Content uploaded by Guohua Li
Author content
All content in this area was uploaded by Guohua Li on Sep 03, 2014
Content may be subject to copyright.
A preview of the PDF is not available
Factors associated with pilot error in aviation crashes
Abstract
The importance of pilot error in aviation crashes has long been recognized. However, understanding and preventing pilot error remains the foremost challenge in aviation safety.
This study aims to identify pilot characteristics and crash circumstances that are associated with the presence of pilot error in a large sample of aviation crashes.
Different data files compiled by the National Transportation Safety Board for 329 major airline crashes, 1,627 commuter/air taxi crashes, and 27,935 general aviation crashes for the years 1983-96 were merged; and the presence of pilot error was analyzed in relation to variables indicating the characteristics of the pilot-in-command, crash circumstance, and aircraft. Multivariate logistic regression modeling was performed to assess the associations of individual variables with the likelihood of pilot error given a crash.
Pilot error was a probable cause in 38% of the major airline crashes, 74% of the commuter/air taxi crashes, and 85% of the general aviation crashes. Among the factors examined, instrument meteorological condition and on-airport location were each associated with a significantly increased odds of pilot error. The likelihood of pilot error decreased as pilot certificate rating increased in commuter/air taxi and general aviation crashes. Neither pilot age nor gender was independently associated with the odds of pilot error. With adjustment for pilot characteristics and crash circumstances, flight experience as measured in total flight time showed a significant protective effect on pilot error in general aviation crashes.
The prevalence and correlates of pilot error in aviation crashes vary with the type of flight operations. Adverse weather is consistently associated with a significantly elevated likelihood of pilot error, possibly due to increased performance demand.
... Almost all the cases where either pilots or passengers suffered injuries happened during the transition from visual flight rules (VFR) into instrument meteorological condition (IMC) in United States (National Transportation Safety Board 1989) or Canada (Transportation Safety Board of Canada 1990). In Australia, 7.8% of the aircrafts involved in weather-related accident were scrapped while 12.1% of the cases of VFR into IMC involved a fatality (Batt and O'Hare 2005); a higher prevalence of pilot errors leading to aviation accidents in IMC conditions was noted also by (Li et al. 2001) due to the incorrect operations and inputs made by the pilots. ...
... Predicting aviation accidents using the environment predictors solely cannot be entirely meaningful and is a subject to cautions since it is excluding human behavior or possible technical malfunctions that might arise after the airplane is facing adverse weather like severe turbulence, lightning or strong winds. Therefore, it is hard to prove that the remaining of the ~ 30% of the model prediction can be attributable to human errors or any other hazard, although (Harizi et al. 2013) found that approximately 21.6% of the human errors appear in unfavorable weather conditions while (Li et al. 2001) estimated this ratio up to 38%. On the other way, (Fultz and Ashley 2016) found only 35% of the fatal aviation accidents to be related to weather. ...
Over 8000 weather-related aviation accidents in the USA from 2000 to 2020 are investigated, in order to find possible connections between atmospheric instability and severe aviation accidents – defined here as events that recorded serious or fatal injuries. We use the bootstrapped median values for multiple meteorological parameters extracted from North America Regional climate reanalysis for each accident cause. To check the possibility of predicting severe accidents, we applied a logistic regression model containing several meteorological variables as predictands. Results show that approximately 67% of the weather-related severe aviation accidents can be explained by meteorological parameters like relative humidity, temperature, visibility or total cloud fraction. The monthly frequency of weather-related accidents shows a 6-months peak, corresponding to the peak of the convective season in North America. Older pilots (> 60 years) are more likely to be involved in a severe accident during adverse weather, while flight experience seems less relevant. Our study aims to better understand which meteorological variables are more significant in the outcome of severe aviation accidents during adverse meteorological conditions, and to investigate the roles of age and experience in pilots' response during extreme weather.
... Experience is a stronger predictor of risky decisions than age (Li et al. 2003). The researchers concluded that pilots who lacked flight experience had a significantly higher chance of being involved in hazardous events (Causse, Dehais, and Pastor 2011;Jarvis and Harris 2010;Goh and Wiegmann 2002;Li et al. 2001). Aviation accidents may occur as a result of inexperienced pilots' inability to effectively obtain environment-related information and decisions to continue flying in adverse weather conditions. ...
... Participants with greater flight hours perceived their risk-taking behaviour was easier to control and thought they were safer pilots. Due to frequent exposure to different weather conditions, pilots with greater flying experience are better able to perceive and control changing external information and make more accurate decisions (Li et al. 2001). In terms of skill level, the ANOVA revealed that flying cadets and first officers had higher intentions to take risks during approach, indicating that pilots with lower skill levels were more likely to accept a higher level of risks under adverse weather conditions. ...
