Chapter One – Introduction
The aim of this project is to implement ‘Crew Resource Management’ (CRM) for aerospace OEMs, argue the benefits, why/how to legislate and how to adapt CRM. This chapter discusses ‘Human Factors’ (HF) and CRM, to get a broad overview of what these are. Furthermore, the original project objectives of the proposal are shown followed by a discussion of ‘Problem Formulation’, involving a discussion of what this project tries to tackle. Next, the cultural limitations of the research are stated. Furthermore, an introduction to the companies participating in the empirical research is included, and a brief summary of the author’s background is shown. The first section involves an overview of the project and the structure of this paper.
1.1 Overview and Structure of the Paper
The four objectives of the project proposal are shown in Section 3 (below). The first objective is considered in Chapter 2, and Chapter 3 includes objectives two and three. The fourth objective is considered in Chapter 4. The chosen method used for the secondary research (literature review) is discussed at the beginning of Chapter 2, and is applicable to all four objectives/chapters. The empirical research method section is located before the empirical research findings section in Chapter 2. Chapter 5 includes the conclusions and recommendations. Appendices are located at the end of the paper.
1.2 Human Factors and Crew Resource Management
The science and the definition of HF may not be widely known and there may be ambiguities about what it is. ICAO (1998) advises of a persistent misconception, that HF is a branch of medicine. HF is based on several scientific disciplines including psychology and engineering according to Harris (2011). There are several different definitions of what HF refers to; one of the definitions is from the Human Factor Society (2015):’ … a multidisciplinary science that includes research from the fields of psychology, biology, sociology and engineering.’ Furthermore, Widdowson and Carr (2002) describe HF as: ‘… a professional discipline concerned with improving the integration of human issues into the analysis, design, development, implementation, and the operational use of work systems’. Furthermore, HF and Ergonomics (ERG) have aspects which are similar, if not identical; therefore, the next chapter will offer a brief discussion about HF and ERG in context.
Crew Resource Management (CRM) is an HF method originally designed for pilots using available resources to prevent incidents/accidents. Lauber (1984) describes CRM as: ‘… using all the available resources — information, equipment, and people — to achieve safe and efficient flight operations.’ CRM has evolved over time, and more aspects of HF are being used in CRM, not just the use of available resources; however, the name CRM has been kept.
1.3 Project Objectives
This project has four objectives which are listed below (Wunderlin, 2017b):
(1) To research HFM and CRM. This will involve secondary research (literature) of current HFM used in aerospace OEMs and other domains in addition to secondary research of CRM in different domains (e.g. air traffic control, aviation maintenance, flight operations and space development/operations). Furthermore, primary research with an OEM in the form of interviews and/or observations to compare and verify the secondary research will be undertaken. Remark: primary research of CRM is not intended as the author has CRM experience. Information gained will be presented in the form of a report with relevant aspects.
(2) To research the potential benefits of implementing CRM in OEMs: financial, safety and efficiency benefits in order to make a business case.
(3) To examine how CRM training may be legislated for, e.g. in EASA Part 21 for Design Organisations Approval (DOA). CRM practitioners (e.g. pilots) have recurrent training which is necessary in order for them to retain their, unlike an OEM employee who may have an academic degree which does not require renewal (a degree does not expire, unlike most pilot licenses).
(4) To explore how CRM could be adapted for aerospace OEMs; to design and propose a broad training outline, and perhaps a new HF approach.
1.4 Problem Formulation
This section aims to give an understanding of the reason for the project and identify what it is trying to tackle. Chapter 3 (benefits) supplements this section, showing evidence of ‘why’ this problem should be addressed.
Major recent developments in aeroplanes have been delayed considerably, that includes the development of the Boeing 787, Airbus 350 and the Bombardier CSeries airliners (Airbus A220) (see Figure 1 below).
Several different reasons for the delays have been identified. One notable issue was the involvement of many suppliers, not only for the production, but also in the design process for the aeroplanes. For example, for the CSeries Aircraft, the total number of suppliers exceeded 90 (Bombardier, 2009).
