A study on human errors of air traffic controllers based on HFACS-AHP

Abstract

Safety is the lifeline of civil aviation. With the advancement of technology, human factors have become the primary factor affecting aviation safety. Among these, human errors in air traffic control (ATC) account for a significant proportion of aviation accidents. In order to investigate human errors in ATC, this paper introduces two common human factors conceptual models – the Reason model and the HFACS model, and analyzes the specific application of the HFACS model in aviation. It is found that the HFACS model can effectively establish a classification system for human errors in ATC. Moreover, this paper combines the HFACS model with AHP (Analytic Hierarchy Process) and uses the HFACS model to establish a first and second level indicator system for human errors in ATC. We calculated the weights of each specific factor using the Analytic Hierarchy Process. The results show that the most significant influencing factor in the first level indicators is organizational factors, while the most significant influencing factor in the second level indicators is Air Traffic Resource Management. Among the 13 second level indicators, the smallest weight is the controller’s violation. Based on the calculated weight results, in order to improve the safety level of ATC, the first step should be to improve the level of ATC organizational management. This is also of great significance for improving the safety level of the entire aviation system. The findings of this study suggest that the HFACS model, combined with AHP, can be an effective tool for identifying and analyzing human factors in ATC and ultimately improving the safety of the aviation industry.

Keywords

Human errors air traffic controllers HFACS AHP aviation safety

1. Introduction

Safety has always been the lifeline of civil aviation. Only by ensuring safety can high efficiency and high returns be achieved. Thanks to the continuous efforts of the aviation industry, the safety level of civil aviation has been greatly improved. From a global perspective, since 1967, the accident rate per million flight hours has been between 1.5 to 3. After entering the 21st century, the accident rate has dropped to about 1 per million flight hours [1]. These achievements are due to the advancement of technology, the development of aviation technology, and the application of new technologies and materials. At the same time, the safety, reliability, and economy of aircraft themselves, operational environments, and software and hardware have also been continuously improved.

Currently, aviation accidents caused by hardware facilities of airplanes have already dropped to a very low level. On the contrary, accidents caused by human factors are showing an upward trend [2]. According to relevant statistics from the International Civil Aviation Organization, approximately 70% to 80% of aviation accidents are directly related to human errors. These errors include not only those of pilots, but also those of personnel such as air traffic controllers, mechanics, and dispatchers who are closely related to flight operations. Air traffic controllers bear great responsibility in communication between aircraft and the ground, and any error in the control process can lead to tragic consequences. The Tenerife airport disaster is a typical accident caused by the error of an air traffic controller and is the most serious aviation accident to date, second only to 9/11. Therefore, in addition to the study of human errors in pilots, the study of human errors in air traffic controllers is also a current research hotspot in aviation.

In order to prevent and mitigate the impact of human errors in ATC, a substantial body of literature has emerged, exploring various factors that contribute to human errors, including cognitive workload, fatigue, and decision-making, among others. Foster et al. believed that the human factor approach is very important for aviation safety management, and a aviation safety management measure had been proposed using Analytic Hierarchy Process and Human Factor Analysis and Classification System, which can address risks in air traffic management in the UK [3]. Miriam et al. onducted a case study on a control center located in a tower in the Philippines to investigate the relationship between fatigue, workload, situational awareness, and control strategies [4]. They found a negative correlation between fatigue and workload of controllers, and significant negative impacts on situational awareness. Abbas et al. proposed a quantitative method for analyzing human errors in air traffic control through TRACEr and CARA, which was used to analyze the human errors of airport tower controllers [5]. They evaluated the probability and specific causes of human errors and found that the probability of errors was directly related to emergency management. Sepideh et al. analyzed the personality traits of 37 air traffic controllers using the Vienna Test System and found that personality traits had a significant impact on controllers’ cognitive levels and decision-making abilities [6]. There were significant differences in personality traits between controllers who had made errors and those who had not.