This paper examined pilots’ risk-taking behavioural intentions based on the theory of planned behaviour, as well as the impact of experience on behavioural intentions in adverse weather conditions. Two hundred and seventy-three airline pilots and flying cadets were divided into two groups and asked to complete a questionnaire based on two decision-making scenarios. This questionnaire measured pilots’ intentions to take risks, along with the attitude towards the behaviour, subjective norms, perceived behavioural control (PBC), risk perception, and self-identity. The results showed that attitude, subjective norm, PBC, and risk perception explained 52% of the variance in behavioural intentions. Additionally, pilots’ risk-taking decisions can be influenced by experience. Inexperienced pilots had a relatively stronger intention to take risks and a more favourable attitude towards risky behaviour. Moreover, pilots were more likely to rely on their own direct experience in the decision-making process.
Practitioner summary: This study examined the pilots’ risk-taking intentions under adverse weather conditions using a questionnaire based on the TPB theory. Results demonstrated that the TPB model can be applied to the risk-taking scenario and that experience can influence pilots’ decisions. These findings have implications for improving flight safety and lowering accident rates.
... As this paper was part of a project evaluating SMS in airline flight operations departments (i.e., the pilot department), a literature review was carried out to determine how previous researchers measured experience and expertise among pilots. Possession of more advanced licenses and working for larger airlines were known to be associated with a lower incidence of error (16,17). However, these variables are of limited relevance for organizational SMS applications where all the participants will be working in the same organization, with identical prerequisite licenses. ...
Introduction
In the aviation industry, safety management has moved away from capturing frontline failures toward the management of systemic conditions through organizational safety management systems (SMS). However, subjective differences can influence the classification of active failures and their associated systemic precursors. With levels of professional experience known to influence safety attitudes, the present research examines whether experience levels among airline pilots had an impact on the classification of causal factors using the Human Factors Analysis and Classification System (HFACS). Differences in the paths of association between categories were evaluated in an open-system context.
Method
Pilots working in a large, international airline were categorized into high (≥10,000 total flight hours) and low (<10,000 h) experience groups and asked to classify aircraft accident causal factors using the HFACS framework. One-way ANOVA tests were carried out to determine experience effects on the utilization of the HFACS categories, and chi-squared analyses were used to assess the strength of association between different categories within the framework.
Results
Results from 144 valid responses revealed differences in the attribution of human factors conditions. The high experience group was more inclined to attribute deficiencies to high-level precursors and found fewer paths of associations between different categories. In contrast, the low experience group presented a greater number of associations and was comparatively more affected by stress and uncertainty conditions.
Discussion
The results confirm that the classification of safety factors can be influenced by professional experience, with hierarchical power distance impacting the attribution of failures to higher-level organizational faults. Different paths of association between the two groups also suggest that safety interventions can be targeted through different entry points. Where multiple latent conditions are associated, the selection of safety interventions should be made with consideration of the concerns, influences, and actions across the entire system. Higher-level anthropological interventions can change the interactive interfaces affecting concerns, influences, and actions across all levels, whereas frontline-level functional interventions are more efficient for failures linked to many precursor categories.
... Up to 85% of manned aircraft incidents are now thought to be due to pilot error, which is proportionally due to technical issues being removed by improved design processes. 50 Because of differences in construction, piloting and application of UA, different proportions of errors may come from pilots or remote operators and other technical parts of the system. 51 ...
In the aviation industry, safety remains vital, often compromised by pilot errors attributed to factors such as workload, fatigue, stress, and emotional disturbances. To address these challenges, recent research has increasingly leveraged psychophysiological data and machine learning techniques, offering the potential to enhance safety by understanding pilot behavior. This systematic literature review rigorously follows a widely accepted methodology, scrutinizing 80 peer-reviewed studies out of 3352 studies from five key electronic databases. The paper focuses on behavioral aspects, data types, preprocessing techniques, machine learning models, and performance metrics used in existing studies. It reveals that the majority of research disproportionately concentrates on workload and fatigue, leaving behavioral aspects like emotional responses and attention dynamics less explored. Machine learning models such as tree-based and support vector machines are most commonly employed, but the utilization of advanced techniques like deep learning remains limited. Traditional preprocessing techniques dominate the landscape, urging the need for advanced methods. Data imbalance and its impact on model performance is identified as a critical, under-researched area. The review uncovers significant methodological gaps, including the unexplored influence of preprocessing on model efficacy, lack of diversification in data collection environments, and limited focus on model explainability. The paper concludes by advocating for targeted future research to address these gaps, thereby promoting both methodological innovation and a more comprehensive understanding of pilot behavior.
General aviation is the most diverse sector in all of aviation, encompassing everything outside of military and airline flying. From its great diversity arises a similar variety of safety challenges. This chapter surveys what we have learned about the human factors of staying safe while we teach and train, conduct business in the sky, and when we fly for fun. Some human factors research will confirm ideas that you have long held—while other fascinating studies will invite you to think about old topics in new ways.