Figure 1: Recent innovative aeroplane developments compared
(Source: Wunderlin, 2016a – adapted from: Airbus, 2016; Boeing, 2016; and Read, 2016)
Furthermore, difficulties in the development of innovative technologies also contributed to delays. For example, the CSeries aircraft has new materials (Bombardier, 2016), new engines (Read, 2016) and a new fly-by-wire system according to Dewar (cited in Warwick, 2013). These very new technologies have, to some extend, contributed to the delays (Hemmendinger, 2016; Bullis, 2013; Dewar, cited in Warwick, 2013), not only in terms of the development of the aircraft, but also new certification methods which have had to be adapted/developed (Zhang et al., 2011 and TCCA, 2015). Therefore, this project proposes CRM for aerospace OEMs to reduce the delays, by increasing efficiency.
Figure 2: Simplified aircraft lifecycle in terms of HF and CRM
(Adapted from: Cross, 2009; CAA, 2016; ICAO, 1998; Koopman, 1999; Sandom and Harvey, 2009)
Efficiency may be achieved as CRM methods successfully improve safety in airline operations. These safety improvements through CRM may be transferred to OEM employees in the form of safety and efficiency improvements. Research, later in this paper, shows the link between safety and efficiency.
CRM and Human Factors Methods (HFM) are used in the lifecycle of an aircraft to some extent. Figure 2 shows the lifecycle of an aircraft in terms of HF and CRM; this figure is a simplified illustration. Nevertheless, there are also some HFM used in the design process, though they may not be considered to be thoroughly applied throughout the process, and are not applied by all people involved. The same applies to CRM which is used by some involved in the flight operations and servicing, e.g. pilots and maintenance technicians. This is supported by the CAA (2016), which argues that the application of CRM in flight operations is inconsistent.
Furthermore, CRM and HFM in the lifecycle of an aircraft are also regulated to some extent. More details about these regulations are presented in Chapter 3 (Legislation). Moreover, the difference between HF and CRM in this lifecycle, is that HF applications are deemed relevant for design purposes, whereas CRM applications are used in operations/maintenance, focusing on the affected person who uses CRM (see Chapter 2 for more). Additionally, the justification for using HFMs in aerospace design are vague in general. This is, perhaps, because the ‘added-value’ may not be visible and it is difficult to present an argument which convinces people to invest in HF according to Harris (2007).
This concludes the problem formulation. The proposed introduction of CRM for all aerospace OEM employees, may help improve the efficiency of OEMs as is discussed in Chapter 3 (Benefits).
1.5 Applicable Countries and Cultural Limitations of the Research
The researcher of this project is based in Swizerland, Europe, working through a university in the United Kingdom. The empirical research was conducted in two European countries, Germany and Switzerland. Furthermore, most of the literature used for the research is from Europe and North America. It is important to acknowledge that the application of the ‘Western’ approach to CRM may not be as successful in other regions of the world, e.g. east Asia, according to Harris (2011), as such regions have a high power-distance culture, unlike Western countries, . In Western countries, people are less inhibited about speaking up, and challenging other people. Therefore, this project is considered to be culturally valid for Western countries, such as the United States, Canada, EU member countries, the United Kingdom, Norway and Switzerland. Additionally, for simplicity, most of the legislation and regulations discussed in this paper are limited to Europe. Some regulations from the United Nations aviation unit, ICAO, are also shown.
1.6 Participating Companies for the Empirical Research
Four companies participated in the empirical research. The author is very grateful for the opportunity to carry out research in these companies. It has been helpful, in terms of getting aspects about HF/CRM used in the real world.
The results of the empirical research are presented as a whole, not distinguishing between companies. More information about anonymity is given in the empirical research methods section in the next chapter. One German and three Swiss companies were used for the empirical research. Below is an introduction to the companies, to give the reader an overview of the companies involved in the research:
The German company, Comco-Ikarus, is an OEM producing ultralight aircraft; it has world-wide sales and a long successful history in this market (Comco-Ikarus, 2018).
RUAG Aviation is a Swiss company and is part of the RUAG Holding. The latter consists of five different divisions including aviation, aero structures, defense, space and ammotec. RUAG Aviation is an OEM and owns the type certificate (TC) of the Dornier 228 aircraft. Furthermore, RUAG also develops/produces aircraft parts e.g. for Airbus; therefore, it is also a supplier. Additionally, the company also services (maintenance) both, military and civil airplanes (RUAG, 2018).
The second Swiss company, Universal Dynamics, produces CNC parts for aerospace companies and is, therefore, a supplier. Additionally, the company has recently developed a UAV (Universal Dynamics, 2018a and 2018b).
