New technology aircraft: training issues

“Glass cockpits” entered commercial aviation in 1979 when the McDonnell Douglas MD80 aircraft entered service. This presented a significant advance in aviation technology as it condensed and displayed systems and flight data to the crew more vividly. Special training for this was minimal as it only required familiarity with the displays - the information and data were the same as for an analogue display.

The really significant development in automation technology came about in 1988 with the introduction of the Boeing 747-400 and the Airbus A320 series aircraft. These proved to be a major change (reducing the number of lights, gauges, indicators and switches in the Boeing 747 cockpit to 365 from 971) and introduction of “fly-by-wire” technology into commercial aircraft. For pilots and engineers it was in many cases their first encounter with computer technology.

The design philosophy was simple; to reduce the workload of the pilot and eliminate the need for a flight engineer. Despite the early concerns associated with inexperienced pilots and engineers in dealing with new technology, ie: the interpretation of data from the various computer systems. There was a need for a level of confidence in automation to be rapidly acquired, and thanks to extensive training and support networks provided by the manufacturers this was achieved.

The level of information available to a pilot on a modern aircraft is very high, you have aircraft (Airbus A380) with 10en flat screen displays in the cockpit, whilst only two screens are scanned by each pilot, the PFD (Primary Flight Display) and the ND (Navigation Display) during normal operation, each screen is categorized and distinct, whereas the other screens are primarily aids, providing support data.

The Boeing 787 has five screens with all information and data can be viewed on the centre (EICAS) MFD (Multi-Function Display) as well as the respective PFD (Primary Flight Display). As with the A380 to ensure that the loss of a screen does not inhibit the crew’s ability to perform, the screens or MFD’s can be selectively interchanged to ensure that there is no loss of visual data.

However, this tends to lead to “information overload” where information in the form of a “Status” message will appear as a result of a valid input, however reaction to the message can overlook the correct procedure to follow, resulting in a “what’s it doing now?” response. For these situations high levels of intense training, particularly cognitive training and a build-up of experience is required. This allows the pilots to more effectively, manage and control the aircraft.

To ensure that this level of automation can be managed the aircraft has a substantial data network of computers with electronic control “checklists”, these allow the pilot to perform sequential tasks, such as  systems control, flight planning, also emergency checklists in case of system degradation, failures, fire, decompression, loss of electrical power supply and engine failure.

The latter of these was graphically demonstrated over a year ago on a Qantas Airbus A380 where the mechanical failure of number 2 engine resulted in significant damage to the aircraft including the severing of data bus cables and electrical wiring, resulting in an unprecedented flow of condition/alert messages, appearing on the ECAM (Engine Condition and Aircraft Monitoring) display.  These messages are displayed in an order of priority depending on the level of severity.

This event resulted in the crew of two pilots (augmented on this occasion by two additional senior pilots) being presented with an intense workload, so far unique in new generation aircraft operations. The level of activity was centred on flying and controlling the aircraft with multiple system failures, adding to this workload were the high number of ECAM messages (approx. 54) that required attention and mostly in sequence to ensure that the aircraft would continue to operate safely under the (manual) control of the pilot. Their combined systems knowledge played a key part in prioritising their actions to successfully land their aircraft. 

Training for such extreme events is very important, whilst the aircraft simulator (synthetic training device) is designed and programmed to be able to induce the many events and conditions that pilots could expect in the course of their flying, it cannot accommodate every conceivable event. It relies on exceptional airmanship, good training and a high level of discipline to fill the void. This is an area where CRM (Crew Resource Management) is a critical factor including “Situational Awareness”. Working as a team in the cockpit the two pilots through training and retraining reduce the workload and chances of errors. Pilots on new generation aircraft are now expected to function as “system managers” adding further pressure on the levels of training.

The adage: “if you think training is expensive, try little or no training!” is quite appropriate.

The case of Qantas QF 32 was different, as this aircraft (Airbus A380) was a fully automated aircraft where the crew had to “compete” with the automated systems to bring the aircraft to a stable enough condition to land safely. There is no specific training for these types of events it relies on exceptional airmanship, cognitive skills and coordination to successfully land the aircraft, (albeit with two extra supernumerary crew members).

This is an area of concern as the emphasis on physically “flying” the aircraft has been reduced drastically and been replaced by “operating” the aircraft, and make “pilots-as-systems-managers”. The argument that the aircraft are far more reliable as more information is available is one that surfaces regularly however, as statistics tends to show from the numerous accidents and incidents around the world on new technology aircraft that this is a shortcoming which has been a contributing factor in a number of cases. Despite this, statistically, flying has never been safer, there are less events and fatalities throughout the world, despite the dramatic increase in commercial operations, this is primarily due to the technological advances in aircraft design and manufacture.

However, to maintain this level of safety, we need to seek out and mitigate new hazards as they arise.  Technology has mitigated old hazards such as inaccurate navigation, systems operation, aircraft systems reliability has improved resulting in the pilot’s physical workload being reduced.

