Flight Operations Quality Assurance Research Articles

Deep Embedding Koopman Neural Operator-Based Nonlinear Flight Training Trajectory Prediction Approach

Accurate flight training trajectory prediction is a key task in automatic flight maneuver evaluation and flight operations quality assurance (FOQA), which is crucial for pilot training and aviation safety management. The task is extremely challenging due to the nonlinear chaos of trajectories, the unconstrained airspace maps, and the randomization of driving patterns. In this work, a deep learning model based on data-driven modern koopman operator theory and dynamical system identification is proposed. The model does not require the manual selection of dictionaries and can automatically generate augmentation functions to achieve nonlinear trajectory space mapping. The model combines stacked neural networks to create a scalable depth approximator for approximating the finite-dimensional Koopman operator. In addition, the model uses finite-dimensional operator evolution to achieve end-to-end adaptive prediction. In particular, the model can gain some physical interpretability through operator visualization and generative dictionary functions, which can be used for downstream pattern recognition and anomaly detection tasks. Experiments show that the model performs well, particularly on flight training trajectory datasets.

An Exploratory Study: Evaluation of the Customer Happiness Based on the Administrative Services Provided by A Low-Cost Airline

The current research as presented in this report, was carried out on a selected organization, which is Fly Dubai Airlines, operating in the United Arab Emirates (UAE). The aim of the study was to analyse whether the company is offering high quality of administrative services, and how well the services are satisfying and making its customers happy. The study attempted to determine the effectiveness of the administrative services suggest improvement its services over time. A mixed methods research approach was considered for the study, based on which a combination of the qualitative and quantitative data were used. Primary and secondary data collection tools were combined. Primary surveys were conducted with employees of Fly Dubai, including employees in different positions in the company. Of the 100 selected members, 30 responded to the questions that were asked, and their responses were analysed based on a comparative analysis technique by plotting the responses on graphs and charts. Secondary sources such as journals, researched publications, books, and articles were also used to obtain information on the current state of quality performance in the airline sector in UAE. It could be obtained from the findings that Fly Dubai has its focus significantly placed on the factor of quality. High service quality is offered by the company, and the company continuously makes efforts to improve its service quality as well. The quality has been improved over time, and the company follows the ISO standards of quality for measuring the quality levels being provided by the company. Quality Management Systems, Safety Management Systems, and the Flight Data Monitoring (FDM) and Flight Operational Quality Assurance (FOQA) could be found as significantly followed by the company. Customers and their satisfactions are considered as extremely important by the company, and the quality measures of the company are therefore targeted towards achieving improved levels of customer satisfaction. The findings also revealed that there is an overall improvement in quality considered in the airline sector in the region. Quality service could be found as one of the key factors leading to the success of the company too, and hence significant importance being given by the management. Also, a positive link could be found to be existing between the quality of service and operational efficiency of the company, with operations influencing the financial performance of the company as well. At the same time, it could be obtained that the company has scope of improvement too, to consider in its quality approaches, to identify, address and fulfil specific quality preferences of its customers and employees. Accordingly, certain recommendations have been suggested to the company. It has been suggested that the company gives important to every customer and every employee, and try to identify their needs and quality demands, which would help the company to achieve overall quality improvement in its services targeting long term effectiveness of the processes. Also, the company has been recommended to consider an integrated organization-wide data quality management and information governance, as well as focus on continuous improvement in its operations to achieve improved service quality on a continuous basis.

A Study on the Flight Operations Quality Assurance System for Predictive Safety Management of the Republic of Korea Air Force

A Study on the Flight Operations Quality Assurance System for Predictive Safety Management of the Republic of Korea Air Force

Semi-Supervised Active Learning for Anomaly Detection in Aviation

Anomaly detection in commercial aviation is an extremely challenging yet crucial task. Accurately detecting operationally significant anomalies can save civilian lives and/or result in significant savings in maintenance cost. The current practice uses manually tuned rule-based mechanisms to flag exceedances from predefined safety boundaries. However, this system cannot identify unknown risks and emerging vulnerabilities. Recently, innovative approaches based on machine learning have been used to automate anomaly detection. However, there are limits to their applicability in the field of aviation due to several challenges: 1) Properly reviewed data are scarce in aviation and, as a result, supervised learning cannot reach optimal performance. 2) Operationally significant anomalies do not coincide with statistically significant ones and, as a result, unsupervised learning fails to provide reliable and robust performance. In this paper, we propose a semi-supervised active learning framework for anomaly detection (SALAD), which detects operationally significant anomalies in flight operational quality assurance data. The developed framework works with vast amounts of unlabeled data as well as a small quantity of labeled data reviewed by subject matter experts to reliably identify safety anomalies in flight operations. Moreover, the model’s active learning strategy allows it to detect unknown anomalies that might emerge in the system. We validate the performance of the SALAD with a real-world case study of anomaly detection during the approach to landing of commercial aircraft. We specifically show that the proposed framework reaches reliable performance when only 1% of the data is labeled and can identify unknown anomalies effectively.

