The global air transport industry agrees that pilots still need manual flying skills despite their highly automated work environment. But it cannot agree on how best to maintain this competency – particularly in instrument flying expertise.
An apparent result of the attrition of so-called “blind flying” skills is the number of loss of control in flight (LOC-I) accidents in recent years. Although extremely rare in the context of how many flights take place worldwide each year, LOC-I resulting in fatal crashes has occurred in a disturbing number of cases since 2000, and represents a high proportion of all the fatal accidents that occur.
These accidents were almost all events that occurred at night or in instrument meteorological conditions when the pilots were forced to take control of the aircraft, and the result has been at least 19 out-of-control crashes of scheduled commercial aircraft that were controllable, causing 2,012 deaths.
This presents the industry with a real cost/benefit conundrum: the statistical risk of fatal accidents is extremely low because of the success of modern cockpit technology – including accurate and reliable flightdeck automation – so pilots hardly ever need to handle the aircraft. Yet when they have to take over control, statistics indicate pilots are not guaranteed to be good at manual flying. The less they fly manually, the more they seem to need costly additional training to refresh lost skills.
The actual risk of events such as these is very low – about one fatal accident per 2 million flights. But a toll of more than 2,000 deaths since the millennium at a rate of 118 passengers and crew a year is difficult to write off as acceptable attrition, especially when the solution appears obvious: train – or select – pilots better.
The problem was set out with admirable clarity in a US Federal Aviation Administration-commissioned study of modern flightdeck pilot performance called The Operational Use of Flight Path Management Systems. And the International Federation of Airline Pilot Associations points out: “Automated systems are tools. Automated systems are not a substitute for well-trained flightcrew fully capable of operating the aeroplane in all aspects of flight.”
Ongoing debate
Debate continues as to whether it is safe to enable pilots to obtain manual flying practice at no cost to the airlines by allowing them to take control of the aircraft on revenue flights, as IFALPA says they must be able to do. No, says Europe, because when airlines allow it they know, from operational flight data monitoring, that there are more high-energy or unstable approaches and missed approaches. But yes, says the USA, because it makes sense to give pilots manual handling practice when conditions are safe.
This cultural difference of approach is striking. The USA’s weather is – most of the time – more benign and less fickle than Europe’s because a large part of the USA is at Mediterranean latitudes, whereas only southern Europe enjoys good weather most of the year. This results in a US air traffic management regime that is more ready to allow flight under visual flight rules. Also, most of the USA’s airspace – with the exception of some major terminal areas – has less intense traffic than Europe, the latter being a smaller space with a higher population density and more big cities in fairly close proximity.
Both Europe and the USA recognise the need to do something, but their resulting strategies have little in common.
The FAA’s Rob Burke, head of flight standards in the Air Transportation Division, says the Colgan Air Bombardier Q400 crash at Buffalo, New York in 2009 was “the straw that broke the camel’s back” for the USA. All 49 on board died when the twin-turboprop’s speed decayed during a night approach, the aircraft stalled, and the crew’s reactions were absolutely the opposite of a stall recovery drill. The aircraft remained stalled and spiralled into the ground.
Since that time, recurrent training in the USA has been required to include a different form of stall recovery than previously taught. It used to be a power-aided recovery on approach to the stall, with minimum height-loss, but now the full stall is examined, and concentration during recovery is on angle of attack reduction and regaining safe flying airspeed, with an acceptance of reasonable height loss to achieve it. Also added to recurrent training, says Burke, is the drill for managing flight despite the loss of reliable airspeed indication.
As for manual flying during commercial flights, Burke says the FAA encourages it under sensible conditions. “There are plenty of places in the USA where you can fly manually safely.” He gives Oklahoma City in good weather as an example. There has been no LOC-I loss in America since the Colgan crash, Burke points out. The FAA has published its philosophy on this issue in a SAFO safety alert for operators. Its central thesis is a reaction to the proposition that “continuous use of autoflight systems could lead to degradation of the pilot’s ability to quickly recover the aircraft from an undesired state”.
“Operators are encouraged to take an integrated approach by incorporating emphasis of manual flight operations into both line operations and training (initial/upgrade and recurrent). Operational policies should be developed or reviewed to ensure there are appropriate opportunities for pilots to exercise manual flying skills, such as in non-RVSM [reduced vertical separation minima] airspace and during low workload conditions.”
FAA statistics show that, in general aviation – the environment in which all civil aviators including airline pilots begin their professional learning – loss of control, usually from a stall, is the commonest cause of fatal accidents. The agency’s reaction has been to institute a programme of learning and awareness, with helpful, plain language advice easily available online.