Currently, there are many studies on human errors in ATC from various perspectives, but most of them focus on a specific aspect or cause of human errors. There is a relative lack of systematic research on human errors in ATC. At the same time, systematic research on the human factors in air traffic control mostly consists of qualitative analysis. Many studies have provided clear classifications and conclusions regarding the factors influencing human errors in air traffic control, but these conclusions are mainly based on theoretical analysis and lack specific data support, indicating a limited amount of quantitative research. Therefore, conducting quantitative analysis of human errors in air traffic control using quantitative methods can deepen the study of the influencing factors and lead to more reliable conclusions.

2. Reason model and HFACS model

In order to systematically study the human errors problem in aviation, many theoretical or conceptual models have been developed, among which the Reason model and HFACS model are widely used.

In 1990, Professor James Reason of the University of Manchester in the UK proposed the famous human factor conceptual model – the Reason model in his famous psychology book “human errors” [7]. The model has gained widespread attention since its inception and is currently utilized in various fields such as human factors engineering, medicine, nuclear industry, aviation, among others. It has also become one of the theoretical models for aviation accident investigation and analysis by the International Civil Aviation Organization, as shown in Fig. 1.

Figure 1.

Reason model.

The Reason model divides the sources of problems and defects in the aviation system into four levels: organizational influences, inadequate supervision, preconditions for unsafe acts, and unsafe acts. The model believes that accidents are caused by system failures, which can be divided into two types: explicit failures and implicit failures. There are many defects or deficiencies in various elements of the aviation system (like the holes in Swiss cheese), some of which may result from poor organizational management, inadequate supervision, and insufficient operator preparation. These defects and deficiencies do not immediately have negative effects on the system and belong to implicit failures. human errors and violations belong to unsafe acts, which may cause explicit failures and directly lead to accidents. However, the Reason model does not provide detailed coding for the defects and deficiencies at the four levels, so although the model is widely used, there are still many inconveniences in specific applications.

Due to the shortcomings of the Reason model in specific applications, Shappell and Wiegman proposed the Human Factors Analysis and Classification System (HFACS) model based on the Reason model, as shown in Fig. 2 [8]. The model was originally developed as an accident investigation and data analysis tool for the US Navy and later began to be applied in the civil aviation field. It has now become the main tool for investigating and analyzing civil aviation accidents. HFACS codes the unsafe acts and their inducing factors at each level of the Reason model in detail, and describes the specific failure contents of the four levels. HFACS was first applied to research on pilot human errors and has been widely used in other aviation fields such as aviation maintenance (HFACS-ME) and air traffic control (HFACS-ATC) after years of research and practice, with good practical effects [9].

Figure 2.

HFACS model.

3. Specific applications of HFACS in aviation

The Reason model, including its derivative HFACS model, is an organizational-oriented model that can effectively analyze the performance, problems, and deficiencies of organizational operations. Therefore, many studies mainly use the Reason model, HFACS model, or their reconstructions. Among them, there are relatively more studies combining the HFACS model with other methods, with Bayesian networks being a common combination. For instance, Meng et al. collected 74 global CFIT accident investigation reports and used the Human Factors Analysis and Classification System (HFACS) and Bayesian networks (BN) to explore the causality and inherent correlation of CFIT accidents. They found that unsafe behavior is the biggest factor in controlled flight into terrain accidents [10]. Lyu et al. introduced the HFACS-BN model based on the HFACS model and used this model to analyze 142 aviation accidents related to ATC that occurred worldwide from 1980 to 2019. They found that unsafe behavior had the greatest impact on ATC performance and provided data support for the management of aviation safety accidents [11]. Additionally, some studies have combined HFACS with other methods. Lower et al. combined the System Theoretical Incident Model and Processes with HFACS and proposed a framework that combined STAMP and HFACS. This framework can analyze the relationship between human, technological equipment, and the environment, and improve air traffic control safety [12].

