be all encompassing. That is, on occasion a pilot may actually still be needed to fly the airplane the way the Wright brothers did.

Crossing the Bridge Safely

Now please don't misunderstand me. I am no Luddite arguing against the eventual denouement of my profession. The piloting profession as it been constituted since Kitty Hawk is in decline and no amount of feather-bedding will change that. And this is most likely a good thing.

Commercial air travel is now safer than it has ever been and has levels of safety which are probably rivalled only by the elevator. Automation has played a large part in this. But like the elevator, the nature of commercial aviation is going to change and drastically so.

The question is not where the profession is headed, but rather how to safely get there. As automation becomes more robust, the need for piloting skills will diminish on a gradual level. The challenge will be how to bridge this ebbing of piloting skills with the gradual increase in the capabilities of automation until such time that pilots are not needed nor desired. 

The proposed study by Rockwell Collins is much less ambitious in its objectives seeking only to explore the feasibility of single pilot operations while a second pilot would be at the ready on the ground to assist if needed:

Under the concept the researchers are studying, aviators on the ground could be assigned to assist solo cockpit pilots on multiple flights, virtually co-piloting during the busiest times through crowded airspace, approach-and-landing maneuvers, or if something goes wrong. “It’s a reasonably new area” to study how the notion may apply to large jets, according to Parimal Kopardekar, the program’s manager based at NASA’s Ames Research Center in northern California. When pilots need a midair rest or bathroom break, those on the ground even may “need to baby-sit the vehicle,” he said. 

Such a dramatic shift won’t happen any time soon, and there is virtual consensus that reduced crews for passenger planes won’t be considered until they are introduced first in the cargo arena. That is unlikely to gain traction much before the end of the next decade, according to experts and airline officials. 

Jets today are designed to have two pilots behind the controls, and retrofitting existing aircraft “may be too expensive and may be too difficult” to obtain regulatory approval, according to NASA’s Mr. Kopardekar. Industry officials say all-new aircraft would be needed with cockpits designed from the start with a single pilot in mind.

It is an open secret that Fred Smith, FedEx founder and CEO has at the top of his bucket list the firing of at least half, if not all his pilots. And there is little doubt that the first single piloted commercial aircraft will be a freighter. Other than that, it is highly unlikely that there will be any single piloted commercial aircraft for at least the next few decades.

Boeing's latest generation technology, the 787 Dreamliner and the forthcoming 737 Max aircraft have all been designed for two pilots as have the latest offerings from Airbus. Given the average technology cycle of about 20 years between major upgrades, only the aircraft introduced to replace these new generation aircraft are likely to be designed for single pilot operations. So while the change may take a few decades, most experts view it as inevitable.

Also necessary for this change will be a wholesale cultural shift in public opinion towards automation. This shift is already underway. A century ago, it might have been unthinkable for any man on the street to consider getting on a train that didn't have an engineer. Now we routinely board completely automated trains taking us to our airline gates.

Likewise, luxury cars are already coming equipped with automatic lane-keeping alert and sudden stop warning systems. Google's driverless cars are already combing the countryside, recording all they see for Google maps. When cars which can parallel park themselves become ubiquitous, a skill many drivers never master, highly automated aircraft overseen by a systems engineer will not seem so far fetched.