 

Training

 

Thorough and effective training is expensive not only in terms of time and resources and time management, but also equipment. The various synthetic training devices such as a Full Motion Simulator or even a Fixed Base Simulator require large capital investment (upwards of $40M in some cases) in addition to the ongoing cost of maintaining and supporting the equipment and the facilities. It is less expensive than actual flight training in an aircraft. Operators that suffer cash flow or revenue generating problems would keep these costs to an absolute minimum, and in some cases at the expense of providing effective training for their pilots.

A new addition to an approach to safety is the introduction of SMS (Safety Management Systems) mandated by regulators to be incorporated into an organisation’s procedures, that specifically calls for a consistent approach to all aspects of training.

There are some basic elements to training we have to accept on new technology aircraft. They include the understanding of:

  • How information is displayed
  • What is the information telling the pilot
  • When required where to locate the appropriate data
  • Managing this data
  • Secondary or flow-on effects of the data
  • Sensitivity of controls

However, there still remains the ongoing issue of how do you ensure that the training that is prescribed by the manufacturers and required by the regulator can effectively instil the confidence to the pilot and to the level of competency that is required. Experience and continuing training (refresher) are two obvious examples, cognitive training is another where during training the instructors will link an action or event to a secondary event to ensure that a thorough appreciation of the action on the part of the pilot under instruction will be realised.

The various phases of flight comprise of:

  • Pre-flight
  • Take-off
  • Cruise
  • Approach and
  • Landing

The workload is different for each of these. During operations, the PF (Pilot Flying) the aircraft needs to be cognizant of the existing conditions during all phases of flight  The PNF (Pilot Not Flying) is required accommodate the workload assigned by the PF who assigns tasks essential to cockpit procedures (ie; observation, communications and call outs).

The risks associated with oversight in the multiple operating tasks, indicators and alerts have been significantly increased, in new generation cockpits. Oversights or failures to respond to these factors can be termed; “Cognitive Dissonance” where simultaneous events or actions required are inadvertently overlooked, or not given the necessary priority. This gives added weight to the principles of disciplined CRM and Situational Awareness, the following represents an event where this was not followed:

In June 2009 with the tragic accident of Air France (AF447) A330 aircraft over the Atlantic where three pilots were unable to recognise errors induced by icing of the Pitot /Static ports resulting from flying into a known storm cell.

The reported lack of training of the two pilots flying in how to deal with an unreliable airspeed indicator or flying the aircraft by hand was a contributing factor. The flight data recorder and the voice recorder of this flight also revealed it was not immediately realized that they were “heading directly towards an area of intense storm activity resulting in the pilot to ‘react irrationally’ and pulls back on the side stick to put the airplane into a steep climb, despite having recently discussed the fact that the plane could not safely ascend due to the unusually high external temperature”.(Ref: French Accident Investigation Authority (BEA) Report: Erreurs de Pilotage (Volume 5).)

This event has raised questions from investigators and regulators as to; “whether training, instrumentation and cockpit procedures can be modified (around the world) so that no one will ever make this same mistake again or for that matter the inclusion of the human element will always entail the possibility of a catastrophic outcome”. Ref: BEA Report.

 

Systems design

 

There are operational differences between design philosophies by various aircraft manufacturers; some of these are subtle and do not pose an excessive burden on training and training resource requirements. However, in some cases where a pilot has to transition from one manufacturer to another these can offer significant challenges.  There is debate as to which of the products are considered most operationally safe and effective, particularly in regard to which is more operator friendly the “Sidestick” cockpit or the “Control Column” cockpit.

Various air accident investigations have revealed that the greater proportion of accidents is human-error, with pilots taking the major proportion of the statistics.

Causes of fatal accidents by decade (percentage)

Cause

1950s

1960s

1970s

1980s

1990s

2000s

All

 Pilot Error

40

32

24

25

27

26

29

 Pilot Error  (weather related)

11

18

14

17

21

17

16

 Pilot Error (mechanical related)

7

5

4

2

4

3

5

 Total Pilot Error

58

57

42

44

53

46

50

 Other Human Error

0

8

9

6

8

8

6

 Weather

16

10

13

15

9

9

12

 Mechanical Failure

21

20

23

21

21

28

22

 Sabotage

5

5

11

13

10

9

9

 Other Cause

0

2

2

1

0

1

1

 

The table above is compiled from the PlaneCrashInfo.com accident database and represents 1,300 fatal accidents involving commercial aircraft, world-wide, from 1950 thru 2009 for which a specific cause is known. Aircraft with 10 or less people aboard, military aircraft, private aircraft and helicopters are not included.