As the HFACS model is a primary means of conducting aviation accident investigation and analysis, many studies focus on this type of problem. A review study by Kaptan et al. showed that the HFACS model is widely used in various fields of accident investigation and is receiving increasing attention. Subsequently, the application of HFACS in marine accident analysis was emphasized, demonstrating the importance of HFACS [13]. Sha et al. proposed a improved HFACS model for bus accidents based on official reports of bus accidents. The model divides the causes of accidents into seven levels, which can better analyze bus accidents [14]. Ma et al. used the HFACS model to statistically analyze 109 general aviation accidents that occurred in China from 1996 to 2021. They found that management and organizational factors are the main causes of general aviation accidents, and there is a significant correlation between the flaws in the organization level and the unsafe behavior of pilots. They identified the key organizational problems, including resource management, organizational processes, failure to correct known problems, insufficient supervision, and regulatory violations [15].

From a specific application perspective, the HFACS model is a very good method for analyzing human errors. Currently, the model is relatively more applied in the study of pilot human errors, while the study of human errors in ATC is less applied. This paper proposes a combination of HFACS and AHP for analyzing human errors in ATC. By using the HFACS model to classify human errors in ATC in detail, and combining it with the Analytic Hierarchy Process, it is possible to quantitatively analyze the weight values of each ATC human errors and judge the degree of impact on air traffic control safety. Effective suggestions for reducing human errors in ATC can be proposed based on this analysis.

4. HFACS and AHP combined analysis of human errors in ATC

4.1 Human errors classification based on the HFACS model in ATC

The HFACS model provides a detailed coding of the organizational influence, unsafe supervision, preconditions for unsafe acts, and unsafe acts in the REASON model. Based on the coding content of the HFACS model and the specific situation of human errors in ATC, a classification of human errors in ATC based on the HFACS model was analyzed, as shown in Table 1.

Table 1
HFACS human errors classification in ATC

Hierarchy	Human factors	Performance
Organizational influence	Air Traffic Resource Management	Human Resources, financial resources, equipment and facilities
	Organizational atmosphere	Organizational structure, organizational culture
	Organizational procedures	Operational management, procedures
Inadequate	Inadequate supervision	Lack of reasonable training, loss of supervisory situational awareness
supervision	Improper operating	Improper team composition, overloaded work, lack of sufficient rest time for
	plans	the team
	Uncorrected problems	Unscheduled maintenance, failure to correct inappropriate behavior
	Non-compliance with	Unreasonable authorization of controller work, violation of procedures, failure
	supervision	to follow rules and regulations
Preconditions	Environmental factors	Physical or technical environment that causes human errors or unsafe events
for unsafe acts	Controller status	Poor mental state	Work pressure outside of work, mental fatigue, lack of concentration
		Poor physical condition	Disease, physical fatigue, physiological disorders
		Limitations of body/mind	Inadequate experience in dealing with complex situations, lack of intelligence and ability, visual limitations
	Personnel factors	Team resource management	Lack of team cooperation, inappropriate shift handover
		Personal preparedness	Inadequate training, self-medication, drinking alcohol, failure to comply with rest time requirements
Unsafe acts	Errors	Skill-based errors	Omission of procedural steps, inadequate skills, improper allocation of attention, distraction
		Decision-making errors	Exceeding the scope of ability, improper handling of emergency situations, improper procedures, inappropriate motivation
		Cognitive errors	Wrong judgment/perception errors, situational diagnostic errors, failure to recognize situations
	Violations	Habitual violations	Violation of instructions, rules and standard operating procedures, failure to follow normal procedures for shift handover
		Occasional violations	Failure to timely release information notification

4.2 Analytic hierarchy process

The Analytic Hierarchy Process (AHP) is a systematic approach that divides a multi-objective comprehensive problem into multiple indicators or criteria, and then divides each indicator (or criterion, constraint) into multiple levels. AHP calculates and analyzes the weight values of each indicator through qualitative indicators’ fuzzy quantitative calculation to achieve a multi-indicator optimization and comprehensive decision-making method [16].

