Study reveals serious deficiencies in instrument scan by Pilot Monitoring

Introduction

Definition of Pilot Monitoring

Monitoring is an overarching process requiring knowledge, skills and attitudes that enable flight crews to perform safely, effectively and efficiently. Monitoring includes observing and creating a mental model by seeking out available information to compare actual and expected aircraft states (IATA Guidance material on improving Flight Crew monitoring). Monitoring requires a combination of cognitive resource allocation, such as attention, and a link to previously acquired knowledge (scripts and scenarios), which allow a pilot to detect, understand, project into the future, and then take the right decision/action.

Study by BEA on approach and go-round

on pilot monitoring

Towards the end of 2000, the BEA (French Authority for Safety Investigators) initiated a study after they observed several accidents or serious incidents caused by a problem relating to aeroplane state awareness during a go-around.

Read the BEA study: Aeroplane State Awareness during Go-Around

These events seemed to have some common features, such as the startle effect, the phenomenon of excessive preoccupation by at least one crew member, poor communication between crew members, and difficulties in managing the automatic systems.

A full flight simulator with eye-tracking devices was used to carry out the study. Following are the results.

VR training is being rapidly adopted in many high-performance environments (Bird, Citation2019; Gray, Citation2019; Hagita et al., Citation2020; Lele, Citation2013; Mathew et al., Citation2021), but methods of assessing and validating the effectiveness of these environments are lagging behind technology uptake.

Adopting and implementing the recommendations of a recent framework for simulation testing (Harris et al., Citation2020) aimed to 1) test an aircrew training simulation and 2) demonstrate some of the methods for conducting evidence-based testing of VR training. The results indicated that expert pilots generally thought the simulation was realistic and potentially useful for training. The construct validity and psychological fidelity of the simulation was also supported as differences in visual expertise were clearly detectable.

Importantly, the performance assessment criteria were not sufficiently sensitive to real-world expertise, indicating that further work is required to develop this simulation aspect. Future work may adopt some of these approaches in testing other VR training applications.

Result of the Simulator session

Zone of interest

Summary

The flight crew had to prioritise their actions to cope with the task overload. When questioned on this subject during debriefing, the PFs were almost unanimous regarding their priorities: “pitch, thrust or management of the flight path”. The replies from the PNF/PMs were more complex and varied. They referred to the “callouts” to be made and to the various “monitoring tasks” such as “positive (rate of) climb, ensure that the aeroplane is climbing, flight path, pitch attitude and thrust”.The crews that experienced difficulties made adaptations to the procedure.

Some adaptations had positive effects (approach to interception altitude); others led to deviations from the expected result (flight path, for example).Assessment on the role of PMThe performance of the monitoring function is essential but insufficient during the go-around. During performance of the go-around studied, the PM focused on the actions to take and not on their monitoring. It is therefore necessary to focus on this issue during initial training in MCC and then assess the results during ongoing and recurrent training.

Recommendation on role of CRM

The study highlighted the difficulties of maintaining a good level of CRM throughout a go-around. The priorities of the PF and PM are different. Their respective workloads limit their interaction and mutual monitoring of actions. Although fundamental, current CRM alone cannot constitute a reliable safety barrier for disruptive elements. In general, whatever the type of recent accident, investigative findings often point to shortcomings in CRM.

mindFly analysis on human factor

It was observed that the instrument scan sequence is no longer taught in the training as in the non-EFIS aircrafts. The “T” scan sequence was integral to the training and critical for instrument flying. With modern EFIS aircraft, the ‘T” scan is lost due to overreliance on automation and low arousal effect. The pilot monitoring scan has deteriorated even further thereby eroding an important barrier.The OEM and training departments need to review and advise the pilots on then new and effective means of instrument scan for various phases of flight.