Waterviolier 15, 4823MA Breda, The Netherlands info@uprt.aero

The LOC-I Problem

The LOC-I Problem

On 1 June 2009, an Airbus A330-200 being operated by Air France on a scheduled passenger flight from Rio de Janeiro to Paris CDG as A447 with a crew of three pilots and in the cruise at FL 350 at night and in Instrument Meteorological Conditions (IMC) became overdue en route and when wreckage was found five days later on the sea surface on track from the last known position, it became clear that aircraft had crashed into the sea with the loss of the aircraft and all 228 occupants. (source: www.skybrary.aero)

Later, official investigation by the French Bureau d'Enquêtes et d'Analyses (BEA) found that there had been a temporary inconsistency between the airspeed measurements, likely following the obstruction of the Pitot probes by ice crystals that, in particular, caused autopilot to disconnect and reconfigure to alternate law. 

This malfunction caused a series of events which led to the stall and loss-of-control of the aircraft. According the investigation report, inappropriate control inputs destabilized the flight path. There was a lack of any link by the crew between the loss of indicated speeds called out and the appropriate procedure. The PNF did not identify the deviation from the flight path in a timely manner and the PF applied insufficient correction to return to a stabilized flight path.

Things got worse as the crew did not identify the approach to stall. Their lack of immediate response resulted in the aircraft exiting the flight envelope: the aircraft stalled. The crew failed to diagnose the stall situation and consequently apply the inputs that would have made it possible to recover from it.

The thorough investigation has clearly yet painfully exposed the cause of the crash. With the knowledge today and in hindsight, it is easy to judge the situation and determine that the loss-of-control could have been avoided. Yet, we have to understand the human factors that played a role in the critical situation, as well as the background and training received by the flight crew. The effect of startle response should also not be underestimated. 

The accident is considered to be one of the major Loss-of-Control In Flight (LOC-I) events. There are many more, including the crash of Colgan Air 3407 and Turkish 1951. Like Air France 447, both events occurred in 2009.

The events led to an increasing concern and discussion on the growing LOC-I problem. In the past decades, and as of today, LOC-I remains the number one cause of lethal accidents in commercial aviation.

Boeing CAST 2017

Source: Boeing Statistical Summary of Commercial Jet Airplane Accidents - Worldwide Operations | 1959-2016


In June 2009 (coincidentally the day after the crash of AF447), a conference on LOC-I was held at the Royal Aeronautical Society (RAeS) in London, UK. Following the conference, the RAeS launched an international industry/government/academia consortium called ICATEE; the International Committee for Aviation Training in Extended Envelopes. The committee was chaired by IDT's president Dr. Sunjoo Advani.

ICATEE was supported through participation from the aircraft industry, simulator manufacturers, training providers, regulators, researchers and airline customers. Participants included Boeing, Airbus, CAE, Opinicus, ETC, APS Emergency Maneuver Training, the FAA, NTSB, NASA, NLR, DLR, ALPA, IFALPA, KLM Flight Training and many others. In total, there were approximately 75 participants.

ICATEE introduced the concept of comprehensive Upset Prevention and Recovery Training. Three levels of training are requires: Awareness, Avoidance and Recognition, and, finally, Recovery.

Training Challenges

The main challenge is to provide realistic training that can be retained by the flight crew. Response to upset conditions demand immediate and correct response by the flight crew, and sometimes this reaction may be counter-intuitive. For instance, the airplane’s response to a stall may be worsened by applying power or continuing to try to maintain altitude (as often prescribed by pilot examination criteria). Training this knowledge, it is said, requires both an academic knowledge, as well as developing the ability to manage the aircraft state through the correct execution of skill-based behaviour. Part of this can be trained in the classroom, and part could be trained in the flight simulator. However, if their response is based on inadequate or incomplete data, simulators may provide a negative training environment.

The Startle Factor

Re-creating the startle factor in flight simulators, in other words the impact of such events that cause a pilot to react in a primal, self-defending manner, is also a significant challenge. In a high-stress situation, a pilot may call upon basic skills more than cognitive and adaptive thinking to resolve the situation, and training these skills is considered essential in preventing LOC-I.

Training in a Realistic Environment

Also, it is difficult to provide the abruptness of accelerations or sustained forces in flight simulators. Alternatives include in-flight training on aerobatic-capable aircraft, or the use of continuous-g simulation devices. Clearly, all of these need to be carefully considered in order that the skill sets are properly developed, and the training remains cost-effective and relevant.

Graduated Approach

ICATEE prepared a strategy that involves a graduated approach to introducing these training requirements. First of all, the current Airplane Upset Recovery Training Aid (AURTA) document was revised. This landmark report, published several years ago through an industry working group, has been of benefit to many airline pilots who choose to read it. It does not, however, cover aircraft below 100 passengers, and is limited to swept-wing jets. Additionally, it is not mandated.

AUPRTA rev 2 was released early 2017 and is forming the basis of IDT's UPRT application Skypointer.

Secondly, ICATEE provided recommendations for enhancing and making better use of current-technology full flight simulators. Enhancements will include data, instructor station feedback and motion cueing.

New tools using modern media will also be recommended, to enhance the knowledge-based skill-sets of pilots.

Finally, the use of aerobatic-capable aircraft will be discussed and considerations given to this type of training as well. Through its liaison with Supra, ICATEE will also investigate the relevance of continuous-g training platforms.


In September 2010, ICATEE provided its first inputs to the FAA/Industry Stall/Stick-Pusher Working Group, recommending how to deal with Startle in simulation, and how to make better use of the instructor station in flight simulator training.

ICATEE’s main output has been through the International Committee for Flight Simulation Training Device Qualification (ICFQ) to ICAO. These recommendations were integrated into ICAO Manual 10011.

The urgency of solving the LOC-I problem is clear, and IDT is proud to make this contribution to airline safety.