In this study, we first scored various types of human factors that cause unsafe incidents in air traffic control through expert judgment. We selected a total of 15 experts to ensure the reliability of the ratings. The selected experts include 10 air traffic controllers with more than 10 years of experience in air traffic control work and 5 university professors with more than 10 years of experience in air traffic management teaching. Then, we used the AHP to calculate and analyze the weight values of various types of human factors in ATC errors, and determined the degree of impact of each human factor on air traffic safety.

4.3 Analysis method and process

To study the impact of various human factors on air traffic safety, we designed a survey questionnaire, which covered various factors of air traffic control personnel at the four levels of the HFACS model. Fifteen aviation experts were asked to rate the degree of impact of human factors in air traffic incidents. Afterward, we analyzed the data using the AHP, and then determined the weight of each human factor that contributes to air traffic incidents.

4.3.1 Establishing a hierarchical structure model

To calculate the weight of the air traffic controller’s work-related errors, we need to decompose the elements into different levels according to the hierarchical relationship between the indicators. Elements in the same level are independent of each other but belong to the previous level. The four levels of the HFACS model were used as the first level, and specific air traffic personnel factors were listed under each level according to the actual situation of air traffic operations, as shown in Fig. 3.

4.3.2 Construction of judgment matrix

For the same level factors under the previous index layer, pairwise comparisons are made to determine the relative importance of each factor. This study adopts a 1–9 comparison scale, as shown in Table 2.

Table 2
Design of the comparative scales

Quantitative value	Element A compared to Element B
1	Equally important
3	Somewhat more important
5	Comparatively more important
7	Very important
9	Absolutely important
2, 4, 6, 8	Midway between adjacent judgments
Reciprocal	aij $=$ 1/aji

Table 3

Scores of experts on factors related to organization influence

$Z$	Organizational influence	Inadequate supervision	Preconditions for unsafe acts	Unsafe acts
Organizational influence	1	3	5	7
Inadequate supervision	1/3	1	3	5
Preconditions for unsafe acts	1/5	1/3	1	3
Unsafe acts	1/7	1/5	1/3	1

Table 4

Scores of experts on factors related to organizational influence

$Z$	Air Traffic Resource Management	Organizational atmosphere	Organizational procedures
Air Traffic Resource Management	1	3	3
Organizational atmosphere	1/3	1	1/2
Organizational procedures	1/3	2	1

Figure 3.

Hierarchy model of human factors in ATC.

After statistical analysis, the scores given by each expert for each level and factor are shown in Tables 3 to 7.

Table 5

Scores of experts on factors related to inadequate supervision

$Z$	Inadequate supervision	Improper operating plans	Uncorrected problems	Non-compliance with supervision
Inadequate supervision	1	3	3	3
Improper operating plans	1/3	1	5	3
Uncorrected problems	1/3	1/5	1	1
Non-compliance with supervision	1/3	1/3	1	1

Table 6

Scores of experts on factors related to preconditions for unsafe acts

$Z$	Controller status	Personnel factors	Environmental factors
Controller status	1	3	3
Personnel factors	1/3	1	2
Environmental factors	1/3	1/2	1

Table 7

Scores of experts on factors related to unsafe acts

$Z$	Skill-based errors	Decision errors	Violations
Skill-based errors	1	3	5
Decision errors	1/3	1	3
Violations	1/5	1/3	1

Using the 1–9 comparison scale and expert ratings, the judgment matrices were constructed as shown in Eq. (1).

$\displaystyle A=\left[{{\begin{array}[]{*{20}c}{a_{11}}&{a_{12}}&\ldots&{a_{1n% }}\\ {a_{21}}&{a_{22}}&\ldots&{a_{2n}}\\ \ldots&\ldots&\ldots&\ldots\\ {a_{n1}}&{a_{n2}}&\ldots&{a_{nn}}\\ \end{array}}}\right]$ (1)

4.3.3 Hierarchical sorting and its consistency test

Using the weighted product method and obtain the weights of each element in the compared index layer, as shown in Eq. (2).

