Selling Ourselves Short

“The Autopilot doesn’t know what a mountain is…”
An SOS for airline pilots?

It is true that we live in an ever changing world. One where the assumptions of the past, fail to live up to the reality of today. Take pilotless planes for example. It seems like only yesterday that some pundits in the aviation industry were predicting that the continuing development of unmanned air vehicles (UAVs) would lead the way to airliners which could fly their passengers across great distances without any pilots being onboard at all. There were some of us sceptics who would respond, “Yes, it is technically possible to build and operate a large four engined UAV with 500+ seats onboard which could fly from A to B without a human hand to guide it — but you try selling the tickets!?”

Now however, the world has moved on. We are into the era of pilotless road vehicles already. In the USA (it happens there first and the rest of the developed world follows suit), they have been road-testing on a grand scale. Trucks for starters. Not just one or two vehicles, but literally dozens of them are out there on the highways trucking up and down the blacktop. America does have a vast number of truck drivers and they are an expensive resource which the industry would love to reduce.

…computers and machines do not need vacations, nor pensions, or dependency leave, not even a dental appointment…

You see much of the country’s commercial goods travel by road and to operate each truck requires manpower. If you talk to any HR person they will tell you that when a company or enterprise employs a person they are not just paying for that person’s time, but also take on the liability for their sickness, holidays, absences and their dependents’ welfare. It doesn’t take a large cognitive leap to see that computers and machines do not need vacations, nor pensions, or dependency leave, not even a dental appointment…

Computer guided machines which operate other machines, do not go absent from their duties in the same way as us ‘Carbon life-forms’. This is why on paper you can see that the principle of driverless machines has such great potential. As a professional aviator I have been aware of the often voiced opinion of many colleagues from an early stage in my own career that we are viewed as an expensive inconvenience to many airline Financial Directors.

The Bean Counters have had a good run in recent years with being able to drive down the costs of providing pilots. Firstly the raising of the retirement age to 65 meant that suddenly there was a surfeit of qualified, experienced professional aviators and consequently there were fewer new entrants required. This in turn had the effect of reduced terms and conditions for pilots — the law of supply and demand.

Then the Low Cost Airlines (LCAs) managed to introduce self-funded training systems for junior pilots such that the expense of type-rating training was transferred from the company to the individual. Also the introduction of ‘Pay to Fly’ schemes — whereby the new entrant First Officer in the airlines actually pays for their own Line Training, meant that the airline business model changed significantly.

Finally reduced salary for junior FOs meant another benefit to airlines balance sheets. Very recently however (the last two years) it has become apparent that there is a shortage of airline pilots. Fewer ex-military pilots are available and the temporary glut of pilots in their early sixties has finished. More significantly perhaps there are fewer volunteers willing to make that six figure investment in a career which now has much less potential financial return. Finally some youngsters are dissuaded by the often rumoured future prospect of pilotless cockpits.

On this last point, I do not think they should worry. Although there is now accelerated development towards pilotless flightdecks for future airliners, the designers are heading up the proverbial blind alley. With the advent of Artificial Intelligence and computer systems so advanced that intelligent robotic vehicles are in use — driverless shuttle trains at airports for example — it is an easy step to visualise a future which include airliners with no flightdeck crew onboard.

Selling the tickets? That would be an easy one, simply give away the tickets at first. There would be no shortage of passengers willing to fly for free from London to Sydney for example. Once the first few weeks of scheduled operations proved ‘trouble-free’, people would start to pay.

The technology is already available to fly airliners automatically from airport to airport, so you have to ask the question, “What’s stopping us?” The answer is simple — the problem is the operational environment. To enable pilotless airline traffic to operate safely the whole air traffic system and ground operations systems would need to change; radically. Recently I experienced a couple of busy multi-sector flying days as an airliner Captain with a huge number of challenges. These ranged from operating to/from small island airports with significant performance limitations and non-standard ATC, (short, narrow runways with reduced margin for error).

Then we had another sector with passengers not arriving at the airport gate despite having checked in with hold baggage — “Find those bags, offload them and let’s go!” I even went down to the cargo hold to assist in the timely locating of the bags myself — we got away on schedule, but only just… Complex ground traffic situations with difficult to interpret R/T instructions at strangely designed airports. Last, but not least we had an in-flight medical emergency in which we had to consider the possibility of an enroute diversion. It is true to say that we as a crew expended considerable intellectual energy on all of this and by the time we had finished this demanding series of flights I was glad to rest on my days off.

“the car ASSUMED that the bus would yield when it attempted to merge…”

Pilotless airliners will only be able to operate safely to and from specially adapted/designed airports where the infrastructure has been completely modified to facilitate them. To change the airport infrastructure worldwide is going to be too big an ask. A perfect example of what I’m talking about occurred only last week with one of the Google self-drive cars. There have been a few accidents with the fleet during the testing, but this one was significant, because it was an event which could not be explained away as a human factor error.

The excerpt from the Google accident report states that ‘…the car ASSUMED that the bus would yield when it attempted to merge back into the traffic lane…’ This is supportive evidence of the proposal that a computer is “only an imperfect human brain”. **

The question the industry experts need to ask now is, “do we WANT imperfect human brains flying our airliners?”

Copyright © James McBride 2016

  • * Professor Katherine Fenton OBE

(note this story appears as Chapter 34 in the recently published book FROM THE FLIGHTDECK available on Amazon — see link below).


One clap, two clap, three clap, forty?

By clapping more or less, you can signal to us which stories really stand out.