"Pilot error (weather related)" represents accidents in which pilot error was the cause but brought about by weather related phenomena. "Pilot error (mechanical related)" represents accidents in which pilot error was the cause but brought about by some type of mechanical failure. "Other human error" includes air traffic controller errors, improper loading of aircraft, fuel contamination and improper maintenance procedures. Sabotage includes explosive devices, shoot downs and hijackings. "Total pilot error" is the total of all three types of pilot error (in yellow). Where there were multiple causes, the most prominent cause was used.
Source: PlaneCrashInfo.com database

If there is a multiple system failure any number of these alerts will be displayed, and the pilot through his:

  • (Boeing) 787 CMCF (Central Maintenance Computing Function) has access to this via his PFD (Primary Flight Display) or the EFB (Electronic Flight Bag). The system is designed to process these faults and perform systems tests. The pilot can also act on the information and take appropriate action.
  • (Airbus) A380 ECAM (Electronic Centralized Aircraft Monitoring) the pilot has access to this system via his PFD (Primary Flight Display) system and ECP (ECAM Control Panel).
  • In addition to these systems on both types of aircraft you have a series of Synoptic pages (systems layout and functioning components) available to the pilot and engineers that provide an expanded view of the condition being alerted. There are pages for every system so as the pilot can choose the system to view.

The crew have to be trained to utilize these display devices, and manufacturers have simplified the systems to ensure that all alerts are displayed prominently to ensure that it captures the attention of the crew.  These displays will be on a MFD (Multi-Function Display) situated centrally between the two pilots, these are the EICAS or ECAM display.

This brings us back to the effectiveness of training programs including the use of synthetic training devices ie: Full motion simulators and Fixed Base simulators. There is no argument that these devices are not only essential but critical to all flight training programs, however the quality of instruction and length of training time is key in these programs.

The importance of engineers to be thoroughly familiar with design and understand the nature of these systems with a greater depth of understanding, than has previously been required has been recognised. To collect data, analyse faults and recognise the true nature of these faults extensive training is required; class room, simulator, CBT (Computer Based Training) associated with PCT (Practical Competency Training), has been developed by training organizations and manufacturers.

The aircraft CMC (Central Maintenance Computers) will display a source of error or fault, however it will not provide details of a secondary cause or effect, the engineer still has to apply his skills and lateral thinking to analyse and diagnose faults that are not identified by the data network ie; short circuits, open circuits or random RF (Radio Frequencies) signals from external sources.

The advent of the Boeing 747-400 also saw the introduction of the “Electronic Logbook” first used extensively by United Airlines and subsequently by other operators. This development saw the demise of the “paper in cockpit” and with ACARS (Aircraft Communication and Reporting System) and SATCOM (Satellite Communication) the aircraft automatically downloads in-flight real time events and data. This also allows engineers on the ground to access and process data as well providing crew in-flight of any requirements or changes that might need to be made. On the ground with the advent of new Wi-Fi technology (Airbus/ Airman and Boeing/AHM) allows staff also to transmit and receive data on the ground.

The Electronic Logbook also allows the defect or check to be written-off and certified by the engineer in the OMT (On-board Maintenance Terminal). This feature also permitted the flight crew to access maintenance data via the FMC and EFB previously recorded and stored. This is of particular importance for flight crew; if a current problem is related to a previous defect, particularly if an MEL (Minimum Equipment List) or CDL (Configuration Deviation List) item was to be applied to the aircraft, affecting operational performance.

An additional device on the Boeing 787 and Airbus A380 aircraft is an independent computer terminal installed in the rear of the cockpit, referred to as the OMT (On-board Maintenance Terminal). This terminal is secured and may only be accessed by authorized engineers (identity and password protected), as it permits changes and or adjustments to the OMS (On-board Maintenance Systems) on the ground. These systems also hold the current manual systems for the aircraft.

The pilots are provided with the EFB (Electronic Flight Bag) that holds all the relevant aircraft and performance data including Flight Manuals required by the crew for operational purposes. As mentioned previously on the Boeing 787 it is also used to access data from CMC (Central Maintenance Computer).

Systems that are independent of aircraft designers have been developed to improve the levels of safety and pilot awareness and are common to all aircraft through mandatory regulations; these includes devices such as:

  • TCAS/TACAS; Traffic (Alert) Collision Avoidance Systems, these are designed to provide information and display on the pilots ND (Navigation Display) screen of traffic around and the vicinity and provide real time positioning of other aircraft in relation to your aircraft.
  • EGPWS; Enhanced Ground Proximity Warning System, providing the pilot with real time information in relation to his relative speed, position and attitude to the ground.
  • S Mode Transponders; provide the identifier to the Control Tower Traffic controller of the identification of the aircraft appearing on his/her screen.
  • WXR (Weather X Band Radar); Enhanced Reactive & Predictive Windshear alerting, and Turbulence Detection providing the pilot with warning of impending windshear or Turbulence conditions.
  • TAWS; Terrain Awareness (Avoidance) and Warning System, providing the pilot with immediate warning on terrain avoidance.

Finally, attempts by manufacturers, designers, human engineering scientists to improve the way we train pilots and engineers have had to accept that Human Nature or more broadly the nature of humans is vast and variable, our personalities, environment, the nature of our employment, family and our basic instincts cannot be defined simply or accommodated, with the advances in technology significant steps in improving the levels of safety can be made…slowly.                                                                                

Peter Marosszeky FRAeS is an Adjunct Senior Lecturer at the Faculty of Science, School of Aviation, UNSW              

comments powered by Disqus

events »