Pilot training has not been sufficiently modified to take account of the loss of the flight engineer nor the age of computer monitoring

Last week an event proved that the mantra about pilots of modern automated aircraft no longer being able to handle an aircraft manually is far from a universal truth. The crew of an Air Transat Airbus A330-200 was left with no engine power a long way from the Azores in the middle of the Atlantic. Coupled with this they had lost their autopilot, flight management system and flight envelope protection.

It was back to basics. But they managed.

An airport runway looks as small as an aircraft carrier's deck when you know you have only one chance. Not just to hit it in the right spot, but at a speed which will allow the aircraft to touch down safely. Any sceptics who said that a fly-by-wire airliner would be impossible to handle without engine power and that the crew would have forgotten how to fly have been proved wrong.

Although some airlines give their crews simulator practice at "dead stick landings", some do not, and it is not a requirement. The cold figures for gliding performance and configuration are there in the flight manual but - happily - it is not a well-thumbed page.

But without pre-judging the Portuguese accident investigators' report on the Air Transat event, it is easy to understand why many people in the industry are asking why a single-point fuel leak should lead to the loss of both engines in a short time. Fuel systems in all airliners can be managed precisely to determine which tank should feed which engine. Alternative solutions are handled with the appropriate use of pumps and fuel shut-off cocks. The final option is to open up the complete fuel plumbing system and have all tanks feeding all engines. That is not normal procedure but, subject to further study by the Portuguese investigation, in this case the system appears to have been open, and so the leak dumped all the fuel from all the tanks.

It would be easy to put the blame for any failure in fuel management technique squarely at the door of the pilots. That is unwise, however, not only because it would cover up other contributory factors within the system, but also fail to take account of distractions to the crew that have not emerged in the evidence yet but which were almost certainly present.

Having used the Air Transat case as an opportunity to raise the subject of modern aircrew skills, responsibilities and training, it is worth looking at the issue generically, and any parallels with this particular incident from this point onward are not intended.

Pilots in modern flight decks need a level of systems knowledge and understanding which compensates for the fact that they have no flight engineer. And with the demise of the flight engineer has come an increase in the size and complexity of modern aircraft, so more knowledge is necessary to manage the systems. Computer monitoring systems like the engine indicating and crew alerting system (EICAS), or the electronic centralised aircraft monitor (ECAM) are great for providing crew with information and even with tactical courses of action, but they do not confer understanding. The latter is essential for knowing how the situation developed and therefore what strategic action to take to keep the flight safe.

And knowledge - the component which is tested in pilots, normally by the use of multiple-choice questions - is of limited use without understanding. Understanding is much more difficult to test, and it is least well tested using multiple-choice questions. The airline pilot type conversion and recurrent training system, as structured today, suggests that the need for understanding has been forgotten or has been sidelined because the cost is too high. High quality, interactive classroom training is rare, and although computer-based training is improving, it alone will never provide the answer.

It takes more time to gain understanding than to gain knowledge, and time is in short supply in modern airline training. Over the last decade there have been increasing numbers of serious accidents which did not need to happen, and many of these showed a lack of crew understanding of aircraft systems - how they worked and even what they did.

The 1994 China Airlines Airbus A300-600 accident at Nagoya, Japan, was not exceptional in this respect, but it showed a crew who did not know how the autopilot worked, how it controlled pitch, and how to disengage it when take-off/go around had been selected. More than 250 people died as a result. Airlines pay for lack of understanding in their crews. It would be better if they paid for training.

Source: Flight International