Research on Aviation Safety Prediction Based on Variable Selection and LSTM.

Accurate prediction of aviation safety levels is significant for the efficient early warning and prevention of incidents. However, the causal mechanism and temporal character of aviation accidents are complex and not fully understood, which increases the operation cost of accurate aviation safety prediction. This paper adopts an innovative statistical method involving a least absolute shrinkage and selection operator (LASSO) and long short-term memory (LSTM). We compiled and calculated 138 monthly aviation insecure events collected from the Aviation Safety Reporting System (ASRS) and took minor accidents as the predictor. Firstly, this paper introduced the group variables and the weight matrix into LASSO to realize the adaptive variable selection. Furthermore, it took the selected variable into multistep stacked LSTM (MSSLSTM) to predict the monthly accidents in 2020. Finally, the proposed method was compared with multiple existing variable selection and prediction methods. The results demonstrate that the RMSE (root mean square error) of the MSSLSTM is reduced by 41.98%, compared with the original model; on the other hand, the key variable selected by the adaptive spare group lasso (ADSGL) can reduce the elapsed time by 42.67% (13 s). This shows that aviation safety prediction based on ADSGL and MSSLSTM can improve the prediction efficiency of the model while keeping excellent generalization ability and robustness.

Predicting and mitigating failures on the flight deck: an aircraft engine bird strike scenario.

Engine damage as a consequence of foreign object debris (FOD) during flight is frequently caused by birds. One approach to minimising disruption caused by this damage is to provide flight crew with accurate information relating to the continuing operational status of the aircraft's engines. Before designing such avionic systems however, understanding of current procedures is needed. Hierarchical Task Analysis (HTA) and Systematic Human Error Reduction and Prediction Approach (SHERPA) were used to identify potential failures that flight crew may make when managing an engine bird strike. Workshops with commercial pilots generated insights into current practice and a commercial pilot SME reviewed outputs for accuracy. Over 200 potential failures were identified, most commonly related to communication. Remedial measures, considering future avionic systems, are proposed to mitigate identified failures. This analysis provides a starting point for future design concepts for assisting flight crew in dealing with engine malfunction due to FOD strikes. Practitioner summary: Hierarchical Task Analysis was conducted to show all tasks involved in dealing with an in-flight aircraft engine bird strike. Systematic Human Error Reduction and Prediction Approach analysis was performed and over 200 possible failures were identified when managing this event. Remedial measures are proposed to help mitigate possible failures.

Human Factors Contributions to the Development of Standards for Displays of Unmanned Aircraft Systems in Support of Detect-and-Avoid.

OBJECTIVE: The aim is to provide a high-level synthesis of human factors research that contributed to the development of detect-and-avoid display requirements for unmanned aircraft systems (UAS). BACKGROUND: The integration of UAS into the U.S. National Airspace System is a priority under the Federal Aviation Administration's Modernization and Reform Act. For UAS to have routine access to the National Airspace System, UAS must have detect-and-avoid capabilities. One human factors challenge is to determine how to display information effectively to remote pilots for performing detect-and-avoid tasks. METHOD: A high-level review of research informing the display requirements for UAS detect-and-avoid is provided. In addition, description of the contributions of human factors researchers in the writing of the requirements is highlighted. RESULTS: Findings from human-in-the-loop simulations are used to illustrate how evidence-based guidelines and requirements were established for the display of information to assist pilots in performing detect-and-avoid. Implications for human factors are discussed. CONCLUSION: Human factors researchers and engineers made many contributions to generate the data used to justify the detect-and-avoid display requirements. Human factors researchers must continue to be involved in the development of standards to ensure that requirements are evidence-based and take into account human operator performance and human factors principles and guidelines. APPLICATION: The research presented in this paper is relevant to the design of UAS, the writing of standards and requirements, and the work in human-systems integration.

Eye tracking as a debriefing tool in upset prevention and recovery training (UPRT) for general aviation pilots.

Upset prevention and recovery training (UPRT) is intended to improve the ability of pilots to recognize and avoid situations that can lead to airplane upsets and to improve their ability to recover control of an airplane that has exceeded the normal flight envelope. To this end, a set of different training contents - from theoretical knowledge of aerodynamics and human factors to practice-based flight training - is necessary. In order to support the debriefing with an objective feedback, and because visual scanning is a core competence, two studies on subjective evaluation of aviation pilots - one conducted in a flight simulator and the other one in-flight - focussed on the practical application of eye tracking as a debriefing tool in UPRT. From a practitioner's perspective, eye tracking appeared to be a useful method in terms of visualising instrument scanning techniques, supporting the instructor with objective debriefing material and fostering self-awareness in human processes. The discussion recommends adjusted UPRT instructor training and further improvements to eye tracking hardware and software. Practitioner Summary: The article focuses on pilot evaluations of eye tracking as a debriefing tool in UPRT and the identification of critical elements in its use. Eye tracking is a promising debriefing tool for UPRT. The discussion points to desirable improvements of eye tracking hardware and software as well as adjustments to instructor training that are pertinent.

