Why having more pilots in the cockpit doesn’t necessarily mean a safer flight
The first aircraft cockpits were rudimentary affairs. They just had a few mechanical instruments to provide the pilot with very basic information. However, already by 1939, cockpits had become quite advanced indeed.
For example, the Boeing 314 Clipper (first flight in 1938) required a five-member crew in the cockpit. Even on the Boeing 707 (1957), the first jet-powered commercial aircraft, there were four crew members: two pilots, an engineer and a navigator.
However, by the 1960s, more advanced jet engines resulted in increased reliability and power, reducing not only the level of maintenance and workload required but also the number of engines and improving safety. Indeed, radio navigation (including later satellite navigation like GPS) removed the necessity for navigation, drastically reducing the overall workload. Thus, early on in this decade, both the engineer and the navigator were no longer necessary positions. Furthermore, technology reduced the already small number of functions assigned to a third crew member.
Labour unions strongly opposed the reduction from a crew of three to a crew of two. Their position was that a ‘third pair of eyes’ improved the likelihood of detecting potential collisions—a belief that had plenty of public appeal. For the general public, having only two people in the cockpit intuitively felt less safe than having three people.
In the early years of the Boeing 737 (1968), aeroplanes flew with both two- and three-member crews, allowing the collection of a significant amount of comparison data. BAC-111 (1963) and DC-9 (1965) were the first large aircraft on the market certified for a crew of two. The results of extensive studies and ongoing comparisons, conducted by Dr Frank Ruggiero from the Boeing Flight Deck Technology group, convinced the company that the ‘third pair of eyes’ did not positively impact avoiding mid-air collisions. In fact, having a three-member crew made flights less safe.
Surprisingly, in the confined environment of a flight deck, a crew member who has a low workload could actually distract the rest of the crew. Thus, contrary to the labour unions’ reasons for having a ‘third pair of eyes’, it could actually lead to decreased operational safety. Specifically, from a psychology perspective, the optimum crew size is the absolute minimum number of people required to complete the work.
In 1981, the US Secretary of Transportation appointed a special task force to investigate and conclusively determine the relative safety of two- and three-member crew flight decks. There were ample safety data to explore, and the task force’s conclusion was that ‘three crew members are not safer than two’.
As a result, in the same year, the FAA deemed that a flight engineer was no longer required on any aircraft and that a two-pilot crew was safe. These achievements were due to the advances in cockpit displays, automation and the dramatic improvement in engine reliability reducing the need for in-flight troubleshooting. By 2001, all but the last of the three-person crew flight deck aircraft had been retired.
Why do we trust human pilots more than their machine counterparts?
Addressing individuals’ preference for human pilots over machines, Andrea Schlapbach, Daedalean’s Chief Commercial Officer, said the following: ‘There are two big reasons that humans are in the roles they are. One is tradition. Just because we did something like that yesterday, we do it still the same way today. Moving away from yesterday’s pattern is tough: it’s a significant effort in costs and time to convince the public, crews, unions and authorities that there is an equivalent level of benefit of safety if you change something big.
The other argument is that human beings are extremely good at coping with unknown situations. Say, Tesla autopilot makes a mistake—each fatal accident will be a big story in the news. But at the same time, there are 1.35 million people killed on the world’s roads a year (roughly 40 thousand in the USA alone), and it’s not a big story; this is acceptable. However, whenever you move to advanced automation, society applies a much higher standard. Killing 1,000 people by human beings in traffic accidents is socially acceptable. Killing one human being by a machine is a big story and not acceptable. In other words, pilots are in the cockpit because humans have much higher credibility for good—but also some self-illusionary reasons.’
Autonomy: the logical next step but also a revolution
Throughout its history, the aerospace industry has seen massive progress, becoming ever more autonomous. By automation, we mean any kind of functionality that is supposed to replace people in one way or another. For example, large aeroplanes used to have five pilots, but now there are only two pilots, and the autopilot and other means of automation keeping the aircraft in a healthy state play a vital role in each flight.
More advanced jet engines brought both more reliability and more power, which reduced demands on the crew while also increasing safety, and radio navigation (including GPS) took away the need for a human navigator. Other workloads were also reduced: the likes of air traffic control (ATC), airway and airspace design, traffic and collision avoidance systems (TCAS) and automatic position reporting (ADS-B) impacted traffic, and autopilots (AP), envelope protection and instrument landing systems (ILS) impacted flying and landing requirements. Computers further increased automation, while digital screens enabled centralisation and reduced workplace complexity.
But for the last thirty years, there really haven’t been any major changes. Full autonomy—meaning pilotless flights—if and when it comes, will be a revolution.