Editor’s note: Exposing the dangers of reduced crew operations continues to be an ALPA priority. This nine-part series will educate, inform, and advocate support for maintaining the most vital aircraft safety feature: two experienced, well-trained, and well-rested professional pilots on the flight deck. Catch up with Part 1 and Part 2.

“Safety is our business. And thanks to the unmatched professionalism and dedication of every pilot flying today, we continuously celebrate our role in maintaining the safest form of transportation today. But to think that someone on the ground, potentially thousands of miles away, or the cold circuits of a computer can match the awareness or responsiveness of two pilots on the flight deck is, at best, a foolhardy notion that presents an unnecessary risk for everyone.”
—Capt. Joe DePete, ALPA President

As more experiments in autonomous flight are being made public, and more manufacturers look to form partnerships to get a foothold in this potential market, there’s a long road ahead before such modes of transport can match the safety record of today’s air transportation system. And that road would require the flying public to become participants in an experiment with no assurances of success or safety. This is a risk no one should have to take.

While recent milestones like the first flight of an autonomous Black Hawk helicopter in February have inspired imaginations and made headlines, the fact is that these tests are done in highly controlled—almost laboratory-type—environments that have been rehearsed and programmed. This sterile airspace presents few challenges to overcome as they’ve been reduced or removed altogether to ensure success. Ask any pilot what ranks among their greatest challenges while flying and they’ll most likely say managing unexpected occurrences like unforecasted weather, avoiding other aircraft, and emergency situations.

In short, an autonomous or reduced-pilot system can have perfect awareness of an aircraft’s situation and still fail to communicate a problem or develop a viable solution. Having two pilots on the flight deck is necessary to handle the tasks involved in flying an airliner, as a wealth of objective evidence shows that single-pilot operations significantly increase pilot workload to the point that safety is compromised due to an accompanying increase in mistakes and task shedding.

A September 2017 NASA paper on the effects of single-pilot operations illustrates this danger to safety. The paper describes a NASA/FAA experiment involving 36 pilots who flew seven flight scenarios aboard a B-737-800 flight simulator—only one scenario of which was nominal—under two-crew, single-pilot, and reduced-crew conditions. The experiment found that pilot workload increased significantly under single-pilot operations in the off-nominal scenarios, which ranged from benign hydraulic leaks to more serious issues such as dual-generator failures. The experiment found a direct correlation between the increased workload and the incidence of pilot errors, with a resulting decrease in overall safety, but offered no remedy to overcome these issues—outside of having two pilots on board.

One proposed solution to offset this increased workload is the use of ground-based pilots. However, an earlier NASA task analysis published in 2015 shows that such assistance doesn’t sufficiently offset the workload increase encountered under single-pilot operations. This task analysis found that under off-nominal conditions, such as a flight diversion, the number of tasks for an onboard pilot assisted by ground-based pilots increased by as much as 24 percent in comparison to the amount normally handled by the captain during standard-two pilot operations. Furthermore, to make financial sense as a replacement for standard two-pilot operations, ground operators (pilots) would have to be responsible for multiple aircraft at any given time. However, according to a NASA experiment that examined this approach, pilots can have difficulty compartmentalizing issues faced by these different aircraft.

Moreover, a 2017 NASA/FAA study on single-pilot and reduced-crew operations further indicated that single-pilot operations aren’t “acceptable” in an emergency because of increased pilot workload: “The pilots could overcome the circumstances presented, but rated the workload, safety, and acceptability as being unacceptable in an emergency condition. There were notable flight performance decrements during [single-pilot operations] compared to two-crew operations that suggest unacceptable reduced safety margins.” Assistance or intervention by a ground pilot would also be complicated by communications transmission delays introduced by the necessary signal encryption. Without such encryption, these signals would be at risk for tampering by unauthorized actors. And even when encrypted, access can still be compromised.

This is why history shows that having at least two fully qualified, highly trained, and well-rested pilots on the flight deck is an airliner’s strongest safety asset. Pilots’ knowledge and previous experiences serve as guidance and fuel motivation to effect a positive outcome to any scenario. Numerous emergencies have, according to computer simulations, been nonsurvivable. And yet, airline pilots have made a difference when computers indicated it was impossible.

Of course, pilots are human and subject to the laws of nature; therefore, having two pilots on the flight deck is the only reliable defense against the possibility that one becomes incapacitated during flight. Though the odds of a pilot becoming incapacitated or impaired in flight are statistically very low, the sheer volume of commercial air traffic globally translates into multiple incidents each year. In standard two-pilot operations, a key responsibility of the pilot not flying—the pilot monitoring—is to monitor the pilot who is flying the aircraft for errors or decline in cognitive capability.

In single-pilot operations, this critical redundancy layer is lost. The ability to reliably monitor pilot health using automated systems will require significant advances in technology, and a ground-based pilot who may be juggling multiple aircraft at any given time simply can’t respond as quickly to a situation in which the onboard pilot becomes incapacitated as would a co-located pilot. And while it’s assumed that a ground-based pilot would take control of the aircraft if the onboard pilot becomes incapacitated, this pilot would then become unavailable for other aircraft that may need assistance—were they assigned to attempt to support multiple flights.

In addition, in reduced-crew or single-pilot operations, instances of pilot incapacitation or impairment could be “catastrophic” and expose everyone to unnecessary risks. The NASA/FAA 2017 study concludes that entirely new automation and autopilot technologies would need to be introduced to address these and other issues associated with reduced-crew or single-pilot operations—especially taking into account that the national airspace is designed with two pilots and their capabilities in mind.

Having two pilots on the flight deck at all times ensures that two sets of eyes and hands are available to quickly identify, prevent, and correct any errors. Although statistics aren’t kept on accidents or incidents averted by pilot action, many recorded incidents of aviation emergencies illustrate situations in which this has been the case.