Unsafe acts and unsafe outcomes in aircraft maintenance

Road safety studies using the Driver Behaviour Questionnaire (DBQ) have provided support for a three-way distinction between violations, skill-based errors and mistakes, and have indicated that a tendency to commit driving violations is associated with an increased risk of accident involvement. The aims of this study were to examine whether the three-way distinction of unsafe acts is applicable in the context of aircraft maintenance, and whether involvement in maintenance safety occurrences can be predicted on the basis of self-reported unsafe acts. A Maintenance Behaviour Questionnaire (MBQ) was developed to explore patterns of unsafe acts committed by aircraft maintenance mechanics. The MBQ was completed anonymously by over 1300 Australian aviation mechanics, who also provided information on their involvement in workplace accidents and incidents. Four factors were identified: routine violations, skill-based errors, mistakes and exceptional violations. Violations and mistakes were related significantly to the occurrence of incidents that jeopardized the quality of aircraft maintenance, but were not related to workplace injuries. Skill-based errors, while not related to work quality incidents, were related to workplace injuries. The results are consistent with the three-way typology of unsafe acts described by Reason et al. (1990) and with the DBQ research indicating an association between self-reported violations and accidents. The current findings suggest that interventions addressed at maintenance quality incidents should take into account the role of violations and mistakes, and the factors that promote them. In contrast, interventions directed at reducing workplace injury are likely to require a focus on skill-based errors.     

Unsafe acts and unsafe outcomes in aircraft maintenance

Road safety studies using the Driver Behaviour Questionnaire (DBQ) have provided support for a three-way distinction between violations, skill-based errors and mistakes, and have indicated that a tendency to commit driving violations is associated with an increased risk of accident involvement. The aims of this study were to examine whether the three-way distinction of unsafe acts is applicable in the context of aircraft maintenance, and whether involvement in maintenance safety occurrences can be predicted on the basis of self-reported unsafe acts. A Maintenance Behaviour Questionnaire (MBQ) was developed to explore patterns of unsafe acts committed by aircraft maintenance mechanics. The MBQ was completed anonymously by over 1300 Australian aviation mechanics, who also provided information on their involvement in workplace accidents and incidents. Four factors were identified: routine violations, skill-based errors, mistakes and exceptional violations. Violations and mistakes were related significantly to the occurrence of incidents that jeopardized the quality of aircraft maintenance, but were not related to workplace injuries. Skill-based errors, while not related to work quality incidents, were related to workplace injuries. The results are consistent with the three-way typology of unsafe acts described by Reason et al. (1990) and with the DBQ research indicating an association between self-reported violations and accidents. The current findings suggest that interventions addressed at maintenance quality incidents should take into account the role of violations and mistakes, and the factors that promote them. In contrast, interventions directed at reducing workplace injury are likely to require a focus on skill-based errors.     

Application of HFACS and grounded theory for identifying risk factors of air traffic controllers’ unsafe acts

Unsafe acts of air traffic controllers (ATCers) are caused by various factors. Based on interview data and case reports, human factors analysis and classification system (HFACS) and the grounded theory were adopted to identify the risk factors of ATCers’ unsafe acts comprehensively. The interview data and the case data issued by the authority were first collected. Then, the above data were encoded to obtain the relevant concepts and categories based on the grounded theory, and the HFACS model is used to classify the concepts and categories. Finally, the relationship between the core category and the secondary category was sorted out in the way of storyline. The results show that the risk factors include environmental factors, organizational influences, unsafe supervision and controllers’ states, and the unsafe acts manifest as errors and violations. Among them, the controllers’ states are intermediate variable, and other factors indirectly affect the controllers’ unsafe acts. The first three risk factors with high frequency in unsafe incidents are technical environment, mental states and business ability. The three most common unsafe acts are giving the wrong order, insufficient situational awareness, and poor work order on-site. Through combining HFACS framework and grounded theory to analyze data, a more clear and comprehensive conceptual model of risk factors of ATCers’ unsafe acts can be obtained.     