The third Swiss company, Aventura AG, is a startup, currently developing an innovative gyrocopter (Aventura AG, 2018).
1.7 Author’s Background
The author has worked for more than eighteen years as an airline pilot, in European and world-wide operations, and has instructed new pilots aspiring to become airline pilots. Furthermore, the author has completed an apprenticeship as a design engineer in an aerospace company, and worked thereafter for other engineering firms, before beginning a career as an airline pilot. The author has worked part-time in order to allow time for study. A university diploma in design and innovation was completed prior to the air safety management studies.
This concludes Chapter 1. Chapter 2, which follows, involves the research, discussions and arguments about HF and CRM with the aim of proposing how to adapt CRM for aerospace OEM employees.
Chapter Two – Objective 1: Human Factors and CRM Research
The intention of this chapter is to give an outline of what Human Factors (HF) and Crew Resource Management (CRM) are. This chapter also represents the basis for the next three chapters; the findings of this chapter will be used and researched further to allow for further exploration of the other objectives of this study. Furthermore, the empirical research methods used to conduct the primary research are discussed followed by the findings themselves. And finally, the last section of this chapter is the discussion.
2.1 Source and Criteria used for the Literature Review
This section states the methods used for the literature review, which are applied in all four chapters, for each objective.
Criteria used for Searches
Many different sources were used in the research, in order to achieve a thorough picture of the subject, and to find relevant information. Recent books and academic journals were searched (not older than 10 years), and older literature was used to supplement this evidence when deemed necessary.
The internet was used as an initial source to research the subject. The City University online library and Google Scholar were the two main search tools used. That led to text books and academic journals which were further researched.
The majority of textbooks were acquired from Amazon and Google Play; additionally, the City University library was used to read some textbooks online.
Many journals were researched; these were accessed from the City University library. Additionally, some journals were found on the internet.
Legislations and Regulations
ICAO and EASA legislations and regulations were researched. Some of the information was found through internet searches, from the respective websites of the regulatory authorities and information was also found in text books and academic journals.
2.2 HF and CRM Literature Review
The literature review in this section is used not only for the preparation of the empirical research, but also to explain HF/CRM, and to get an insight into its application in other industries/domains. Furthermore, HF will be explained using the SHELL model, followed by a discussion about CRM. Also, ‘Human Error’ and automation/men-machine will be discussed. Additionally, Human Factors Engineering is also discussed followed by ‘just-culture’. Initially, however, HF, Human Performance and Ergonomics will be considered.
2.2.1 Human Factors, Human Performance and Ergonomics
There are many different definitions of what HF are; some of the definitions are mentioned above in the introduction chapter. The integration of the human into systems with the aim of improvement, may be used as a brief description of the term. Eurocontrol summarised (Figure 3) the interpretations of HF to include the interaction of men-machine, organisation/staffing, training/development, procedures/roles/responsibilities, teams/communication and the recovery from failures. Furthermore, human limitations and capabilities are studied in HF, with the intention to design systems with a minimised mismatch between the requirements and capabilities of the human.
Figure 3: Interpretation of the field of Human Factors
(Source: Eurocontrol, cited in Dahlström et al., 2008)
By doing so, Human Performance (HP) is not only improved, but optimised according to Rodrigues et al. (2012). Additionally, HF also include the relationship and interaction between humans according to Hawkins (2016). That is one important aspect, as the original concept of CRM is to use all available resources, e.g. the communication between pilots, thus maximising the benefits of the interaction aspect. Moreover, HP describes how well humans’ carry out assigned tasks, and measures how quickly and how accurately these are performed. HP may be split into seven different categories: physical, physiological, psychological, psychosocial, hardware, task and environmental factors (Rodrigues et al., 2012).
Table 1: Educational backgrounds, NASA-Ames
(Source: NASA, 1992, cited in Hawkins 2016)
HP and HF may be thoroughly discussed using these seven categories; this paper will, however, use a simpler approach; the SHELL model will be used not only to explain and discuss HF, but also as a reference to illustrate the context for the other objectives of this project.
Moreover, psychology may be considered the predominant science in HF. That is, as many HF practitioners have a psychology background and most of the HF literature is authored by psychologists according to Hawkins (2016), Table 1 shows the backgrounds of HF practitioners.