$\displaystyle W_{i}=\mathop{\sum}\limits_{j=1}^{n}\frac{a_{ij}}{n}$ (2)

The corresponding column vectors for each indicator were obtained using Eq. (3).

$\displaystyle\overrightarrow{W}=(W_{1},W_{2},\ldots W_{n})^{T}$ (3)

Then normalized to obtain the relative weights of each indicator. After obtaining the relative weights, consistency testing was performed on the judgment matrix. If the consistency test did not pass, the judgment matrix needed to be revised.

First, the maximum eigenvalue $\lambda_{\max}$ of the judgment matrix was obtained by substituting W and AW into Eq. (4).

$\displaystyle{\lambda}_{{\max}}=\mathop{\sum}\limits_{{i}=1}^{n}\frac{({{AW}_{% i}})_{i}}{{nw}_{i}}$ (4)

Then, the consistency index CI was used to test the consistency of the judgment matrix, as shown in Eq. (5).

$\displaystyle\textit{CI}=\frac{{\lambda}_{{\max}}-n}{{n}-1}$ (5)

If the CI value was 0, the judgment matrix was considered completely consistent. If the CI value was not 0, the consistency ratio CR was calculated to further judge the consistency of the judgment matrix, as shown in Eq. (6).

$\displaystyle\textit{CR}=\frac{\textit{CI}}{\textit{RI}}$ (6)

If the CR value was less than 0.1, the judgment matrix was considered to have passed the consistency test; otherwise, it was considered to have failed the consistency test and needed to be revised. The average random consistency index RI could be obtained by referring to Table 8.

Table 8

RI value

$n$	1	2	3	4	5	6	7	8	9
RI	0	0	0.58	0.90	1.12	1.24	1.32	1.41	1.45

Tables 9 to 13 show the relative weights, CI and CR values of the factors related to air traffic controller personnel that have been calculated.

Table 9

Relative weights of main human factors

	Weight	Eigenvector	CI	CR
Organizational influence	0.5579	2.3559	0.0395	0.0444
Inadequate supervision	0.2634	1.0996
Preconditions for unsafe acts	0.1219	0.49198
Unsafe acts	0.0569	0.2299

Table 10

Relative weights of organizational factors

	Weight	Eigenvector	CI	CR
Air Traffic Resource Management	0.5889	1.8219	0.02695	0.0518
Organizational atmosphere	0.1592	0.4814
Organizational procedures	0.2518	0.7665

Table 11

Relative weights of factors for inadequate supervision

	Weight	Eigenvector	CI	CR
Inadequate supervision	0.4592	2.0818	0.0882	0.0991
Improper operating plans	0.3156	1.3623
Uncorrected problems	0.10895	0.4414
Non-compliance with supervision	0.1163	0.4835

Table 12

Relative weights of factors for the prerequisites of unsafe acts

	Weight	Eigenvector	CI	CR
Controller status	0.5889	1.8219	0.02695	0.0518
Personnel factors	0.2518	0.7665
Environmental factors	0.1592	0.4814

Table 13

Relative weights of factors for unsafe acts

	Weight	Eigenvector	CI	CR
Skill-based errors	0.6334	1.9454	0.0193	0.0371
Decision errors	0.2605	0.7899
Violations	0.1061	0.3196

4.3.4 Calculation of comprehensive weight

After obtaining the weights of each indicator level, it is necessary to summarize and calculate the comprehensive weight of each indicator in the evaluation indicator system. The comprehensive weight of the secondary indicators is calculated by multiplying the weight of the secondary indicator relative to the primary indicator by the corresponding weight of the primary indicator, and then summing up the weights of all indicators and ranking them accordingly. The results are shown in Table 14.

