Should the planes have a single pilot at cruise level?
Dr. Jose Sanchez-Alarcos (Eur. Erg.)
Aviation Psychologist. EASA external expert in Aircraft Design and Production, Aircraft maintenance, Mgmt & Org, New and emerging technologies and others. Member of European Commission Human Factors Group in ATM.
Should the planes have a single pilot at cruise level?
Let me to begin with an autobiographical paragraph: Before entering the Human Factors area, back in the early 2000s, I worked in Human Resources and, among other things, I oversaw workforce analysis, planning and adjustment.
A very common standard in workforce planning is 80%, that is, to consider that, if 80% of the working day is spent performing the tasks inherent to the job, the workforce will be properly sized. It is a standard that has numerous exceptions, one of which is linked to the peaks in the task flow.
Can we think of a firefighters' staff defined in such a way that 80% of their time is spent putting out fires? And a police staff defined in such a way that 80% of their time is spent arresting criminals? Obviously, such workforces would be understaffed since they would not have the capacity to handle a peak workload.
What about the pilots? It is curious that the suggestion of having only one pilot at cruise level appears precisely at a time when, from different sides, they are trying to treat the pilot as an emergency resource, leaving, as much as possible, flight management in the hands of Technology.
Let us follow, then, the logic of dealing with the pilot's job as something whose sense is, during the cruise phase, mainly in the attention to possible emergencies: Their workload is very high when they are on the ground or close to the ground while, at cruise level, this workload normally decreases a lot, but...let's underline the "normally". A search for critical cruise events in the ASRS database returns 5,553 crisis situations, as the picture above shows.
?Let's add that a serious aviation event can develop in an extremely short time and its management may require the full attention of at least two people.
Different events have shown that two people represent an absolute minimum that, in some cases, may be exceeded. Consider cases such as Swissair 111, where a fire on board required preparing a diversion to a nearby, unknown airport, searching radio frequencies and approach routes, dropping fuel to reduce the weight of the aircraft and, in addition, seeing if there was any way to extinguish the fire. There wasn't. Would things have been different with one more person? Hard to know.
Other more fortunate cases such as AC143, QF32 or U232 tell us that, in a phase of flight when the workload is usually low, situations may arise that require more than two qualified hands to deal with them; that’s exactly the same situation of policemen, firemen or any other profession that, by its nature, deals with emergency situations.
领英推荐
The promoters of the idea of a single pilot at cruise level justify it on the scarcity of pilots and reassure the public by talking about advanced technological support for this single pilot. Would the rationale of scarcity be used to reduce the number of pilots on board from one to none?
As for technological support, it has an interesting peculiarity that is reflected in an old joke: Some people define a banker as a man who lends us an umbrella and, when it starts to rain, asks us to give it back to him. Similarly, technology tends to make tasks easier when they are originally easy. However, it can become an added problem when the situation is difficult and does not allow us to perform actions that would be necessary, or when it behaves abnormally.
The popularization of Artificial Intelligence means only added pressure to a situation whose effects were already visible. With or without Artificial Intelligence, although more so when it is present, systems can have strange behaviors derived from their programming and, when these appear, instead of making tasks easier for the human pilot, they add a new one that is not exactly minor.
It cannot be forgotten that Information Systems lack common sense and leaving a single pilot in their hands as an alternative to an expert discussion of an unknown situation and the subsequent management of that situation is not reassuring or a guarantee of safety.
It can be argued that there is another pilot inside the aircraft and that, under this model, all that is being done is to avoid carrying backup crews. Be that as it may, but in that case, let me recall AF447:
The captain was out of the cockpit and, when he returned, the situation had become so unmanageable that, even when he was able to correctly diagnose what was going on, it was too late to recover the aircraft: An event that should not have gone too far - the freezing of a sensor - caused anomalous reactions in the system and they completely confused the pilots in the cockpit; when someone with more experience entered the cockpit, his greater knowledge was not accompanied by the availability of the time that would have been required to solve the problem.?
The last issue is the most obvious: What happens in the event of incapacitation - are they going to use practices like the railroads' "dead man's pedal"? It didn't work in the Waterfall case in Australia and it might not work in an aircraft where an event can quickly reach the level that makes it unmanageable.
On the other hand, how do you reconcile the idea of a single pilot at cruising altitude with the idea that the pilot should not be left alone in the cockpit, put in place after the GermanWings 9525 case? Does he have to be accompanied by a cabin crew member? Doesn't it make more sense for him to be accompanied by someone who, when faced with an emergency, can help assess and resolve it when, if it is serious, it is going to require the assistance of both?
Professional Pilot/Army Reserve Officer
2 天前Well said!
Flight Safety Specialist at British Airline Pilots Association
1 年We have seen the "benign" cruise phase be the area of some of the greatest avoidable tragedies. The causes vary, but automation, or essential elements of it, either caused it or failed to prevent it . Some in certification have already admitted that the existing software found in aircraft in service, cannot be tested to be absolutely certain of its response in all circumstances. Examples already exist of software unable to cope with even dealing with RNP approaches as designed, relying on the crew to intervene and resolve the issues. If AI is introduced there is an even greater lack of clarity as to what the system will do and why ? So, how will theses systems be certified? How will their performance be verified in service? If they use AI technology , how will it be prevented from acquiring " bad habits" ? Finally, has anyone really thought through the issues of incapacitation? Combine that with degraded systems ? If you only have one pilot , where do they get the experience needed to safely operate and command a half billion dollars plus of insurance liability on each flight ?
Captain of A320 family. LTC, TRI, CRM and LOFT Instructor.
1 年No. They shouldn’t. Pilots = Humans . We have the ability to deal with lots of treats during a emergency situation. And by the way. An emergbecy situation in real live it’s never hapen lime in a flight simulator. This is why aviation needs two pilots in the cockpit.