Aviation as a system of systems: Preface to the special issue of human factors in aviation

Aviation is a system of systems. Maier (1998) characterised a ‘system of systems’ as possessing five basic traits: operational independence of elements; managerial independence of elements; evolutionary development; possessing emergent behaviour; having a geographical distribution of elements. In the context of aviation, these systems have distinct operational independence (aircraft operations; maintenance; air traffic management/control) and each of these aspects has managerial independence (they are offered by independent companies or national providers); however, they are bound by a set of common operating principles and international regulations for design and operation. All aspects of aviation encompass technical, human and organisational aspects. It is a socio-technical ‘system of systems’ encompassing critical human factors considerations such as usability, training, design, maintenance, safety, procedures, communications, workload and automation.

It is fair to say, though, that the aviation ‘system of systems’ was never designed, it is a legacy system that has evolved over the past century. All the components in aviation are themselves open systems (i.e. they must interact with their environment). Open Systems Theory is derived from General Systems Theory (von Berthalanfry 1956); however, these organisations are only selectively open, in that they interact with their environment but also need boundaries in order to exist. For example, civil airlines operate into a wide range of airports (none of which they own), aircraft maintenance is often provided by third parties, aircraft ramp servicing is almost invariably provided by a range of external suppliers and air traffic management/air traffic control (ATC) is provided by the air traffic service providers from the countries into which they either operate or overfly. In the operation of civil aircraft, there are a great number of inter- and intra-organisational boundaries that information and resources must cross in this system of systems.

Leveson (2002, 2004) has proposed that the nature by which systems of systems remain in a dynamic equilibrium is via the control and communication of constraints. In Systems-Theoretical Accident Model and Processes, accidents are considered to result from inadequate control or enforcement of safety-related constraints (occurring during the design, development or operation of the system) and not from individual or component failures. Safety is a product of control structures embedded in an adaptive socio-technical system. Accidents are viewed as control failures.

The papers in this special issue of Ergonomics address all aspects of the aviation system. Some adopt a macro-ergonomics approach investigating system-wide issues in the safe operation of aircraft. Other papers are much more focused in their aims and objectives, addressing a very specific issue in human performance. However, all papers can be viewed from a wider, socio-technical system perspective.

Macro ergonomic papers addressing systemic issues

Working in an air traffic management context, Vogt et al. (Human factors in safety and business management) adopt an organisationally based approach, arguing that through greater business awareness, ergonomists can make inroads both into reducing cost and simultaneously increasing safety. These improvements in performance are a result of adopting a careful measurement and monitoring programme, which integrates human factors into a wider, ‘balanced scorecard’ approach for organisational performance as a whole. Increases in performance are often a result of ‘soft’ factors, such as reduced absenteeism through increased employee well-being, rather than ‘harder’, procedural or equipment-based interventions. They propose an approach by which such less tangible human factors can be measured and their impact assessed. In contrast, Leva et al. (The advancement of a new human factors report – ‘The Unique Report’ – Facilitating flight crew auditing of performance/operations, as part of an airline's safety management system), this time working with a commercial airline operator, adopt a different approach to the previous soft, organisationally based safety management perspective. In this case, emphasis is firmly based on improving organisational safety performance through the collection, collation and distribution of safety-related information in a consistent and accessible format. In common with the previous paper, though, is the objective to obtain auditable (measurable) performance data.

Walker et al. (From ethnography to the EAST method: a tractable approach for representing distributed cognition in air traffic control) also adopt a macro-ergonomic approach to the study of ATC in their contribution. However, in this case the emphasis is focused upon the slightly more tangible command and control aspects of the system. Nevertheless, it is argued that the wider ATC system has an emergent property, in the form of distributed situation awareness (DSA) that is not apparent through the study of its individual components: any complex distributed socio-system such as ATC needs to be studied from several viewpoints. It is proposed that event analysis for systemic teamwork (EAST) provides a methodology for doing just this. The application of this method is illustrated using a case study. In another paper in this special issue, Griffin et al. (Investigating accident causation through information network modelling) demonstrate how the EAST methodology can also be used in a retrospective manner to analyse air accidents, in this case a reappraisal of the 1989 Kegworth accident. Again, the emphasis is placed upon understanding the network of information nodes and systemic failures across both human and non-human aspects of the wider system, which resulted in a lack of DSA, rather than in identifying particular, individual failings.

Micro-ergonomic contributions

The theme of DSA is also implicit in the contribution from Grote et al. (Adaptive coordination and heedfulness make better cockpit crews), who examined aspects of crew resource management in the performance of higher and lower performing crews. In this case, they examined ‘heedfulness’ (adaptive levels and types of coordination with regard to workload) on the flight deck. Implicit coordination based upon a shared mental model of the flight situation was as important as explicit coordination in producing good performance.

Tiewtrakul and Fletcher (Are aviation English language proficiency standards enough to address the problems caused by regional accents? A study of errors in ATC–pilot communications) investigated the effects of accented English on the comprehension of air traffic instructions. All ATC communication globally is undertaken using ‘aviation English’; however, comprehension of these instructions can be severely compromised (both on the flight deck and on the ground) by the accents of the speakers. This is not a trivial problem when it is considered that voice communications are the principal way in which safety constraints are communicated between components throughout the air traffic system. DSA cannot be achieved if communication is impaired in such a way.