Moreover, HF and Ergonomics (ERG) are closely related or may even be almost identical sciences. Hawkins (2016), however, argues that HF have a wider meaning than ERG; the aspects of HP and system interfaces are not generally considered in ERG, unlike in HF. Furthermore, the origins of ERG predominately take physiological aspects into account, whereas psychology is the basis for HF, according to Noyes (2009). HF developed in the United States, and ERG emerged in Europe initially (Sandom and Harvey, 2009). The expression HF will be used in this paper, to include aspects which some literature may refer to as ERG. For the direct associations of the human-hardware integration, however, the term ERG will be used, as ERG may be understood by the general public and the empirical research involves related questions.
2.2.2 Human Factors; the SHELL Model
Figure 4: Components of the SHELL model
(Source: Wunderlin, 2017a – adapted from: ICAO 2013a; Hawkins 2016)
The SHELL model was chosen to explain HF in more depth, as it shows the different interactions/influences, of humans in a simplistic way. The SHELL model emerged in the early 1970s and was originally called SHEL (Edwards, 1972); the second ‘L’ was added later to illustrate the interaction between people. The model has the following different components: Lifeware, Sofware, Hardware and Environment. Figure 4 shows the SHELL model according to Hawkins (2016) including explanations of the interfaces between the components.
Table 2: Components of the SHELL model
(Adapted from: ICAO 2013a; Hawkins 2016)
A simple explanation of the SHELL model in practical terms is as follows: the human at the centre of the SHELL model may be a pilot. A pilot flies an aeroplane and is using procedures (software) to operate the airplane (hardware) and low air density (environment) may be a performance-influencing factor. The pilot will communicate and interact with other humans (lifeware) e.g. cabin crew and air traffic controllers. The individual components are listed in Table 2, and explanations and/or examples are shown to get a better understanding of the different components.
to explain in greater depth, the Lifeware component in the middle of the model is the human, and may be called the hub; it is the most flexible and valuable component according to Hawkins (2016). The other components need to be matched or adapted to the central human component. The connections between the components are not straight or simple; a careful match is necessary to avoid suboptimal performance and prevent stress in the system according to Abbott (2001).
Therefore, good products/services will be designed to make the hub (the human) work as effectively as possible. This effective design of components is also called Human Centered Design (HCD) and will be discussed later.
The four different connections between the components are discussed in more detail below, according to Abbott (2001), Rodrigues et al. (2012), ICAO (2013a) and Hawkins (2016):
The L-L interface, is the interaction between people and is concerned with teamwork, inter-personal cooperation, leadership, and personality. The L-L interface was the main focus of CRM when it first appeared, ‘using all available resources’ in the form of other people in the means of communications. Additionally, when people work in groups, their performance and behaviour can be affected by group influences, which is addressed with CRM training.
The interface between lifeware and software includes aspects which are non-physical. Computer software is an obvious one, as the name suggests. Furthermore, the layout of checklists and manuals are other software components. Symbology is also considered software; this can be seen, for example, at airports, where different signs with symbols help passengers find their way. Symbols are also shown on instrument screens in the cockpit illustrating flight information. Software problems are often more difficult to solve compared to L-H problems and, as such, are less tangible.
The interface between lifeware and hardware, commonly refers to, and is most concerned with, the expression man-machine. A good adaptation of hardware to the human is also sometimes called good ERG. Nevertheless, all four interfaces (not only L-H) are addressed in HF and ERG (see Section ‘2.2.1 Human Factors, Human Performance and Ergonomics’). The design of a seat is one example of the L-H interface. A good seat design will prevent the Human from having an uncomfortable experience, and in some cases even prevent injury in the case of an accident or spinal issues from long-term usage.
Environmental aspects can have an influence on HP. Notably, the air (pressure, moisture levels, temperature, …) and the noise can adversely influence the human. Measures to reduce/avoid these effects for people on board aeroplanes have been designed and implemented. That includes pressurised cabins in aircraft and noise cancelling headsets. Furthemorer, another aspect which has gained importance is the disturbance of biological rhythms such as sleep disturbances due to air travel and crossing time zones.