Safety in the nonoperating room anesthesia suite is not an accident: lessons from the National Transportation Safety Board.

PURPOSE OF REVIEW: To review the findings of National Transportation Safety Board-related aviation near misses and catastrophes and apply these principles to the nonoperating room anesthesia (NORA) suite. RECENT FINDINGS: NORA is a specialty that has seen tremendous growth. In 2019, NORA contributes to a larger proportion of anesthesia practice than ever before. With this growth, the NORA anesthesiologist and team are challenged to provide safe, high-quality care for more patients, often with complex comorbidities, and are forced to utilize deeper levels of sedation and anesthesia than ever before. These added pressures create new avenues for human error and adverse outcomes. SUMMARY: Safety in modern anesthesia practice often draws comparison to the aviation industry. From distinct preoperational checklists, defined courses of action, safety monitoring and the process of guiding individuals through a journey, there are many similarities between the practice of anesthesia and flying an airplane. Consistent human performance is paramount to creating safe outcomes. Although human errors are inevitable in any complex process, the goal for both the pilot and physician is to ensure the safety of their passengers and patients, respectively. As the aviation industry has had proven success at managing human error with a dramatic improvement in safety, a deeper look at several key examples will allow for comparisons of how to implement these strategies to improve NORA safety.

Airline Safety Improvement Through Experience with Near-Misses: A Cautionary Tale.

In recent years, the U.S. commercial airline industry has achieved unprecedented levels of safety, with the statistical risk associated with U.S. commercial aviation falling to 0.003 fatalities per 100 million passengers. But decades of research on organizational learning show that success often breeds complacency and failure inspires improvement. With accidents as rare events, can the airline industry continue safety advancements? This question is complicated by the complex system in which the industry operates where chance combinations of multiple factors contribute to what are largely probabilistic (rather than deterministic) outcomes. Thus, some apparent successes are realized because of good fortune rather than good processes, and this research intends to bring attention to these events, the near-misses. The processes that create these near-misses could pose a threat if multiple contributing factors combine in adverse ways without the intervention of good fortune. Yet, near-misses (if recognized as such) can, theoretically, offer a mechanism for continuing safety improvements, above and beyond learning gleaned from observable failure. We test whether or not this learning is apparent in the airline industry. Using data from 1990 to 2007, fixed effects Poisson regressions show that airlines learn from accidents (their own and others), and from one category of near-misses-those where the possible dangers are salient. Unfortunately, airlines do not improve following near-miss incidents when the focal event has no clear warnings of significant danger. Therefore, while airlines need to and can learn from certain near-misses, we conclude with recommendations for improving airline learning from all near-misses.

Hot Air Balloon: An Unusual Cause of Multicasualty Trauma Incident.

Hot air balloon incidents are few and far between compared with the total number of flights. Nevertheless, hot air balloon incidents may produce severe trauma involving several patients and are linked to significant mortality. The prehospital management of injured patients starts after having secured potential surrounding dangers, such as fire or explosion. In the context of a rescue by helicopter, close attention must be paid to potential obstacles, like trees or electrical wires, and the risk of aspiration of the balloon envelope into the rotor. Patients involved in such incidents are often split up in a closed perimeter around the crash point. The severity of the trauma depends essentially on the height of the fall. The most frequent traumatic lesions involve fractures of the lower limbs, the spine, and the pelvis as well as severe burns caused by the balloon fire. Because of the number of patients present, an initial triage is usually required at the site. The use of rescue helicopters can be helpful. They can perform aerial reconnaissance, provide on-site high-level resources, enable access to the patients even in hostile environments, and quickly transport them to trauma center hospitals.

[Shuttle Challenger disaster: what lessons can be learned for management of patients in the operating room?]. / Catastrophe de la navette Challenger: quels enseignements pour la prise en charge des patients au bloc opératoire?

For many years hospitals have been implementing crew resource management (CRM) programs, inspired by the aviation industry, in order to improve patient safety. However, while contributing to improved patient care, CRM programs are controversial because of their limited impact, a decrease in effectiveness over time, and the underinvestment by some caregivers. By analyzing the space shuttle Challenger accident, the objective of this article is to show the potential impact of the professional culture in decision-making processes. In addition, to present an approach by cultural factors which are an essential complement to current CRM programs in order to enhance the safety of care.