Table 14
Weights of indicators for human errors in ATC

Primary indicator	Weight	Secondary indicator	Intra-level weight	Comprehensive weight	Ranking
Organizational influence	0.5579	Air Traffic Resource Management	0.5889	0.3285	1
		Organizational atmosphere	0.1592	0.0888	4
		Organizational procedures	0.2518	0.1405	2
Inadequate supervision	0.2634	Inadequate supervision	0.4592	0.121	3
		Improper operating plans	0.3156	0.0831	5
		Uncorrected problems	0.10895	0.0287	10
		Non-compliance with supervision	0.1163	0.0306	9
Preconditions for unsafe acts	0.1219	Controller status	0.5889	0.0718	6
		Personnel factors	0.2518	0.0307	8
		Environmental factors	0.1592	0.0194	11
Unsafe acts	0.0569	Skill-based errors	0.6334	0.036	7
		Decision errors	0.2605	0.0148	12
		Violations	0.1061	0.006	13

4.3.5 Results analysis

Among the four primary indicators of organizational influence, inadequate supervision, preconditions for unsafe acts, and unsafe acts, the organizational factor had the highest weight, accounting for over 50%. The indicators were subsequently ranked based on their weights, with inadequate supervision, preconditions for unsafe acts, and unsafe acts having lower weights, with the weight for unsafe acts being less than 10%. Among the secondary indicators of organizational influence, air traffic control resource management had the highest weight of 0.5889, which exceeded 50%. It was followed by organizational procedures and organizational atmosphere. Among the secondary indicators of inadequate supervision, inadequate supervision had the highest weight of 0.4592, followed by improper operating plans, non-compliance with supervision, and uncorrected problems. Among the secondary indicators of preconditions for unsafe acts, the weight of controller status was the highest at 0.5889, followed by personnel and environmental factors. Among the secondary indicators of unsafe acts, the weight of Skill-based errors was the highest at 0.6334, which also exceeded 50%, followed by decision errors and violations.

The comprehensive weights of all thirteen secondary indicators were ranked as follows: Air Traffic Resource Management, organizational procedures, inadequate supervision, organizational atmosphere, improper operating plans, controller status, Skill-based errors, personnel factors, non-compliance with supervision, uncorrected problems, environmental factors, decision errors, and violations. The comprehensive weight of the organizational factor was 0.3285, which was close to one-third of all human factors in the entire air traffic control system and was the most significant factor among all factors. The weight of violations was only 0.006, which was the smallest among all factors and had a significant difference compared to the second-to-last weight.

5. Conclusion

The results of the Analytic Hierarchy Process indicate that the organizational factor is the most significant human factor in the entire air traffic control system, with its weight accounting for more than half of all human factors in the system. The secondary indicator with the highest weight within the organizational factor was air traffic control resource management, which was also the indicator with the highest comprehensive weight among all thirteen indicators. Unsafe acts had the lowest weight among the four primary indicators, and violations had the lowest weight among all thirteen indicators. Therefore, for air traffic controllers, they are not willing to engage in unsafe behavior. The occurrence of errors by controllers is attributed to the presence of vulnerabilities that permeate through various levels of the HFACS model. This study employed the expert rating method, which has certain limitations due to the subjective nature of expert evaluations. However, compared to other qualitative analysis methods, the expert rating method still has its advantages. In future research, the use of eye-tracking devices, physiological recorders, and other tools can be considered to capture the states of air traffic controllers during errors, which would provide more reliable data.