2.2.3 ‘Human Error’
‘Some believe that they need to keep beating the «human» until the «human error» goes away.’ (Dekker, 2014)
There are several different definitions of what ‘Human Error’ (‘HE’) is, and the expression is even considered inappropriate in some literature. Above is a quote from Sidney Dekker, which sums up in a few words, the misconception of how ‘HE’ may be removed. Additionally, Dekker (2012) suggests that ‘HE’ may be considered the symptom and not the cause of mishaps. Nevertheless, McFarland and McRandal (2016) describe ‘HE’ as: ‘… defined as a human action with unintended consequences’, or: ‘Any action, performed by a person, which exceeds a system’s tolerance’. Furthermore, ICAO (2013a) argues that contributions to ‘HE’ involve:'”… a mismatch between the liveware and the other four components …’, referring to the SHELL model.
Moreover, ‘HE’ may be considered an expression which is misleading and wrong. If determined to be more critical, a particular ‘HE’ may have a connection to an organisation, meaning the error happened because of decisions within the organisation, resulting in the operator taking actions leading to the particular ‘HE’. That is supported by Dekker (2014), who argues that more complex stories about organisations are behind stories of ‘HE’. Dekker (2014) also states: ‘… errors are symptoms of trouble deeper inside a system’. Furthermore, Hawkins (2016) argues, in the context of the SHELL model, that badly designed equipment (hardware) and/or badly designed procedures (software) may induce ‘HE’. Improvements in hardware/software may be helpful in such cases, as the ‘HE’ may be repeated by other humans. Furthermore, Hawkins (2016) also argues that accidents arising due to ‘HE’ may lack appropriate HFM in the working environment. Therefore, HF and CRM may be considered an ‘instrument’ for preventing ‘HE’. The preferred management of error will be discussed below (Just-Culture).
Nevertheless, the expression ‘HE’ will be used in this paper in spite of the debate about terminology outlined above. As much of the HF and CRM literature uses the expression ‘HE’ and it may be considered helpful for educational purposes, it will be used in a simplistic way. The next sub-chapter will give some examples.
2.2.4 CRM and Threat/Error Management (TEM): Pilots
This sub-section illustrates Crew Resource Management (CRM) used by pilots; it also addresses Threat and Error Management (TEM). Additionally, CRM is also used by other professions and will be discussed below.
CRM Definition and History
CRM originally emerged in the 1970s, and its intention was to prevent incidents and accidents in airlines, caused by pilots (Rodrigues et al., 2012). Similarly to HF, CRM doe snot have one clear definition. The definition from Lauber (1984) is given in the introductory chapter: ‘… using all the available resources — information, equipment, and people — to achieve safe and efficient flight operations.’ The original name of CRM was ‘Cockpit Resource Management’ and it was introduced by NASA; the name was eventually changed to ‘Crew Resource Management’, as other crew members, outside the cockpit, were also included, according to Helmreich and Foushee (2010).
CRM has undergone evolution according to Salas et al. (2001). The focus of the first evolution was on the individual styles and behaviours, focusing on psychological testing. Group dynamics in the cockpit was the focus of the second evolution. The broadening of the scope was central to the third evolution, when factors outside the cockpit were taken into account, training CRM for cabin crew and maintenance technicians. The fourth evolution integrated CRM with technical training. Concepts such as Advanced Qualification Program (AQP) and Line Orientated Flight Training (LOFT) which led to proceduralisation and integration of CRM were included. The fifth evolution has acknowledged the inevitable existence of ‘HE’, and managing ‘HE’ is its predominant focus (Flin et al., 2008).
HF may be considered a better expression of the respective CRM training and methods, because of the evolution of CRM. While the original intention of CRM was ‘to use all available resources’ referring mainly to the ‘L-L’ interface in the SHELL model, HF looks at all the components in the SHELL model. Nevertheless, the name CRM has stayed and is used in several domains and industries, referring to the HFMs of operating personnel like pilots and air traffic controllers. The CAA (2016) supports the idea that CRM has a wider scope nowadays: ‘… Although retaining the title «CRM», such training now covers a much wider scope that is often referred to under the umbrella terms of «non- technical skills» and «human factors».’
Threat and Error Management
Threat and Error Management (TEM), similarly to CRM, is conceived for pilots. It can be described as safety management activities oriented for the cockpit, according to Harris (2011). In practical terms, TEM may be defined as predicting/avoiding operational threats and errors by thinking ahead (CAA, 2016). The fifth generation of CRM acknowledges the inevitable existence of ‘HE’. As mentioned above, TEM is concerned with the management of error in general. Furthermore, all pilot training events should cover TEM according to EASA-FCL (CAA, 2016). In this paper, the expression CRM will combine CRM and TEM, for simplicity.