Kallus et al. (The taskload-efficiency-safety-buffer triangle (TEST) – Development and validation with air traffic management) describe the development and validation of a methodology to measure and visualise the balance between task demands, safety and efficiency in air traffic controllers. The methodology can be applied on an individual basis or on a wider basis. From an overall system-wide perspective, three generic, antagonistic parameters can be applied to evaluate system functioning: safety; performance; cost. Aviation authorities are concerned solely with safety aspects of aircraft design, pilot training and airline operations. However, commercial organisations, be they air traffic providers, airlines or maintenance organisations, are required to balance the requirement for safety against both cost and performance considerations. Until relatively recently, the aviation human factors discipline has concentrated almost exclusively on the safety aspects of system functioning, but this paper, along with several others in this special issue, begins to examine the contribution of ergonomics in a wider, business context, where it is acknowledged and accepted that safety is not the only goal (although it should be the primary goal) and other factors have to be considered – see also the papers by Vogt et al. and Ward et al.

The maintenance system is an often overlooked component in aviation human factors but is crucial to maintaining safety. Rankin (1997) reported that improper maintenance contributed to 15% of all commercial jet aircraft accidents. Based upon Boeing data, Marx (1998) estimated that solely in the US, 48,000 non-airworthy flights per year were dispatched as a result of a maintenance error. Ward et al. (A performance improvement case study in aircraft maintenance and its implications for hazard identification) again demonstrate how efficiency gains can be made, while at the same time improving safety performance. As in several previous papers, the emphasis is again in integrating human factors considerations into a wider socio-technical system model to identify blockers to good maintenance performance and practice. Blockers to performance were recorded, categorised and monitored. What is more, to resolve these issues, the workforce was empowered to produce local solutions to many of the problems.

Few airliners are now routinely flown manually; however, pilots need to maintain their manual flying skills proficiency to pass their regular ‘base checks’. This has resulted in a number of accidents resulting from pilots ‘hand flying’ the aircraft on routine flights to practise for their simulator check rides (and getting it wrong)! Ebbatson et al. (The relationship between manual handling performance and recent flying experience in air transport pilots) describe the development of a sensitive methodology, based upon control input measures in the frequency domain, which can distinguish between pilots who have recent manual flying experience and those who have not. It is suggested that such measures are well suited to the continual surveillance of pilot performance as part of a flight data monitoring programme allowing for early intervention. In a further related study of human skilled performance in micro-gravity and high ‘g’ environment, Mierau et al. (Exaggerated force production in altered Gz-levels during parabolic flight: the role of computational resources allocation) found that inferior motor performance was not necessarily related to inadequate cognitive resources, as had been previously suggested.

Should the untoward actually happen, in the rare event of a crash it is comforting to know that the vast majority of aircraft accidents are survivable. However, in the event of an accident when it comes to evacuating the aircraft the control of many aspects of the situation now revert to the control of the passengers in the aircraft, especially those near the emergency exits. Wilson and Muir (The effect of overwing hatch placement on evacuation from smaller transport aircraft) describe the detrimental effects of the incorrect placement of the exit on evacuation rates in a commuter aircraft. This is one of the few occasions when a vital aspect of control resides externally to a component directly involved in the running of the aviation system.

Sport or recreational aviation is often overlooked in many treatments of human factors in aviation, especially gliding, which is particularly under-researched. This special issue concludes with a paper by Jarvis and Harris (Development of a bespoke human factors taxonomy for gliding accident analysis and its revelations about highly inexperienced UK glider pilots) on this topic. It is argued that gliding requires a unique accident categorisation system: adopting or adapting existing taxonomies cannot accommodate many of the unique aspects of this form of sport aviation. Once a bespoke taxonomy is developed and applied retrospectively to a sample of UK gliding accidents, it is argued that several existing assumptions about accident causation in gliders cannot be empirically supported and there is a case to change several aspects of the instructional process, particularly in the light of the analysis of accidents occurring to low hours pilots.

Conclusions

Human factors as a discipline has come of age. It is still a relatively new discipline with its naissance in the 1940s in the aviation domain commencing with the work undertaken in the UK and North America during and shortly after WWII (see Chapanis 1999). For the majority of this time, the discipline has essentially been building its applied science base, drawing heavily from experimental and social psychology, engineering and aerospace medicine, and providing bespoke solutions to localised problems. However, as a result it has also tended to become a somewhat fragmented discipline. While increasing levels of specialisation served to develop the science, it has also often mitigated against its coherent application. The macro-ergonomic papers in this special issue that look at the system-wide issues in the aviation industry clearly demonstrate that there is both a great deal of independence and interdependence between aspects of the aviation system and that these relationships need to be properly understood. Human factors in aviation must also avoid its natural inclination to rush and claim the moral high ground by marking its territory solely within the realm of safety. The communication of constraints between system components is vital for safe operations but system-wide studies can also identify opportunities for performance and efficiency gains. These studies of systemic function provide a coherent framework into which the smaller, more focused studies may be placed.

The first 60 years of the study of human factors in the aviation domain has produced significant safety benefits. The challenge for the next 60 years is to continue to improve safety while at the same time also striving to use ergonomic principles simultaneously to enhance organisational efficiency and performance.