The U.S. commercial air tour industry: a review of aviation safety concerns.

The U.S. Title 14 Code of Federal Regulations defines commercial air tours as "flight[s] conducted for compensation or hire in an airplane or helicopter where a purpose of the flight is sightseeing." The incidence of air tour crashes in the United States is disproportionately high relative to similar commercial aviation operations, and air tours operating under Part 91 governance crash significantly more than those governed by Part 135. This paper reviews the government and industry response to four specific areas of air tour safety concern: surveillance of flight operations, pilot factors, regulatory standardization, and maintenance quality assurance. It concludes that the government and industry have successfully addressed many of these tenet issues, most notably by: advancing the operations surveillance infrastructure through implementation of en route, ground-based, and technological surveillance methods; developing Aeronautical Decision Making and cue-based training programs for air tour pilots; consolidating federal air tour regulations under Part 136; and developing public-private partnerships for raising maintenance operating standards and improving quality assurance programs. However, opportunities remain to improve air tour safety by: increasing the number and efficiency of flight surveillance programs; addressing pilot fatigue with more restrictive flight hour limitations for air tour pilots; ensuring widespread uptake of maintenance quality assurance programs, especially among high-risk operators not currently affiliated with private air tour safety programs; and eliminating the 25-mile exception allowing Part 91 operators to conduct commercial air tours without the safety oversight required of Part 135 operators.

Effects of distance flown on pilot decision making in continued flight into deteriorating weather conditions.

INTRODUCTION: Continuing flight into adverse weather remains a significant problem in general aviation (GA) safety. A variety of experiential, cognitive, and motivational factors have been suggested as explanations. Previous research has shown that adverse weather accidents occur further into planned flights than other types of accident, suggesting that previous investment of time and effort might be a contributing factor. The aim of this study was to experimentally determine the effect of prior commitment on general aviation pilots' decision-making and risk-taking in simulated VFR flights. METHOD: Thirty-six licensed pilots 'flew' two simulated flights designed to simulate an encounter with deteriorating coastal weather and a developing extensive cloud base underneath the aircraft as it crossed a mountain range. After making a decision to continue or discontinue the flight, pilots completed a range of risk perception, risk taking, and situational awareness measures. RESULTS: Visual flight rules were violated in 42% of the flights. Prior commitment, in terms of distance already flown, led to an increased tendency to continue the flight into adverse weather in the coastal 'scud running' scenario. Continuing pilots perceived the risks differently and showed greater risk tolerance than others. These 'bolder' pilots also tended to be more active and better qualified than the others. CONCLUSIONS: There are undoubtedly multiple factors underlying any individual decision to continue or discontinue a flight. The willingness to tolerate a higher level of risk seems to be one such factor. This willingness can increase with time invested in the flight and also seems to be related to individual flight qualifications and experience. PRACTICAL APPLICATIONS: All pilots might benefit from carefully structured simulator sessions designed to safely teach practical risk management strategies with clear and immediate feedback.

The Untapped Potential of Narrative as a Tool in Aviation Mental Health and Certification.

INTRODUCTION: Work-related stress is common in pilots, with broad implications, including the potential development of mental health symptoms and sometimes even psychiatric disease. This commentary argues for the use of narrative as a tool to promote preventive health behaviors in pilots and combat misinformation about aeromedical certification related to mental health.Hoffman WR, McNeil M, Tvaryanas A. The untapped potential of narrative as a tool in aviation mental health and certification. Aerosp Med Hum Perform. 2024; 95(3):165-166.

Identifying training deficiencies in military pilots by applying the human factors analysis and classification system.

Without accurate analysis, it is difficult to identify training needs and develop the content of training programs required for preventing aviation accidents. The human factors analysis and classification system (HFACS) is based on Reason's system-wide model of human error. In this study, 523 accidents from the Republic of China Air Force were analyzed in which 1762 human errors were categorized. The results of the analysis showed that errors of judgment and poor decision-making were commonly reported amongst pilots. As a result, it was concluded that there was a need for military pilots to be trained specifically in making decisions in tactical environments. However, application of HFACS also allowed the identification of systemic training deficiencies within the organization further contributing to the accidents observed.

Analysis on coupling dynamic effect of human errors in aviation safety.