CRM Applications, Methods and Models
This short section illustrates some examples of the practical use of CRM. Firstly, inevitable ‘HE’ will be addressed in the form of the ‘error-troika’ in connection with TEM and Situational Awareness (SA). Additionally, ‘ANC’ and the Swiss cheese model will be shown.
‘Error Troika’: three different levels of managing ‘HE’ are suggested in TEM: avoiding the error, trapping the error and mitigating the consequences of errors (CAA, 2016), as illustrated in Figure 5 below.
SA is used by pilots to prevent and handle errors. The CAA (2016) describes SA as ‘knowing what is going on’; additionally, Weiner (2016) suggests that the same method is also used by healthcare practitioners. Tools like SOP checklists, planning, drills and more are used in practical terms.
Figure 5: The error troika
(Adapted from: Helmreich et al., 1999b)
For example, planning a flight, e.g. in connection with weather, may be considered thinking ahead (avoid). Using the weather radar of an aircraft in order to detect and avoid CBs may be considered understanding (trapping). Noticing entry into a turbulent area, turning on the fasten seatbelt sign and changing the aeroplane’s lateral and/or vertical flight path may be considered noticing (mitigating), according to Dahlström et al. (2008). Figure 33 below illustrates the connection between SA and TEM. Figure 33 appearing here is a copy of the original in Chapter 4 (‘Objective Four: Adaptation of CRM for OEMs’).
Figure 33: SA in context with TEM
(Adapted from: CAA, 2016; Dahlströhm et al., 2008; Helmreich et al., 1999a; Weiner, 2016)
Furthermore, ‘ANC’ is a CRM method, which stands for ‘Aviate Navigate Communicate’. ANC is a very simple but an important tool used by pilots, as it contributes to flight safety in a effective way. The application is a prioritisation tool: flying the aircraft is most important, next is navigation and third is any communication, that includes with air traffic control, cabin crew and passengers (CAA, 2016). As an example, a functional Lockheed Tristar crashed in the 1970s in the Everglades, United States. It only had an inoperative light bulb, which indicates the position of the landing gear. The flight crew tried to solve that minor problem; however, while doing so, nobody was flying the plane and it crashed (NTSB, 1973).
Moreover, several models are used in CRM training and one of those is the ‘Swiss cheese model’ (see Figure 6, below). According to Reason (2000), the system and the person approach can be used to determine error causation. The system approach looks deeper into the system including processes from organisations.
Figure 6: Swiss-cheese model
(Source: Wunderlin, 2016b — adapted from Reason, 1990)
In contrast, the person approach blames humans for ‘HE’ . Nevertheless, the systems approach sees errors as consequences from deeper inside the system and considers that multiple causes contribute to error. That is reflected in the Swiss-cheese model (Figure 6) and shown as an analogy, with objects flying through cheese slices (barriers). Holes move and change in size, and a mishap happens when all the layers are aligned (Reason, 1990). Moreover, protective layers can be added for error prevention, e.g. a third or fourth engine of an aircraft would give more back up, in the face of potential engine failures. Nevertheless, Dekker (2014) appreciates the simplicity of the model and suggests that finding contributors to accidents and categorising them is a strength of the model. However, Dekker (2014) also states shortcomings of the model: a ‘linear’ perception of barrier models may be dangerous; it is difficult to predict accidents; explanation of the organisational, bureaucratic and social context is difficult; systems are complex – just adding barriers may not work.
CRM Training Content
It is not practical to state a full pilot CRM program in this paper, as that would take too long. Therefore, the previous sub-section illustrates the practical application of CRM with a few examples. Additionally, Table 3 is added to give an overview of what both Europe (JAA) and the United States (FAA) recommend for pilot CRM training. It is worth mentioning that although the wording is different, both regions have a similar approach to pilot CRM training. Additionally, the table also illustrates the wider scope of CRM and more HF aspects are addressed than in the original CRM concept, e.g. fatigue, automation and stress reduction.
