Abstract
This paper adopts a socio-technical approach to aviation in order to examine how Human Factors (HF) influence airline operational safety and efficiency. The SHELL model is deployed to analyze the Tenerife airport disaster in order to define and present the main HF that caused it. Aviation is a complex system so the causes of an accident should be investigated within its all interconnected elements and interfaces. Tenerife collision was a starting point for the development of HF measures and approaches. This paper presents the main features of TEM, SMS, LOSA and CRM that evolved following the tragedy to enhance safety and efficiency.
Introduction
Human Factors garnered enthusiasm from scientists and aviation experts following the increased number of accidents that occurred in the 1970s. The industry of commercial aviation reached the conclusion that the majority of them were caused by human failure rather than technical malfunctions (Wise, Hopkin and Garland, 2009). Human Factors (HF) science is a multidisciplinary field that was established to define and explain which personal errors in interaction with other elements lead to adverse consequences for aviation safety and efficiency. Further investigation of the Tenerife airport disaster showed that contributory factors came from the environment, personal skillsets, mental conditions and communication issues, rather than one serious mistake of a particular individual. Different actors made some minor mistakes that eventually resulted in a horrible tragedy, which proves the socio-technical perspective of the aircraft system. The famous theoretical model, SHELL, designed to shed light on HF’s negative influence on the airlines’ operation, will be used further to analyze the accident. HF measures that emerged and were implemented following the infamous Tenerife collision also will be presented to show how lessons were learned in order to prevent further accidents.
HF cannot be limited to the professional skills of the individuals involved. It is rather related to their relationships with other people, machines, and the environment. Aviation should be seen as a “system of systems” where different elements have their operational and managerial independence (Dogan, Pilfold and Henshaw, 2011). Manufacture and design of aircraft have improved following the last decades that decreased technical reasons for accidents, while the number of civil aviation accidents due to human factors increased significantly (Han and Wang, 2016). The SHELL model that was initially presented by Edwards in 1972 mainly shows how HF interacts with the aviation environment (Hawkins, 2017). It is a theoretical framework that encompasses socio-technical elements of the aviation system. This combination of social and technical aspects influences the behavior and performance of individuals in the workplace.
The SHELL model will be further applied to analyze the Tenerife accident. In 1977, the deadliest runway collision took place on the island of Tenerife, killing a total of 583 people (Cookson, 2009). The two B-747 charter flights, KLM 4805 and Pan Am 1736 were diverted to Los Rodeos Airport due to a bomb explosion in Las Palmas Airport they were initially bound for. When it was reopened, the KLM failed to lift off because of the collision of planes that occurred. HF, such as insufficient decisions and communication failure, became the main cause of this accident. However, many other contributory factors led to the tragedy.
The L-E interface presents and explores a social and natural environment. According to Bruggink (2000), visibility was inferior and continued to deteriorate during the time Flight 4805 was utilized for refueling. Its captain had been working as an instructor at Flight Training Department for the six previous years, whereas the first officer (FO) received the B-747 qualification check from him two months before the accident (Bruggink, 2000). The KLM crew suffered from an autocratic leadership style under which the captain made all decisions. The FO and flight engineer were intimidated by his position and mitigated speech. This poor leadership led to miscommunication between the flight crew.
For instance, when the engineer wanted to clarify if Clipper 1736 left the runway, the captain told him in the affirmative that it did. Communication and coordination issues usually occur when dictatorial and experienced captains work in pairs with junior co-pilots. The steep cockpit authority gradient should be addressed by CRM training to see that captains are receptive to FO inputs, and FOs are assertive if they spot danger in the captain’s action.
The next interface is L-H what explores technology and equipment issues. The airport itself was a small regional one that is typically not capable of handling international air traffic. Both crews were not familiar with the airport, taxiway exits did not have markings, and ground radar was not available. The Clipper 1736 was instructed to exit the runway with the help of the third taxiway before the accident.
However, the size of the pavement needed for Boeing to make two 145-degree turns in order to reach C-3 was less than the plane required. The distance from the tower to the third exit was more than 500 meters, so the controller reasonably could not concentrate on the visual channel of communication (Weick, 1990). The planes also were forced to use the runway for taxiing instead of taxiways, which were occupied by other diverted planes. There were some technical problems with radio communication, as some near-simultaneous transmissions caused electromagnetic interference making it impossible for the KLM crew to hear the Pan Am reporting that they continue taxiing down the runway. The L-S interface comprises HF related to system procedures, computer programs, standard operating procedures (SOP) and checklist layout. The main cause of the collision was the decision of the KLM captain to take off without receiving proper clearance from the control tower. The “Stand by for takeoff” procedure initiated by ATC was also neglected or misunderstood by him.