Predicting unsafe behaviors at nuclear power plants: An integration of Theory of Planned Behavior and Technology Acceptance Model

Unsafe behaviors, including both errors and violations, have been identified as key contributors to accidents at nuclear power plants. However, the mechanisms behind such unsafe behaviors are not fully understood. This study aimed to investigate how workers' attitude and perception factors would predict errors and violations at nuclear power plants by proposing and validating an unsafe behavior model. The proposed model applied the Theory of Planned Behavior as the ground theoretical model and added factors from Technology Acceptance Model to capture workers’ perception of work regulations. To examine the validity of the proposed model, a total of 178 questionnaires were distributed at two nuclear power plants in China and 171 valid questionnaires were returned. The Structured Equation Modeling (SEM) analysis indicated that the proposed model fitted the data well. The results showed that perceived ease of use and perceived usefulness in following work regulations contributed to a positive attitude, which helped reduce the occurrence of both errors and violations. Moreover, errors were further affected by subjective norm while violations were not. Perceived behavior control was not a significant factor of either errors or violations.These findings suggest that to reduce errors and violations, policymakers should focus on methods to improve the perceived usefulness and perceived ease of use of work regulations, and promote a positive attitude towards safety.     

The evolution of the HFACS method used in analysis of marine accidents: A review

The importance of accident investigations carried out in every field where operators play a vital role is increasingly recognised. Many researchers argue that understanding accident formation is the most important way to prevent future disasters. In this research, an analysis of the modified Human Factor Analysis and Classification System (HFACS) structures developed for use in the analysis of marine accidents was conducted. These structures include HFACS-PV (Passenger Vessels), HFACS-MA (Maritime Accidents), HFACS-Coll (Collisions), HFACS-SIBCI (ship collision accidents between assisted ships and icebreakers in ice-covered waters) and HFACS-Ground (Groundings). In this study, revisions in HFACS structures were examined. It was found that the accident factors were classified at different levels to facilitate the application of the original HFACS framework. The first of the remarkable differences among the basically developed methods is the level of external factors (first level), where the accident factors arising from national and international rules are classified. The second is the level of operational conditions (last level). It has been observed that the precondition for the unsafe acts level has been revised in all methods examined. This study will guide researchers in choosing an HFACS structure suitable for the area they will study, as well as revealing different aspects of the modified methods examined in marine accident analysis.     

A quarter of a century of the DBQ: some supplementary notes on its validity with regard to accidents

This article synthesises the latest information on the relationship between the Driver Behaviour Questionnaire (DBQ) and accidents. We show by means of computer simulation that correlations with accidents are necessarily small because accidents are rare events. An updated meta-analysis on the zero-order correlations between the DBQ and self-reported accidents yielded an overall r of .13 (fixed-effect and random-effects models) for violations (57,480 participants; 67 samples) and .09 (fixed-effect and random-effects models) for errors (66,028 participants; 56 samples). An analysis of a previously published DBQ dataset (975 participants) showed that by aggregating across four measurement occasions, the correlation coefficient with self-reported accidents increased from .14 to .24 for violations and from .11 to .19 for errors. Our meta-analysis also showed that DBQ violations (r = .24; 6353 participants; 20 samples) but not DBQ errors (r = ? .08; 1086 participants; 16 samples) correlated with recorded vehicle speed.

Practitioner Summary: The DBQ is probably the most widely used self-report questionnaire in driver behaviour research. This study shows that DBQ violations and errors correlate moderately with self-reported traffic accidents.     

A systems approach to organizational error

Abstract

The past two decades have seen a significant number of large-scale disasters in a wide range of hazardous, well-defended technologies. Despite their differences, the root causes of these accidents have been traced to latent failures and organizational errors arising in the upper echelons of the system in question. A model of the aetiology of these organizational accidents is outlined. The model describes two interrelated causal sequences: (a) an active failure pathway that originates in top-level decisions and proceeds via error-producing and violation-promoting conditions in the various workplaces to unsafe acts committed by those at the immediate human-system interface and (b) a latent failure pathway that runs directly from the organizational processes to deficiencies in the system's defences. The paper goes on to identify two sets of dependencies associated with latent failures and violations. Organizational errors increase the likelihood of operator error through the active failure pathway and, at the same time, enhance the possibility of adverse outcomes through defensive weaknesses. Violations have a narrower range of consequences. Non-compliance with safe operating procedures increases the likelihood of error by taking perpetrators into regions of operation in which neither the physical regime nor the hazards are well understood. Violations, by definition, also take perpetrators ‘closer to the edge,’ and thus increase the chance that subsequent errors will have damaging outcomes. The paper concludes by indicating two ways in which the model has been applied in industrial settings: (a) through the development of proactive measures for diagnosing and remedying organizational processes known to be implicated in accident causation, and (b) an accident investigation technique that guides investigators and analysts to the organizational root causes of past accidents.     

