David Learmount/LONDON
Airline accidents are occurring because many pilots do not understand their aircraft, according to Germany's Darmstadt University. The institution suggests that the course for an air transport pilot's licence (ATPL) should involve degree level studies to gain the depth of understanding needed for modern airline flying. The university is not alone in its belief.
The idea was first floated by the German and Spanish pilots' associations. Vereinigung Cockpit (VC, also known as the German Cockpit Association) and the Asociacion Espanola de Pilotos (AEP) are working together with universities in their respective countries to create an ATPL curriculum that is not "20 years out of date", as the VC describes the existing training requirements.
Accidents resulting from insufficient pilot knowledge of aircraft systems and procedures - H3-category human factors accidents - are on the rise, according to the International Civil Aviation Organisation (ICAO). The proportion of serious H3 events involving "third-generation" aircraft has quadrupled, says Darmstadt's Professor Dr Gerhard Faber. This illustrates the need for a more complete education for those who choose a career handling and managing these progressively larger, higher performance, more complex machines, he says.
Recent examples of H3 accidents include:
the February 1996 Birgenair crash into the sea near Puerto Plata, Dominican Republic; the December 1995 American Airlines accident at Cali, Colombia; the April 1994 China Air Lines (CAL) crash at Nagoya, Japan.The first two involved Boeing 757s and the third an Airbus A300-600. The pilot error - or pilot confusion - was different in each case, but all three events involved pilot ignorance of some aspect of the automated flight control or flight management systems (FMS). In the Cali accident, confusion about why the FMS was turning the aircraft in an unexpected direction resulted in controlled flight into terrain (CFIT). In the Birgenair and CAL accidents, which ended in loss of control, the pilots showed lack of comprehension of what the aircraft was doing and why. Their problems included FMS mode confusion, poor knowledge of how to interpret systems information presented on displays, and ignorance of the interfaces between autopilot, data sensors and control surfaces.
DEFINITIVE STUDY
The only comprehensive study of H3 accidents was produced by the US Federal Aviation Administration-led human factors team in 1996. The ensuing report, The interfaces between flightcrews and modern flight-deck systems, examined in detail pilot training for highly automated modern airliners and concluded that it was inadequate. Training in systems knowledge and in the skills required to make the best use of automated flight systems (AFS) were highlighted as poor.
Most pilots first encounter highly automated flightdecks during type conversion, which, by definition, assumes piloting skills and is seen as "differences training", rather than a course involving fundamentals. Type conversion can be a pilot's first encounter with an FMS, however.
Given the brevity of the type conversion course, it is tacitly accepted by the airline community that "on the job" experience will complete the training task. The human factors team observed a particularly high reliance upon on-the-job training where sophisticated flightdecks are concerned. This has led to increasingly tight national crewing regulations for highly automated aircraft types. These rules specify that the captain/co-pilot mixture shall not consist of two pilots who both have low levels of FMS or aircraft type experience.
From the airline's point of view, the task of type conversion training is to show the pilot how to operate the aircraft and its systems, and to provide an adequate level of knowledge about the whole machine. Some airlines acknowledge a "classic" pilot's total lack of AFS knowledge by providing the flight manual to read a month before the conversion course begins.
The concept of teaching generic AFS principles, highlighting what the system does well and when its use should be avoided, as well as introducing automation-specific human factors issues such as mode confusion, is low on the list of instructional priorities. Yet the FAA report gives several examples of confused crews, just before a fatal accident, engaging the autopilot at a time when manual flying was the best or only option. Since the report was written, there have been other fatal examples of what appears to be crew mode confusion, combined with injudicious automatic flight systems usage, including the 16 February China Airlines Airbus A300-600 crash at Taipei, Taiwan.
Vereinigung Cockpit, setting the rationale for what it calls University Level Training for Pilots, states: "Especially in the case of system disturbance, pilots need complete knowledge of the aircraft systems and subsystems in order to make sound decisions-third generation civil jets have a considerably higher system complexity than older models. Operators of such highly automated hybrid systems need better training [and] higher qualifications."
As for the flightdeck, the FAA report says: "Current standards for type certification and operations have not kept pace with changes in technology and increased knowledge about human performance." Lack of situational awareness was a major area of concern. Although the advent of terrain awareness products such as the enhanced ground proximity warning system promises to reduce the number of CFIT accidents, technology does not have the answer to other forms of crew confusion.
The FAA report says the issue cannot be resolved merely by identifying a few operational bullet points and providing individual cures. It explains: "We need system solutions, not point solutions to individual problems."
Changing the system starts with investing in the people involved, it says. What is needed is "-investments in people: designers, users, evaluators and researchers. For example, flight training investments should be rebalanced to ensure appropriate coverage of automation issues."
EARLIER TRAINING
VC and AEP, working with Darmstadt and Spain's Rego university, agree. For pilots, they want training to start earlier and be more comprehensive - the equivalent of a three- or four-year bachelor of science (BSc) degree course rather than the 18 months ab initio training given today. It should be recognised, they say, that the digital equipment that was supposed to make the pilot's task simpler has, because of its own complexity and the removal of the flight engineer, introduced a need for the pilot to learn a new range of skills without dropping any of the traditional disciplines.
The University of North Dakota (UND) has since the late 1980s offered a four-year degree course in aeronautics, which combines its Spectrum ab initio pilot training course with a choice of aeronautical or air transport related studies. The graduate emerges with a multi-engine licence, full instrument rating, multi-crew experience and either industry or aeronautical studies that go beyond the pilot licence requirements.