Table 3: JAA and FAA CRM curricula recommendations for flight crew
(Source: Flin et al. 2008)
Acceptance of CRM and Motivation
The acceptance of CRM and motivation may be considered crucial factors. Hawkins (2016) even argues that motivation of the lifeware component is the most significant characteristic, referring to the SHELL model. Additionally, according to ICAO (1998) advice, flight safety is affected significantly by behaviours and attitudes. Furthermore, motivation may be a reflection of what a person actually does and could do (ICAO, 1998). CRM training acceptance can be linked directly to better crew performance and attitude changes (Chidester et al., 1991). Additionally, a resistance to CRM training existed when it first emerged and sometimes the expression ‘charm schools’ was used by attendees (Flin, 2010).
More literature is available on this subject, confirming the issue of acceptance and required motivation; in addition, Section ‘3.1 Benefits’ will supplement this argument.
2.2.5 HF and CRM in Context
HF and CRM have been discussed in the previous sub-sections to give an understanding of the context and application of the two subjects. CRM is considered to be the application of HF; additionally, it was stated above that CRM has evolved and has a wider usage compared to HF than the original intention. The name, however, stayed, and it may be considered to be HF as used by pilots. This is supported by Dahlström et al. (2008) who argue that CRM is an applied form of HF. Pilots are operational staff, following procedures, and they judge how tasks can be accomplished successfully as such, and when necessary, they deviate from procedures in order to reach the intended and safe outcome (Harris 2011; Hawkins 2016; Rodrigues et al., 2012). The project proposal (Wunderlin, 2017b) states that an introduction of CRM for aerospace OEMs require an adaptation of CRM; a transfer of existing CRM training is not sufficient according to Flin et al. (2008). Aerospace companies have many different jobs and some may be considered ‘operational’ jobs, some may not. An aerospace design engineer’s job may have some ‘creative’ aspects; therefore, certain CRM methods used by pilots may be little or no help for design engineers. There are, however, certain HF aspects which are valid for many jobs and domains.
2.2.5 Automation; Men and Machine
‘Getting technology to replace unreliable people is an attractive idea, and is wide-spread. But technology introduces new problems as well as new capacities.’ (Dekker, 2014)
Humans make mistakes, and when they happen, often the expression ‘HE’ is used. One tempting approach to prevent ‘HE’ is to remove the ‘human’, and replace it with a machine; it is, unfortunately, not that simple. The quote at the beginning of this sub-chapter is shown, as it illustrates in a few words the challenge of automation and the men/machine problematic/opportunity. Additionally, Dekkers’ quote is supported by Abbott (2001), who argues that new technology often introduces other problems while solving some. This sub-section is a discussion about automation/men-machine in anticipation of the aspects concerning CRM and HF.
One approach to address this issue is called the ‘compensatory principle’. This approach is also called ‘MABA-MABA’ in some literature, meaning Men Are Better At – Machine Are Better At (Dekker and Woods, 2002). Reduction of ‘HE’ is done by determining ‘MABA-MABA’ and allocating tasks to men or machines respectively, according to the strengths, guided by the compensatory principle (Sandom and Harvey, 2009).
Table 4: The principles of the Fitts list
(Source: Sandom and Harvey, 2009)
The compensatory principle has been addressed and used for a long time, and a considerable amount of knowledge about the characteristics of man-machine is available. The application of the compensatory principle may be performed using lists comparing men/machine characteristics. One list was published as early as the early 1950s by Fitts (1951) (see Table 4). Nevertheless, there is also opposition to the compensatory principle.
Dekker and Woods (2002) question the compensatory principle based lists, as the false assumption is made that machines/computers and people have fixed weaknesses and strengths. Furthermore, Dekker and Woods (2002) argue that for human-automation, ‘get along together’ may be the preferred approach rather than refining/re-inventing compensatory based methods.
Moreover, there are also other principles which address the men-machine problematic/opportunity; two additional principles are the ‘left-over’ and the ‘complementary’ principles. Firstly, the left-over principle lets the machine do as much work as is feasible; the rest will be done by the human according to Hollnagel (2009). The complementary principle looks for guidelines to compliment the human, instead of replacing the human with machines according to Grote et al. (1995). Table 5 is illustrated to show a comparison between the three discussed men-machine principles.