From the reviews of the First Edition."An interesting, useful, and well-written book on logistic regression models . . . Hosmer and Lemeshow have used very little mathematics, have presented difficult concepts heuristically and through illustrative examples, and have included references."—Choice"Well written, clearly organized, and comprehensive . . . the authors carefully walk the reader through the estimation of interpretation of coefficients from a wide variety of logistic regression models . . . their careful explication of the quantitative re-expression of coefficients from these various models is excellent."—Contemporary Sociology"An extremely well-written book that will certainly prove an invaluable acquisition to the practicing statistician who finds other literature on analysis of discrete data hard to follow or heavily theoretical."—The StatisticianIn this revised and updated edition of their popular book, David Hosmer and Stanley Lemeshow continue to provide an amazingly accessible introduction to the logistic regression model while incorporating advances of the last decade, including a variety of software packages for the analysis of data sets. Hosmer and Lemeshow extend the discussion from biostatistics and epidemiology to cutting-edge applications in data mining and machine learning, guiding readers step-by-step through the use of modeling techniques for dichotomous data in diverse fields. Ample new topics and expanded discussions of existing material are accompanied by a wealth of real-world examples-with extensive data sets available over the Internet.
The role of human error in commercial and general aviation accidents and the techniques used to evaluate it are reviewed from a human-factors perspective. Topics addressed include the general decline in accidents per million departures since the 1960s, the increase in the proportion of accidents due to human error, methods for studying error, theoretical error models, and the design of error-resistant systems. Consideration is given to information acquisition and processing errors, visually guided flight, disorientation, instrument-assisted guidance, communication errors, decision errors, debiasing, and action errors.
Stress is “the nonspecific response of the body to any demand made upon it,” that is, the rate at which we live at any one moment. All living beings are constantly under stress and anything, pleasant or unpleasant, that speeds up the intensity of life, causes a temporary increase in stress, the wear and tear exerted upon the body. A painful blow and a passionate kiss can be equally stressful.
Five highly experienced professional pilots performed instrument landing system approaches under simulated instrument flight conditions in a Cessna 172 airplane and in a Link-Singer GAT-1 simulator while under the influence of orally administered secobarbital (0, 100, and 200 mg). Tracking performance in two axes and airspeed control were evaluated continuously during each approach. The data from the airplane and simulator were compared. Error and RMS variability were about half as large in the simulator as in the airplane. The observed data were more strongly associated with the drug level in the simulator than in the airplane. Further, the drug-related effects were more consistent in the simulator. Improvement in performance suggestive of learning effects were seen in the simulator, but not in actual flight. It is concluded that the GAT-1 simulator is a useful and sensitive device for studies of the effects of mild stress on pilot performance, but extrapolation of simulator data to the flight environment must be approached with considerable caution.
The number of fatal accidents involving public transport aircraft has increased significantly in recent years and, because more and more "wide-bodied" aircraft have been coming into service, this has resulted in a rapid increase in the number of fatalities. A combined attack on the problem by all concerned with flight safety is required to improve the situation. The collection and analysis of aircraft accident data can contribute to safety in two ways; by giving an indication of where to concentrate future effort and by showing how successful past efforts have been. An analysis of worldwide accident statistics by phase of flight and causal factor show that the largest percentage of accidents occurs in the approach and landing phase and are caused by "pilot error". Further research is needed to find out why pilots make errors and how such errors can be eliminated.
Although successful collision avoidance is presumed a learned skill, midair collision accidents typically involve highly experienced pilots. It is hypothesized that pilot complacency, arising from a history of incident-free flights, plays a role in the midair collision process. Pilot risk perception is explored using Laplace's Law of Succession, and the resulting perceived collision probability compared to reality. The two converge at around 5,000 hours of flight experience.
General aviation and air taxi approach phase accidents, which occurred during Visual and Instrument Flight Rules (VFR and IFR, respectively) over the last 25 years, were analyzed. The data suggest that there is a 204% higher risk during the approach and landing phase of VFR flights, than during similar IFR operations (14.82 vs. 7.27 accidents/100,000 approaches). Alarmingly, the night single pilot IFR (SPIFR) accident rate is almost 8 times the rate of day IFR, 35.43 vs. 4.47 accidents/100,000 approaches, and two and a half times that of day VFR approaches, 35.43 vs. 14.82 accidents/100,000 approaches. Surprisingly, the overall SPIFR accident rates are not much higher than dual-pilot IFR (DPIFR), 7.27 vs. 6.48 accidents/100,000 approaches. The generally static ratio of the statistics for SPIFR/DPIFR accident rates may be accounted for by little or no change in general aviation cockpit technology during the last 25 years, and because IFR operational flight task management training has not kept pace.