Since June 2016 it has instituted a programme it calls Airman Certification Standards. The FAA explains: “The ACS improves the practical test standards [PTS] for the private pilot certificate by adding task-specific knowledge and risk management elements to each PTS area of operation and task.
“By integrating knowledge and risk management requirements with skill tasks,” the FAA continues, “the ACS offers a comprehensive presentation of the standards for what an applicant needs to know, consider, and do in order to pass both the knowledge and practical tests for a certificate or rating.
“This format helps applicants, instructors, and evaluators understand what the FAA expects in each phase of the certification process, from the FAA knowledge exam to the practical test. It also helps everyone understand how knowledge, risk management, and skill work together.”
This represents progress toward competency-based and evidence-based training, rather than box-ticking during the licensing process. Burke explains that the FAA, working with industry, carried out for the GA industry a tailored version of what the now-renowned CAST (Civil Aviation Safety Team) did in the 1990s to promote a new data-driven safety strategy in commercial air transport.
And because the lessons learned in early flying are more likely to stay with a pilot than any other aviation knowledge, it is important that trainee pilots not only pass tests, but understand fully what they learn, so that misperceptions do not follow them through an entire career.
The EASA recognises exactly the same problems, and it published a Safety Information Bulletin about the effects of automation on skills almost four years ago. It is working now with the industry’s Air Training Policy Group to update it.
Common approach
The thesis of the existing bulletin is actually the same as the FAA’s, but it is characteristically more cautious in its wording: “Operators are encouraged to consider incorporating emphasis of manual flight operations, as a means of maintaining basic flying skills, into training (initial and recurrent) and, when feasible, line operations.” That feasibility of incorporating it into line operations is understood to be one of the issues that will be addressed in the update, which should be published this year.
The caution exists because there are not many equivalents of Oklahoma City in Europe, and VFR does not apply anywhere within European controlled airspace. The following warning is part of the present SIB: “It is also important that pilots clearly understand the circumstances under which automated systemshaveto be used, such as during high workload conditions, while operating in traffic congested airspaces, or when following airspace procedures that require the use of autopilot for precise operations.” Those “precise operations” are common in European airspace.
Basically, European carriers would rather their pilots were given extra manual handling practice in the simulator than the aircraft. They have not actually banned manual approaches, but the advice is that locations where manual approaches or departures are to be approved should first be risk assessed, and they should not be flown completely raw – that is without auto-thrust and flight director, as well as without autopilot.
The safety management theory behind this approach appears admirable, but it does not indicate management confidence in the abilities of the pilots. Pilots need to be prepared psychologically as well as physically for the shock of an unexpected autopilot disconnect late at night. IFALPA comments: “Operators and regulators should not interfere with captain’s authority nor prevent crews from deciding when to practise their manual flying skills (based on good judgement, and deemed safe and practical) during line operations.”
Meanwhile, additional time in simulators that is purely for manual skills improvement, and which is not a part of recurrent testing or assessment, would clearly be desirable. In Europe Ryanair has voluntarily added an interesting layer to its recurrent training system, while Emirates has adopted its own training scheme to bolster manual skills.
The Ryanair Controlled Training scheme allows the low-cost carrier’s pilots to book time on one its company-owned fixed-base flight simulation devices. The flightdeck has all the capabilities of a Boeing 737-800 full-flight simulator, including an advanced visual system. It has an instructor operating station, but that will normally be unmanned, and pilots “fly” with their colleagues after choosing from a menu of “lesson plans”.
If the crew has trouble with a scenario, they can freeze the simulator and discuss it, or try again, but there is a time limit for the sortie.
“Our hope is that crews come to the FTD [flight training device], practice their skills, improve their knowledge and leave feeling good about themselves,” says Ryanair head of training Andy O’Shea. Sessions are recorded, but he says the company has “no desire or intention to review each session.”
Emirates gives its pilots half an hour of a flight profile that entails a demanding, manually flown instrument and visual approach that may entail an air traffic control go-around; some of it flown without the flight director.
IFALPA welcomes initiatives like this, but adds: “To help keep flightcrews from becoming overly dependent on automation, pilot skills in manual flight operations must receive regular exercise in the aeroplane during normal line operations. Maintaining pilot skills in manual flying operations only in a simulator is not the answer. It is important for pilots to practise their ability to manually manage the aircraft’s trajectory and energy during all weather conditions daily, and not just periodically in a simulator.”
While it seems that every part of the industry acknowledges there is a problem, universal agreement on how to deal with it may never be achieved.
Pilots: have you registered yet for our virtual jobs fair?
This free event for pilots is taking place online on 18-20 April 2017.Register today to secure your place.
Source: Flight International
