Enhancing Aircraft Safety through Observations and Pilot Verbalisations
Embargoed until: 2019-02-09
MetadataShow full item record
Safety in complex work environments is often an area of focus in the research community. Aviation is a multi-faceted field, but it remains one of the safest work environments in the world. For instance, despite passenger numbers increasing worldwide over many decades, the year 2013, when 3.1 billion passengers flew worldwide, is the safest year on record for aviation (Aviation Safety Network, 2017). That year resulted in only 29 fatal crashes and 265 deaths – one per 11,501,886 miles flown. Despite these impressive statistics, every disaster affects the lives of many people and attracts the attention of the general public, who demand answers. As a result, aviation has continually sought to discover new methods for increasing safety. Early in the twentieth century, when aviation was developing as an industry, accidents were decreasing steadily due to technical improvements, such as advances in aircraft structures and materials, aerodynamics, and systems. In the 1970s, as safety improvements plateaued, the link between human error and aviation accidents became increasingly apparent. This was emphasised in 1977, when 583 people were killed in a collision between two aircraft on a runway in Tenerife. The collision was found to have been the result of poor decision-making and miscommunication by pilots. In the aftermath of the crash, the Royal Dutch Airline KLM (the main protagonist) created the KLM Human Factors Programme. Recognising the part humans play in safety, the purpose of this programme is to improve communication and resource management skills among pilots. In an ongoing effort to improve safety, the aviation industry continues to develop more initiatives based on “human in the cockpit” principles, both in design and in operation. One such strategy is the Line Operations Safety Audit (LOSA). Developed in the 1990s at the University of Texas, the goal of this initiative is to audit pilots’ behaviour during normal flight operations for the purpose of improving air safety. It is a methodology whereby observers sit in aircraft flight decks and unobtrusively monitor pilots’ communications, the threats they face, the errors they make, and how they mitigate these threats and errors. In other words, it is a snapshot of pilot responses to factors impacting safety in normal flights as opposed to a line check by an examiner, where pilots display “angel behaviour” during a flight assessment. Many commercial airlines now use LOSA in their multi-crew operations in order to identify any patterns in behaviour, communication or systems that may need to be changed to improve aviation safety. Whilst commercial aviation generally has a very good safety record, single-pilot operations lag considerably behind, therefore safety in single-pilot operations needs to be further investigated. Past initiatives for improving the safety of single-pilot operations have been less than successful, so new and innovative approaches must be considered. To date, LOSA has not been used in single-pilot operations, and no studies have investigated its applicability in the single-pilot context. This thesis examines the possibility of adapting the LOSA methodology into the single pilot environment, a variation here known as LOSA:SP. The first study conducted for this thesis by the author, implemented the LOSA methodology with pilots working for an emergency services company in Australia. It was clear from the study that the threat and error conceptual framework and observational methodology that are key elements of LOSA were applicable to these single-pilot operations. The results initially supported the view that with minor modifications, the LOSA methodology may be applied to single-pilot operations. In the LOSA:SP study, when pilots were observed to talk aloud during flight, it was referred to as verbalisation. During the study, it became apparent that some pilots used verbalisation as a method for improving their cognitive performance. Further, this verbalisation actually assisted the unobtrusive observers during the LOSA:SP study because it made available to the observers the process of how the pilots’ reached certain decisons. This is significant because two-pilot crews normally discuss their reasoning and decision-making out loud to each other, but this form of verbalisation does not always occur in single-pilot flights. For verbalisation to be viable in the single-pilot environment, limitations must be considered, including safety. A further study was conducted using transcripts and data from a current research project exploring verbalisation in multi-crew operations under simulator conditions. The research aim was to examine the viability of verbalising during different workload phases of flight, such as taxi, take-off, and landing. A method consisting of both concurrent and retrospective verbalisation was used to examine a captain and first officer (second in command) over two simulated flights. The pilots were asked to verbalise during simulated scenarios, and each simulation was audio-visually recorded. Soon afterwards, the pilots were shown these recordings in a separate debriefing room and asked to comment on their performance. Special attention was paid to verbalisations made during three different degrees of workload: low, medium, and high. A total of five overall themes emerged from this study. The most significant were that pilots said that verbalisations during periods of low and medium workload assisted them with their cognitive processes, but verbalisations during periods of high workload were minimal and were perceived by pilots as interfering with their tasks. This second study suggested two findings: that safety could be compromised if pilots were forced to verbalise under periods of high workload and that retrospective verbalisation (or debriefings) were extremely helpful in gaining context supporting thoughts and actions that allowed participants to realise and learn from mistakes and good practices. To determine whether these findings were applicable to SPO, a third study, in the form of a literature review and thematic analysis of LinkedIn data and Focus Group transcripts was undertaken. This study identified that verbalisation is used by many single pilots and flight instructors as a means of enhancing cognitive ability. Yet mulit-crew operators had significantly different views, believing that verbalisation was less useful. Additionally, the literature review uncovered many other applications of verbalisation in other transport and safety conscious industries. The LinkedIn data and Focus Group correspondents also gave opinions on the viability of the LOSA methodology for single pilot operations resulting in a wide variety of discussion. In discussing these results in relation to the application of LOSA in a single-pilot environment, several considerations are suggested in this thesis. One is that the value of verbalisation depends on the situation in which it is used. Another important factor to consider is the familiarity pilots have with verbalisation and how comfortable they are verbalising. Requiring a pilot to verbalise for the purpose of a LOSA:SP may reduce pilot performance in high workload scenarios (with obvious safety consequences). This limits complete transferability of the LOSA methodology to single-pilot operations, where having access to pilot thought processes is fundamental. However, retrospective debriefing may be a useful technique for regaining some of the merits of LOSA:SP as a safety intervention. The willingness of a company to introduce a LOSA seems to suggest an increased awareness of safety, which can only be positive and many respondents found success in adaptations of the methodology to certain routes or conditions. Even without cross-talk between crews, LOSA:SP has sufficient merit for industry to be interested in replicating it.
Thesis (PhD Doctorate)
Doctor of Philosophy (PhD)
School Educ & Professional Studies
The author owns the copyright in this thesis, unless stated otherwise.