Moreover, the Dutch flight crews had just started to operate under new strict regulations that define duty time. As a result, the KLM crew strived not to exceed the duty time limits to avoid possible loss of licenses and fines. At the time of the accident, they had approximately 9 hours of duty time, while Pan Am was already on duty for 11 hours. Controllers were also working since 10 AM, so fatigue and stress seemingly contributed to the accident. Workload pressure may explain why parties made crucial mistakes in communication, while KLM company’s pressure on their crew to take off explains the imprudence of their captain. Cognitive overload is a situation when an individual is mentally tired and does not able to process properly more data because of the lack of working memory (Woods, Patterson and Roth, 2002). The ATC controllers obviously were suffering from it, which decreased their performance and control efficiency.
The interface L-L examines the interactions needed for teamwork in the workplace. Radio communication issue that became the main reason for the tragedy is one of them. The final five minutes were decisive when miscommunication occurred between pilots of both flights and the control tower. Voice recorder transmissions revealed that all parties faced particular challenges in communicating with each other despite using English. The ATC staff had a heavy Spanish accent that allegedly contributed to the misunderstanding. The primary communication failure occurred when the ATC issued a clearance to the KLM, which actually was not a takeoff clearance. The crew incorrectly interpreted it because the message confusingly included the verb “take off.” The FO then repeated the clearance to the tower and stated that “we are now at takeoff” (Cookson, 2009, p. 10). It is where the cultural differences played an important role, as the FO allegedly meant “we are now at the process of taking off.” The Dutch language allows using the infinitive form of the verb in order to point at the action which is currently performed.
In that case, the FO’s first language (L1) intervened the communication, whereas Spanish speaking controller just added a missing word “position” to the last sentence according to the context available to him. He believed that the KLM crew is still waiting for the proper clearance and added, “Ok. Stand by for takeoff, I will call you” (Bruggink, 2000, p. 3). Only the word “Ok” was clearly heard by Flight 4805 because of transmission interference. This situation can serve as an example of expectation bias. It is a strong belief that a particular outcome is going to occur in accordance with previous experiences and inappropriate assessment (Salas and Maurino, 2010). The KLM captain was too confident that the control tower issued takeoff clearance, taking into consideration his recent experience in simulation flights.
The ATC, in their turn, expected that the radio transmission is working well, the KLM flight would wait a while until Pan Am accomplished the safe exit. As a result, both parties failed to question the received information. Moreover, the lack of situational awareness (SA) of ATC controllers was clearly visible in them being unfamiliar with precise aircraft positions and not being able to predict future collisions. Expectation bias usually negatively influences SA and reduces the chance of proper future event anticipation. The fact that controllers instructed the Pan Am to use the C-3 intersection instead of the C-4, which required an easier maneuver, proves they had the wrong mental picture of the whole situation in a stressful time.
The Tenerife runway collision encouraged international aviation organizations to improve HF measures to prevent further tragedies. For instance, new flight rostering systems proved to increase both efficiency and safety and decrease fatigue, stress, anxiety with the help of a new workload approach (Harris, 2006). Communication was standardized to avoid similar situations involving ambiguous language. Aviation English courses became compulsory for non-native speakers during their flight training, whereas ICAO established language Proficiency Requirements (LPR). The KLM crew spoke both Dutch and English while ATC used Spanish and English. The LPR reemphasized the importance of one and standardized language for the reduction of communication errors.
Moreover, the Tenerife airport disaster, together with the United 173 crash over Portland in 1978, forced the NTSB to design a special training that ultimately emerged as Crew Resource Management (CRM). The accidents which occurred in the 1970s pointed at the fact that the inability of the crew to respond to stressful situations rather than aircraft handling skills is the leading cause of disasters. The first CRM training following the NASA seminar was conducted by United Airlines (Dahlström, Laursen and Bergström, 2008). In the beginning, it was an experimental measure that was met with resistance because its content challenged the commander’s authority. Nevertheless, now it is an essential and mandatory part of the training, helping to enhance flight safety.
CRM courses very often take place in conjunction with MCC, command course, or basic training. Moreover, recurrent training also requires CRM courses to be conducted. In general, CRM deals with the cognitive and interpersonal skills needed for successful operations within an aviation system. It improves cognitive skills that help professionals to solve problems, make appropriate decisions and maintain situational awareness. A new generation of CRM addresses strategies for managing errors and deals with the effects of stressors as work overload, emergencies and fatigue (Helmreich, Merritt and Wilhelm, 1999). It works as a countermeasure to performance degradation and personal vulnerability.