Computer vision and long short-term memory: Learning to predict unsafe behaviour in construction

Predicting unsafe behaviour in advance can enable remedial measures to be put in place to mitigate likely accidents on construction sites. Prevailing safety studies in construction tend to be retrospective and focus on examining the conditions that contribute to unsafe behaviour from a psychological perspective. While such studies are warranted, they can also not visually comprehend the dynamic and complex conditions that influence unsafe behaviour. In this paper, we aim to contribute to filling this void and, in doing so, combine computer vision with Long-Short Term Memory (LSTM) to predict unsafe behaviours from videos automatically. Our proposed approach for predicting unsafe behaviour is based on: (1) tracking people using a SiamMask; (2) predicting the trajectory of people using an improved Social-LSTM; and (3) predicting unsafe behaviour using Franklin's point inclusion polygon (PNPoly) algorithm. We use the Wuhan metro project as a case to evaluate our approach’s feasibility and effectiveness. Our adopted SiamMask method outperforms current techniques used for tracking people. Additionally, our improved Social-LSTM can achieve higher accuracy on trajectory prediction than other methods (e.g., Social-GAN). The research findings demonstrate that our developed computer vision approach can be used to accurately predict unsafe behaviour on construction sites.     

Cognitive failure analysis for aircraft accident investigation

Abstract

The present studies were undertaken to investigate the applicability of an information processing approach to human failure in the aircraft cockpit. Using data obtained from official aircraft accident investigation reports, a database of accidents and incidents involving New Zealand civil aircraft between 1982 and 1991 was compiled. In the first study, reports were coded into one of three error stages proposed by Nagel (1988) and for the presence of any of 61 specific errors noted by Gerbert and Kemmler (1986). The importance of decisional factors in fatal crashes was noted. Principal components analysis suggested the presence of five different varieties of human failure. In the second study, a more detailed error taxonomy derived from the work of Rasmussen (1982) was applied to the data. Goal selection errors emerged as the most frequent kind of cognitive error in fatal accidents. Aircraft accident reports can be a useful source of information about cognitive failures if probed with an appropriate, theoretically-based, analysis of information processing errors. Such an approach could provide the accident investigators with a useful tool, and lead to a more complete understanding of human error in aviation.     

Aberrant driving behaviour: errors and violations

The present study was a replication of the research of Reason et al. (1990). Its aim was to confirm the distinction between driving errors and violations in a Western Australian driving population. Sixty-one male drivers and 74 female drivers completed a questionnaire containing items on driver demographics, driving penalties incurred, driving convictions and accident history and driver behavioural aberrations drawn from the Driver Behaviour Questionnaire (DBQ). In agreement with Reason et al. factor analysis revealed three factors; in the present study these were general errors, dangerous errors, and dangerous violations. Young drivers committed more dangerous errors and dangerous violations than older drivers. Females reported more dangerous errors than males. Males reported more dangerous violations than females. Drivers who reported a high level of road exposure and those who reported having been convicted for speeding reported more dangerous violations. Differences in the results of the two studies can largely by accounted for by differences in the representation of age and gender in the two populations studied.     

Evaluating non-technical skills and mission essential competencies of pilots in military aviation environments

Abstract

To develop and validate a classification of non-technical skills (NTS) in military aviation, a study was conducted, using data from real operations of F16 aircraft formations. Phase 1 developed a NTS classification based on the literature review (e.g. NOTECHS) and a workshop with pilots. The Non-TEChnical-MILitary-Skills (NOTEMILS) scheme was tested in Phase 2 in a series of Principal Component Analysis with data from After-Action-Review sessions (i.e. 900 records from a wide range of operations). The NTS were found to make a good prediction of Mission Essential Components (R² > 0.80) above the effect of experience. Phase 3 undertook a reliability analysis where three raters assessed the NOTEMILS scheme with good results (i.e. all r_wg > 0.80). To look into the consistency of classifications, another test indicated that, at least, two out of three raters were in agreement in over 70% of the assessed flight segments.

Practitioner Summary: A classification scheme of Non-Technical Skills (NTS) was developed and tested for reliability in military aviation operations. The NTS scheme is a valuable tool for assessing individual and team skills of F-16 pilots in combat. It is noteworthy that the tool had a good capability of predicting Mission Essential Competencies.     