Table 5: Comparison of the three automation philosophies
(Source: Hollnagel, 2009)
Furthermore, automation and the interaction between men and machine may be a considerable part of CRM and HF. Two of the four interfaces of the SHELL model may be considered men-machine interfaces to some extent (Lifeware-Hardware and Lifeware-Software). Flight safety can be affected in an adverse way and accidents may happen when the automation/men-machine interaction breaks down. As an example, an aircraft crashed and all 264 occupants were killed in 1994, because of inputs given to the autopilot by the pilots. The aircraft stalled because of a combination of high engine thrust, out-of-trim condition and too great a flab retraction (Abbott, 2001). The interactions between man and machine need to be addressed to prevent such accidents. Human limitations and capabilities need to be considered in the design process (HFE) and humans need to be trained (CRM).
2.2.6 Usage of CRM in other Industries/Domains
CRM was introduced originally for pilots, and over time, more industries and domains have adopted CRM. Fields where safety is deemed a high priority use CRM nowadays, such as rail and sea transportation, also nuclear/chemical industries and medicine (Dahlström et al., 2008). Furthermore, other airline employees, e.g. cabin crew, were eventually also CRM trained; additionally, maintenance technicians and air traffic controllers are now trained in CRM as well (Rodrigues et al., 2008).
2.2.7 Human Factors Engineering: Methods in Design
A discussion about Human Factors Engineering (HFE) will be presented in this sub-section. Aerospace companies use HFE for the design of their vehicles. There is, unlike CRM for pilots, not much standardisation, nor are there many regulations. Chapter 3 will state some regulations currently in place for aerospace OEMs.
HFE, like other previously mentioned HF disciplines, addresses aspects concerning humans, and HFE used in design may neither have a simple nor clear definition. Nevertheless, Cardosi and Murphy (1995) consider engineering psychology to be a close cousin of HFE. Furhermore, Cardosi and Murphy (1995) define HFE as: ‘… the discipline that applies knowledge of human capabilities and limitations to the design of technological systems.’ Additionally, Harris (2007) describes HFE as how performance can be optimised, in the system of men-machine, by integrating the human into the system. In addition, Human Centered Design (HCD), also known as ‘User Centered Design (UCD)’, is where the design process places the end-user at the centre. HCD may be considered another name for HFE according to Harris (2007).
HFM are used in HFE; nevertheless, Stanton et al. (2013) argue that HFMs are sometimes only used by the very person who designed them, and a limited use in general may apply. Furthermore, the process for designing, e.g. aircraft cockpits, may be considered nonstandard, complex, largely unwritten and variable according to NASA (1995b). That may be, as several disciplines are required in the process, not only HFE (Abbott, 2001).
2.2.8 Punishment for Mistakes; Just-Culture
CRM training itself may not be sufficient for the successful implementation of CRM; a supportive company structure may also be required. Such a supportive structure may be achieved by implementing a ‘just-culture’. Reducing and preventing mistakes has been discussed in several sections above. It was stated that ‘HE’ may be considered inevitable, and methods used to avoid, trap and mitigate consequences are being used, e.g. by pilots.
Not many human actions that result in mishaps are deliberate. Additionally, judging in hindsight and drawing conclusions is easy according to Dekker (2014). Figure 7 below, the tunnel model, shows the lead up to and implications of a mishap. Pilots come to work to do a good job and their decision-making is influenced by several factors, when safety critical events happen. Furthermore, a continuous pace and ‘being inside the tunnel’ adds further difficulties.
Therefore, a simple judgment in hindsight is considered inappropriate according to Dekker (2014). Additionally, Dekker states: ‘To understand error, take the view from inside of the tunnel and stop saying what people failed to do or should have done” and “Don’t ask who is responsible, ask what is responsible!’
Nevertheless, sometimes deliberate actions do lead to mishaps, and not all actions are tolerated in a just-culture. An intended action which leads to an intended consequence may be considered sabotage; however, an intended
Figure 7: The tunnel model
(Adapted from: Dekker, 2014)
action which leads to an unintended consequence may be tolerated (Marx, 2014).
Culpability models, such as those shown in Figure 8 below, are used in support of the management in just cultures. A line must be drawn to distinguish between acceptable and not acceptable actions according to Reason (1997). Additionally, Dekker (2012) advises that who draws the line is more important than where to draw the line.
Moreover, an effective reporting environment is a key aspect of a just-culture. That is: learning from mistakes is practiced in a just-culture and that is only possible if people do the reporting. In a just-culture, reporting is encouraged as there is no retribution in most cases (Dekker, 2014).
Figure 8: Just-culture culpability model
(Adapted from: Baines Simmons, 2008)