With the continuous advancement of aviation technology, the reliability of air traffic control facilities has been improved, leading to a decrease in aviation safety incidents caused by hardware equipment. As a result, human error has become the primary cause of aviation accidents. Air traffic management, as an essential component of aviation transportation, plays a crucial role. The issue of human error among air traffic controllers has therefore become increasingly significant. According to the above analysis, to reduce human errors among air traffic controllers and improve the safety level of air traffic control, improvement measures should be proposed primarily from the level of organizational factors. Firstly, air traffic control resource management should be strengthened, and reasonable policies and procedures should be formulated to make the entire air traffic control system’s organizational management more scientific and enable various software and hardware resources to be effectively utilized. Secondly, a good organizational atmosphere should be formed to ensure that problems in the operation of air traffic control organizations can be detected and effectively resolved. Thirdly, it is important to strengthen the oversight of the air traffic control system. It is crucial to promptly identify and address any minor issues and vulnerabilities within the system. Proactive monitoring and timely resolution of these issues are essential to prevent them from escalating into serious safety concerns. Waiting until the problems reach a critical level before taking action may lead to new air traffic control safety issues. Only when the safety level of human factors in the air traffic control system is improved can the defects and shortcomings in inadequate supervision and preconditions for unsafe acts be reduced, ultimately reducing the unsafe behavior of air traffic controllers and reducing human errors, thus ensuring aviation safety.

References

Chan

. Development of effective human factors interventions for aviation safety management. Frontiers in public health. 2023; 11: 1144921.

Karunakaran

Babu

. Impact of human factors in aircraft accident mitigation and aircraft maintenance training needs in post COVID-19 aviation. Aircraft Engineering and Aerospace Technology. 2022; 94(8): 1296-1302.

Foster

Plant

Stanton

. Maladaptation in air traffic management: Development of a Human Factors methods framework. Human Factors and Ergonomics in Manufacturing & Service Industries. 2023; 33(1): 118-146.

Miriam

Rosemary

. Effect of Fatigue in Air Traffic Controllers’ Workload, Situation Awareness, and Control Strategy. The International Journal of Aerospace Psychology. 2022; 32(1): 1-23.

Abbas

Maryam

. Classification and Quantification of Human Error in the Air Traffic Control: A Case Study in an Airport Control Tower. International Journal of Occupational Safety and Ergonomics. 2020; 27(4): 1222-1234.

Sepideh

Vahid

Morteza

, et al. Evaluating differences in cognitive functions and personality traits among air traffic controllers with and without error history. Safety Science. 2021; 139: 1-8.

Salmon

Lenne

Stanton

, et al. Managing error on the open road: The contribution of human error models and methods. Safety Science. 2010; 48(10): 1225-1235.

Shappell

Detwiler

Holcomb

, et al. Human error and commercial aviation accidents: An analysis using the human factors analysis and classification system. Human Factors. 2007; 49(2): 227-242.

Wiegmann

Shappell

. Human error perspectives in aviation. International Journal of Aviation Psychology. 2001; 11(4): 341-357.

10.

Meng

. A Hybrid Model Integrating HFACS and BN for Analyzing Human Factors in CFIT Accidents. Aerospace. 2022; 9(11): 711-731.

11.

Lyu

Song

. Human Factors Analysis of Air Traffic Safety Based on HFACS-BN Model. Applied Sciences. 2019; 9(23): 5049.

12.

Lower

Magott

Skorupski

. A system-theoretic accident model and process with human factors analysis and classification system taxonomy. Safety Science. 2018; 110: 393-410.

13.

Kaptan

Sarialioglu

Ugurlu

, et al. The evolution of the HFACS method used in analysis of marine accidents: A review. International Journal of Industrial Ergonomics. 2021; 86: 103225.

14.

Sha

Zhang

, et al., Systematic Analysis of the Contributory Factors Related to Major Coach and Bus Accidents in China. Sustainability. 2022; 14(22): 15354.

15.

Wang

Sven

, et al. Cause analysis of unsafe acts of pilots in general aviation accidents in China with a focus on management and organizational factors. International Journal of Occupational Safety and Ergonomics. 2022; (5): 1-14.

16.

Chen

Yang

Chang

. A model integrating fuzzy AHP with QFD for assessing technical factors in aviation safety. International Journal of Machine Learning and Cybernetics. 2014; 5(5): 761-774.