Stray Example Sheltering by Loss Regularized SVM and <Emphasis Type="Italic">k</Emphasis>NN Preprocessor

This paper presents a new model developed by merging a non-parametric k-nearest-neighbor (kNN) preprocessor into an underlying support vector machine (SVM) to provide shelters for meaningful training examples, especially for stray examples scattered around their counterpart examples with different class labels. Motivated by the method of adding heavier penalty to the stray example to attain a stricter loss function for optimization, the model acts to shelter stray examples. The model consists of a filtering kNN emphasizer stage and a classical classification stage. First, the filtering kNN emphasizer stage was employed to collect information from the training examples and to produce arbitrary weights for stray examples. Then, an underlying SVM with parameterized real-valued class labels was employed to carry those weights, representing various emphasized levels of the examples, in the classification. The emphasized weights given as heavier penalties changed the regularization in the quadratic programming of the SVM, and brought the resultant decision function into a higher training accuracy. The novel idea of real-valued class labels for conveying the emphasized weights provides an effective way to pursue the solution of the classification inspired by the additional information. The adoption of the kNN preprocessor as a filtering stage is effective since it is independent of SVM in the classification stage. Due to its property of estimating density locally, the kNN method has the advantage of distinguishing stray examples from regular examples by merely considering their circumstances in the input space. In this paper, detailed experimental results and a simulated application are given to address the corresponding properties. The results show that the model is promising in terms of its original expectations.     

A socio-technical analysis of functional properties in a joint cognitive system: a case study in an aircraft cockpit

Abstract

In a socio-technical work domain, humans, device interfaces and artefacts all affect transformations of information flow. Such transformations, which may involve a change of auditory to visual information & vice versa or alter semantic approximations into spatial proximities from instruments readings, are generally not restricted to solely human cognition. This paper applies a joint cognitive system approach to explore a socio-technical system. A systems ergonomics perspective is achieved by applying a multi-layered division to transformations of information between, and within, human and technical agents. The approach uses the Functional Resonance Analysis Method (FRAM), but abandons the traditional boundary between medium and agent in favour of accepting aircraft systems and artefacts as agents, with their own functional properties and relationships. The joint cognitive system perspective in developing the FRAM model allows an understanding of the effects of task and information propagation, and eventual distributed criticalities, taking advantage of the functional properties of the system, as described in a case study related to the cockpit environment of a DC-9 aircraft.

Practitioner Summary: This research presents the application of one systemic method to understand work systems and performance variability in relation to the transformation of information within a flight deck for a specific phase of flight. By using a joint cognitive systems approach both retrospective and prospective investigation of cockpit challenges will be better understood.

Abbreviations: ATC: air traffic control; ATCO: air traffic controller; ATM: air traffic management; CSE: cognitive systems engineering; DSA: distributed situation awareness; FMS: flight management system; FMV: FRAM model visualize; FRAM: functional resonance analysis method; GF: generalised function; GW: gross weight; HFACS: human factors analysis and classification system; JCS: joint cognitive systems; PF: pilot flying; PNF: pilot not flying; SA: situation awareness; SME: subject matter expert; STAMP: systems theoretic accident model and processes; VBA: visual basic for applications; WAD: work-as-done; WAI: work-as-imagined; ZFW: zero fuel weight     

遥感图像飞机目标分类的卷积神经网络方法

目的 遥感图像飞机目标分类,利用可见光遥感图像对飞机类型进行有效区分,对提供军事作战信息有重要意义。针对该问题,目前存在一些传统机器学习方法,但这些方法需人工提取特征,且难以适应真实遥感图像的复杂背景。近年来,深度卷积神经网络方法兴起,网络能自动学习图像特征且泛化能力强,在计算机视觉各领域应用广泛。但深度卷积神经网络在遥感图像飞机分类问题上应用少见。本文旨在将深度卷积神经网络应用于遥感图像飞机目标分类问题。方法 在缺乏公开数据集的情况下,收集了真实可见光遥感图像中的8种飞机数据,按大致4∶1的比例分为训练集和测试集,并对训练集进行合理扩充。然后针对遥感图像与飞机分类的特殊性,结合深度学习卷积神经网络相关理论,有的放矢地设计了一个5层卷积神经网络。结果 首先,在逐步扩充的训练集上分别训练该卷积神经网络,并分别用同一测试集进行测试,实验表明训练集扩充有利于网络训练,测试准确率从72.4%提升至97.2%。在扩充后训练集上,分别对经典传统机器学习方法、经典卷积神经网络LeNet-5和本文设计的卷积神经网络进行训练,并在同一测试集上测试,实验表明该卷积神经网络的分类准确率高于其他两种方法,最终能在测试集上达到97.2%的准确率,其余两者准确率分别为82.3%、88.7%。结论 在少见使用深度卷积神经网络的遥感图像飞机目标分类问题上,本文设计了一个5层卷积神经网络加以应用。实验结果表明,该网络能适应图像场景,自动学习特征,分类效果良好。     