Line Operations Safety Audit (LOSA) is a safety management tool designed to gather information on and manage everyday operations’ errors. LOSA methodology requires observing normal multi-crew operations to define flight crew performance (Earl et al., 2012). The collected data by trained observers helps to assess pilot strategies for managing errors, undesirable states and threats. Observations showed that some deliberate errors concerning procedures are not dangerous, whereas communication with ATC is a more rare error but brings more severe consequences (Earl et al., 2013). ICAO implemented LOSA as the main tool to develop countermeasures to human errors.
LOSA is generally based on the Threat and Error Management (TEM) concept. It is a broad safety theory that focuses on human performance in aviation operations (Maurino, 2005). Its main field of consideration is the interaction between operational context and people. The TEM model can be applied as an analysis tool for a single event and also identify systemic features within a broad set of interactions (operational audits). The main three components of TEM are threats, errors and undesired aircraft states (Maurino, 2005). Threats and errors are perceived as “normal” everyday issues that should be managed by the flight crew while undesired aircraft states usually come from their inability to manage it in the early stages. For instance, the contaminated runway is an environmental threat; failure to cross-verify automation inputs is a procedural error; proceeding towards the wrong taxiway is an example of an undesired aircraft state.
The ultimate purpose of TEM is to propose countermeasures to decrease its three components’ adverse effects on flight operations safety. The main system-based countermeasures applied by flight crews are Standard Operation Procedures (SOPs), Checklists, Briefings, Ground Proximity Warning System (GPWS) and Airborne Collision Avoidance System (ACAS) (Maurino, 2005). NextGen program is an example of modern countermeasures that aimed to accomplish the shift from a ground-based system of air traffic control to one that is based on satellite technology (Salas and Maurino, 2010). Safety Management System (SMS) is another approach that controls safety risks in operations and identifies areas of safety improvements. Risks and quality management methods help to establish safety policy, conduct safety planning and implementation, safety assurance that provides hints for further safety improvement within the organization.
To conclude, the Tenerife tragedy sheds light on the wide range of HF that adversely affects the complex socio-technical aviation system. The SHELL model helped to explain different levels of interfaces between individual and environment, software, hardware and other people in the workplace. Such HF as poor communication, work and cognitive overload, poor leadership, the influence of various stressors and lack of situational awareness all contributed to the accident. The technology development concerning aircraft and cockpit designs currently stresses the importance of HF. As a consequence of the tragedy, HF measures like CRM, TEM, LOSA and SMS emerged to increase the performance and safety of aircraft operations considering both flight crews and ATC.
Reference List
Bruggink, G. M. (2000). ‘Remembering Tenerife’, Air Line Pilot, 69(7), pp. 18-23.
Cookson, S. (2009) ‘Zagreb and Tenerife: airline accidents involving linguistic factors’, Australian Review of Applied Linguistics, 32(3), pp. 22-1.
Dahlström, N., Laursen, J. and Bergström, J. (2008) Crew resource management, threat and error management, and assessment of CRM skills.
Dogan, H., Pilfold, S.A. and Henshaw, M. (2011) ‘The role of human factors in addressing systems of systems complexity’, 2011 IEEE International Conference on Systems, Man, and Cybernetics, pp. 1244-1249.
Earl, L. et al. (2012) ‘Developing a single-pilot line operations safety audit’, Aviation Psychology and Applied Human Factors, 2(1), pp. 1-14.
Han, X. and Wang, L. (2016) ‘San cai human factors analysis model of civil aviation maintenance’, Advances in Engineering Research, 93, pp. 288-292.
Harris, D. (2006) ‘The influence of human factors on operational efficiency’, Aircraft Engineering and Aerospace Technology, 78(1), pp. 20-25.
Hawkins, F. H. (2017) Human factors in flight. 2nd ed. London: Routledge.
Helmreich, R.L., Merritt, A.C. and Wilhelm, J.A. (1999) ‘The evolution of crew resource management training in commercial aviation’, The International Journal of Aviation Psychology, 9(1), pp.19-32.
Maurino, D. (2005) Threat and error management (TEM).
Salas, E. and Maurino, D. (2010) Human factors in aviation. 2nd ed. London: Academic Press.
Weick, K.E. (1990) ‘The vulnerable system: an analysis of the Tenerife air disaster’, Journal of Management, 16(3), pp.571-593.
Wise, J. A., Hopkin D. V. and Garland D. J. (2009) Handbook of aviation human factors. 2nd ed. Boca Raton: CRC Press.
Woods, D., Patterson, E. and Roth, E. (2002) ‘Can we ever escape from data overload? A cognitive systems diagnosis’, Cognition Tech Work, 4, pp. 22–36.