In the USA, all US nationals are self-sponsored, and the UND admits that candidates for a piloting job usually go just for the ab initio course, especially with the present buoyant job market for trainee pilots. If they begin a degree course and are offered an "internship" with an airline, they often drop the academics and go for the licence qualifications.
US airlines have a tradition of putting a high value on pilot log-book hours, no matter how they have been gained, says the UND. Large carriers have been accustomed to getting their pilots from the military and the commuter airlines. The commuters recruit from the area of general aviation, where instructional and utility flying can provide lots of hours, if not the ideal background.
Now that the US air transport industry has been rapidly expanding for several years, carriers are having to consider a source of pilots which Europe has always accepted: the fully trained (but low-hour) candidate.
It is not only expansion that has prompted US airlines to rethink their traditional pilot sources. There is about to be a "disproportionately high" level of retirements among post-Vietnam former military pilots, some of whom are now pushing the legal age limit of 60, says the UND. "Airlines, particularly commuters, are now moving toward the acceptance of low-time pilots, providing they have a good integrated training background," it says.
For these pilots, educational background is important: they do not have to be at degree level, says the UND, but they must be "college graduates who have learning capability". The University acknowledges that the main problem with a four-year degree course for pilots is that that it is not sufficiently responsive to the supply and demand dictates of the industry, which is highly impatient with the lag that such a long course builds in.
WHAT AIRLINES WANT
In Europe, on VC's doorstep, Lufthansa says it likes its pilots to have appropriate degrees, although it admits they are not essential to piloting. If one of its flightcrew has gained a relevant degree before, during or after qualification, the airline will encourage the individual to make use of it as a management pilot. In 1990, Lufthansa sent some of its pilots to the London, UK-based European Business School for nine months of management studies and the German flag carrier may do so again. Meanwhile, the carrier's attitude is that further education for pilots is "-not vital for the moment. When young people are training to be pilots, the pilot training is the most important".
British Airways' attitude is similar. It chooses pilots for their command potential as well as for their flying skills, and while higher education may contribute to captaincy potential, it is not seen as essential. Sponsorship for the further education of BA's pilots is an option. Chief training captain Capt Gordon Holdaway says: "We have always been happy to provide educational sponsorship for pilots providing that we believe it will be useful for their career development." There is always a need for some management pilots, he points out.
BA offers £20,000 ($33,000) a year in sponsorships to pilots who want to take the MSc course in airline management at London's City University. The airline does not press its flightcrew to do such courses, says Holdaway, explaining that it is "a huge commitment" of at least 10-15h of study a week while the applicants continue their jobs as line pilots. About 10 people a year take up these postgraduate courses, says BA.
Meanwhile, City University, working with two UK-based flying training schools, the European College of Aviation (ECA) and the Oxford Air Training School, is devising a three-year course leading to a BSc in air transport operations. Graduates will emerge with the degree and a "frozen" air transport pilot's licence (acquisition of 1,500 flying hours confers the licence's full privileges).
The first year will involve academic and aviation studies and the second will consist of flying training at either of the two schools. The third year, taken at the University, is described by City as "group and individual project work associated with the airline industry". The total cost is estimated at £50,000. ECA marketing chief Colin Green says the captaincy course will aim to produce flightcrew with a high captain and management-pilot potential.
Until about a year ago, BA had a module called airline business training (ABT) in the ab initio course for its sponsored cadet pilots. The intention was to encourage business and customer awareness in young pilots. Now, according to BA head of flight operations strategy and resources Paul Douglas, the ABT module has been moved from the ab initio phase to the early line-flying stage soon after type conversion.
BA provides a one-week residential "captains for the future" command development course for pilots approaching captaincy. Douglas says that this course is designed to instil the management skills "-which we, rather naively, used to assume they automatically had".
FLIGHTDECK FODDER
This may be good, character-developing education, but it is nothing like the flightdeck and systems academic study that German and Spanish pilots would like to see as a fundamental part of commercial pilot's licence qualification. The aim of most existing postgraduate studies is to broaden the pilot's areas of knowledge rather than to deepen them.
In the UK, where, until recently, the Government was offering tax concessions to people who paid for training towards a pilot "national vocational qualification" (NVQ), the airlines so manifestly failed to take part in the system that the NVQ course for pilots has now lapsed.
Originally intended to be similar to a degree, but heavily vocational in content, the NVQ encompassed all the existing study and practical skills training involved in becoming a pilot, while adding some air transport-related, but more academic, studies.
The UK Aviation Training Association, which believes that the NVQ for pilots might be revived, says that it is likely to comprise ab initio training, plus a study module on a subject such as "safety critical systems management".
Airline co-operation is required, however. Because this failed to materialise at the first attempt, the evidence suggests that most carriers just want flightdeck "fodder" as cheaply as possible. Meanwhile, with cadet pilot sponsorship becoming increasingly rare, ab initio pilots themselves do not want anything that adds to the expense of becoming a pilot, or which increases the time it takes to start earning a salary.
PILOT CRUSADE
This may make depressing reading for Germany's VC and Spain's AEP pilots, but it is unlikely to stop their crusade. These pilots want to able to take a fully fledged degree in the sciences associated with operating modern airliners. This, says Darmstadt, would enable students in countries where university courses are still fully state-funded to gain a higher level of expertise at no extra cost to themselves or their eventual employers.
Figures for the increase in H3-category human factors accidents are incontrovertible, thus confounding the hopes of airlines and manufacturers that pilots could be replaced by quickly trained, low-cost system operatives.
Those who wring their hands in despair about human factors accidents are now confronted with a plausible proposal for reducing this risk by increasing the depth of pilots' professional knowledge. It will be interesting to see how the industry meets this challenge.
Source: Flight International