Human factors have increasingly been the leading cause of aircraft accidents. In most cases, human factors are not working alone, instead they are coupled with complex environment, mechanical factors, physiological and psychological factors of pilots, and organizational management, all of which form a complex aviation safety system. It is vital to investigate the coupling impact of human errors to avoid the occurrence of aviation accidents. In view that the Human Factors Analysis and Classification System (HFACS) provides a hierarchical classification principle of human errors in aviation accidents, and the System Dynamics (SD) approach is helpful to describe the risk evolution process, this paper establishes a hybrid HFACS-SD model by employing the HFACS and the SD approach to reveal the aviation human factors risk evolution mechanism, in which the HFACS is first used to capture the causal factors of human errors risk, and a coupling SD model is then built to describe the evolution of aviation human factors risk supported by historical data. The eigenvalue elasticity analysis is taken to identify critical loops and parameters that have a substantial impact on the system structural behavior, and the influence of parameters and loops is assessed. Simulation results show that the evolution trend of the accident rate can be replicated by the proposed HFACS-SD model, and the structural dominance analysis can efficiently identify critical loops and parameters. Simulation results further show that, with the recommended safety enhancement measures, the stability of the aviation system is increased, and thus lowering the overall accident rate.

Cause analysis of unsafe acts of pilots in general aviation accidents in China with a focus on management and organizational factors.

Objectives. General aviation (GA) safety has become a key issue worldwide and pilot errors have grown to be the primary cause of GA accidents. However, fewer empirical studies have examined the contribution of management and organizational factors for these unsafe acts. Flawed decisions at the organizational level have played key roles in the performance of pilots. This study provides an in-depth understanding of the management and organizational factors involved in GA accident reports. Methods. A total of 109 GA accidents in China between 1996 and 2021 were analysed. Among these reports, pilot-related accidents were analysed using the human factors analysis and classification system (HFACS) framework. Results. The significant effects of managerial and organizational factors and the failure pathways on GA accidents have been identified. Furthermore, unlike traditional HFACS-based analyses, the statistically significant relationships between failures at the organizational level and the sub-standard acts of the pilots in GA accidents were revealed. Conclusions. Such findings support that the GA accident prevention strategy that attempts to reduce the number of unsafe acts of pilots should be directed to the crucial causal categories at HFACS organizational levels: resource management, organizational process, failure to correct a known problem, inadequate supervision and supervisory violations.

Pilot performance and workload whilst using an angle of attack system.

Loss of control in flight is the primary category of fatal accidents within all sectors of aviation and failure to maintain adequate airspeed - leading to a stall - is often cited as a causal factor. Stalls occur when the critical angle of the aircraft is exceeded for a given airspeed. Using airspeed as an indicator of the potential to stall is an unreliable proxy. Systems that measure the angle of attack have been routinely used by military aircraft for over 50 years however rigorous academic research with respect to their effectiveness has been limited. Using a fixed-base flight simulator fitted with a simulated, commercially available angle of attack system, 20 pilots performed normal and emergency procedures during the circuit/pattern in a light aircraft. Experimental results have shown that pilot performance was improved when angle of attack was displayed in the cockpit for normal and emergency procedures during the approach phase of flight in the pattern/circuit. In relation to pilot workload, results indicated that during the approach phase of flight, there was a moderate but tolerable increase in pilot workload. The use of such a display may assist pilots to maintain the aircraft within the optimum range and hence reduce occurrences of unstable approaches. Overall, fewer stall events were observed when angle of attack was displayed and appropriate pilot decisions made during emergencies. These results provide a new perspective on pilot workload and aviation safety.

Prediction of severity of aviation landing accidents using support vector machine models.

The purpose of this study was to apply support vector machine (SVM) models to predict the severity of aircraft damage and the severity of personal injury during an aircraft approach and landing accident and to evaluate and rank the importance of 14 accident factors across 39 sub-categorical factors. Three new factors were introduced using the theory of inattentional blindness: The presence of visual area surface penetrations for a runway, the Federal Aviation Administration's (FAA) visual area surface penetration policy timeframe, and the type of runway approach lighting. The study comprised 1,297 aircraft approach and landing accidents at airports within the United States with at least one instrument approach procedure. Support vector machine models were developed in using the linear, polynomial, radial basis function (RBF), and sigmoid kernels for the severity of aircraft damage and additional SVM models were developed for the severity of personal injury. The SVM models using the RBF kernel produced the best machine learning models with a 96% accuracy for predicting the severity of aircraft damage (0.94 precision, 0.95 recall, and 0.95 F1-score) and a 98% accuracy for predicting the severity of personal injury (0.99 precision, 0.98 recall, and 0.99 F1-score). The top predictors across both models were the pilot's total flight hours, time of the accident, pilot's age, crosswind component, landing runway number, single-engine land certificate, and any obstacle penetration. This study demonstrates the benefit of SVM modeling using the RBF kernel for accident prediction and for datasets with categorical factors.