Speech variability effects on recognition accuracy associated with concurrent task performance by pilots

Abstract

Pairs of pilots performed an aircraft warning classification task using an isolated-word speaker-dependent speech recognition system to record their responses. Task-induced stress was manipulated by means of different scoring procedures for the warning classification task and by inclusion of a competitive manual control task (Voorhees and Bucher 1984, Voorhees and Bucher 1985). High quality tape recordings were made of pilots’ speech spoken to the recognizer while performing under the different types of task loading. This speech was then analysed spectrographically for timing differences in the pilot's manner of speaking as a function of task loading, and of recognition accuracy associated with different types of task loading. Recognition errors were recorded by type for an isolated-word speaker-dependent system and by an off-line technique for a connected-word speaker-dependent system. While recognition errors increased with task loading for the isolated word system, there was no such effect for task loading when tape recordings of the same utterances were recognized by a connected word system. The results suggest areas for further improvements to speaker-dependent recognition algorithms and provide predictions of the types of recognition errors that can be expected with current recognition technology used by pilots under moderate to high levels of task loading.     

The causes of causes: determinants and background variables of human factor incidents and accidents

Abstract

1448 German Air Force pilots took part in an anonymous investigation of critical flight incidents. A 315-item questionnaire was applied, which included relevant preflight and inflight data, physiological and psychological parameters (behaviour variables, subjective state ratings before and during critical situations, pilot errors and their consequences) as well as environmental factors. Extensive data analysis revealed various human errors, which can be intepreted as a four-dimensional error-structure: vigilance errors, perception errors, information processing errors and sensorimotor errors. Determinants and background variables of human factor incidents and accidents were discussed in a model of basic man–machine system interaction patterns. Preventive measures to limit and control human factor incidents and accidents are proposed in a four-step programme.     

Errors in routine driving tasks: a model and proposed analysis technique

Abstract

Models of behaviour and cognitive errors are briefly reviewed. A model derived from the work of Reason and Rasmussen is applied to the analysis of accidents at a crossroad in Friesland where the bulk of accidents occurred to drivers familiar with the location. Detailed application of the predicted error mechanisms to the steps of the driving task allows testable hypotheses to be formulated to differentiate between a number of possible causes of the accidents. The implications for research into accidents during routine driving tasks are discussed.     

Driving errors,driving violations and accident involvement

Abstract

A survey of over 1600 drivers is reported, the results of which are consistent with those reported in an earlier study (Reason et al. 1990), which identified a three-fold typology of aberrant driving behaviours. The first type, lapses, are absent-minded behaviours with consequences mainly for the perpetrator, posing no threat to other road users. The second type, errors, are typically misjudgements and failures of observation that may be hazardous to others. The third type, violations, involve deliberate contraventions of safe driving practice. In the present study the survey instrument used, the Driver Behaviour Questionnaire, was also shown to be reliable over time. Each type of behaviour was found to have different demographic correlates. Most importantly, accident liability was predicted by self-reported tendency to commit violations, but not by tendency to make errors or to have lapses. The implications for road safety are discussed.     

A method for investigating human factor aspects of aircraft accidents and incidents

Approximately 70% of aircraft accidents and incidents have in the past been attributed to human error, and the importance of human error as a major contributory factor is now universally recognized. A satisfactory technique for the investigation of human error type accidents and incidents has not yet been standardized. This paper describes a comprehensive procedure which has been designed to assess human behaviour in instances of aircraft accidents and incidents. The aim of such a human factors investigation is to understand why the accident or incident occurred and so lead to a better understanding of human error, and as a result enhance flight safety.

The use of questionnaires and interviews to assess the subjective data available is discussed, and it is recommended that the most appropriate technique for the assessment of behavioural data is the use of an interview together with a check list. The importance of an early interview with all personnel associated with an accident or an incident by a human factors specialist trained in interviewing techniques is emphasized.

The check list described has been developed after reviewing the relevant literature and has been modified by personal experience derived from the investigation of Army aircraft accidents and incidents in which human factors have played an important part. The check list is based on a systems approach to understanding human error and includes such headings as ‘stress’ (including life events), ‘fatigue’, ‘arousal’ and ‘personality’. It is hoped that further improvements will be made to the check list to enable it to be used not only within the specific area of aircraft accidents and incidents, but also within the broader context of any accidents which may be attributed to human errors and as a result reduce the number of accidents and identify areas where further research may be useful.