Irregular deviation of flight control surface monitoring for jet transport aircraft

PurposeThe purpose of this paper is to present the irregular deviation examination of flight control surfaces and the potential problem diagnosis of irregular deviations for the jet transport aircraft. A four-jet transport aircraft at transonic flight in cruise phase is the study case of the present article.Design/methodology/approachThe standard lift-to-drag ratio (L/D) and flight dynamic models are established through flight data mining and the fuzzy logic modeling technique based on the flight data of quick access recorder available in the Flight Operations Quality Assurance (FOQA) program of the airlines. The irregular deviations of flight control surfaces are examined by the standard L/D model-predicted results through sensitivity analysis. The contribution values in L/D deficiency are predicted by the deviations and the L/D derivatives of all influencing variables in Taylor series expansion. The potential problems due to irregular deviations can be excavated by the flight dynamic models through the analysis of in-flight stability and controllability.FindingsThe magnitude of stabilizer angle to the deficiency of L/D is the largest among the four control surfaces and elevator is the second one through the judgment of contribution values in L/D deficiency. The stabilizer has irregular deviations with obvious endplay problems of jackscrew, as found in the present study. The stabilizer is suggested to have the unscheduled maintenance for the flight control rigging.Research limitations/implicationsThe specific transport aircraft of the standard L/D model should be the best one in L/D performance among all transport aircraft in the fleet of the airlines. The present method is a new concept to monitor the irregular deviation of flight control surface. The study case of the four-jet transport aircraft at transonic flight in cruise phase is illustrated as the standard L/D mode. The required flight data of monitored flight is requested to eliminate the biases through compatibility checks. The flight data of study case in the present study is also illustrated as monitored flight data.Practical implicationsTo diagnose the irregular deviations of flight control surface deflected angles with contributing to the L/D deficiency estimation is an innovation to improve the flight data analysis of FOQA program for airlines. If the irregular deviation problems of control surfaces can be fixed after rigging in maintenance, the goal of flight safety and aviation fuel saving will be achieved.Social implicationsThe flight control surface rigging of unscheduled maintenance is not expected to coincide with an airline’s peak season or unavailable space in hangar. The optimal time of unscheduled maintenance for the flight control rigging will be easily decided through the correlations between excessive fuel cost and flight safety.Originality/valueThis method can be used to assist airlines to monitor irregular angular positions of flight control surfaces as a complementary tool for management to improve aviation safety, operation and operational efficiency.

Estimating runway veer-off risk using a Bayesian network with flight data

Risk assessments in airline operations are mostly qualitative, despite abundant data from programmes such as flight data monitoring (FDM) and flight operations quality assurance (FOQA). In this paper, features relating to runway excursion causal factors are extracted from flight data from over 310,448 flights from Airbus A320 series aircraft flown on a European network. The data is combined with meteorological data to provide additional features.Bayesian networks are then learnt from the feature set, and two network learning algorithms are compared, Bayesian Search and Greedy Thick Thinning (GTT). Cross-validation of the resulting networks shows both algorithms produce similarly performing networks, and a subjective analysis concludes that the GTT algorithm is marginally preferred.The resulting networks produce relative probabilities, which airlines can use to quantitatively assess runway veer-off risk under different scenarios, such as different meteorological conditions and unstable approaches.This paper's main finding is that by utilising existing data sources, such as FDM and weather databases, airlines can create and use Bayesian networks alongside their existing qualitative risk assessment methods to provide quantitative risk assessment and understand the effect of different conditions on those risks. This is not possible with current methods in use by airlines.The method described can be extended to other operational safety risks, such as runway overrun.

Deep Spatio-Temporal Neural Networks for Risk Prediction and Decision Support in Aviation Operations

AbstractThe maintenance and improvement of safety are among the most critical concerns in civil aviation operations. Due to the increased availability of data and improvements in computing power, applying artificial intelligence technologies to reduce risk in aviation safety has gained momentum. In this paper, a framework is developed to build a predictive model of future aircraft trajectory that can be utilized online to assist air crews in their decision-making during approach. Flight data parameters from the approach phase between certain approach altitudes (also called gates) are utilized for training an offline model that predicts the aircraft’s ground speed at future points. This model is developed by combining convolutional neural networks (CNNs) and long short-term memory (LSTM) layers. Due to the myriad of model combinations possible, hyperband algorithm is used to automate the hyperparameter tuning process to choose the best possible model. The validated offline model can then be used to predict the aircraft’s future states and provide decision-support to air crews. The method is demonstrated using publicly available Flight Operations Quality Assurance (FOQA) data from the National Aeronautics and Space Administration (NASA). The developed model can predict the ground speed at an accuracy between 1.27% and 2.69% relative root-mean-square error. A safety score is also evaluated considering the upper and lower bounds of variation observed within the available data set. Thus, the developed model represents an improved performance over existing techniques in literature and shows significant promise for decision-support in aviation operations.

Critical Parameter Identification for Safety Events in Commercial Aviation Using Machine Learning

In recent years, there has been a rapid growth in the application of data science techniques that leverage aviation data collected from commercial airline operations to improve safety. This paper presents the application of machine learning to improve the understanding of risk factors during flight and their causal chains. With increasing complexity and volume of operations, rapid accumulation and analysis of this safety-related data has the potential to maintain and even lower the low global accident rates in aviation. This paper presents the development of an analytical methodology called Safety Analysis of Flight Events (SAFE) that synthesizes data cleaning, correlation analysis, classification-based supervised learning, and data visualization schema to streamline the isolation of critical parameters and the elimination of tangential factors for safety events in aviation. The SAFE methodology outlines a robust and repeatable framework that is applicable across heterogeneous data sets containing multiple aircraft, airport of operations, and phases of flight. It is demonstrated on Flight Operations Quality Assurance (FOQA) data from a commercial airline through use cases related to three safety events, namely Tire Speed Event, Roll Event, and Landing Distance Event. The application of the SAFE methodology yields a ranked list of critical parameters in line with subject-matter expert conceptions of these events for all three use cases. The work concludes by raising important issues about the compatibility levels of machine learning and human conceptualization of incidents and their precursors, and provides initial guidance for their reconciliation.

Evaluation of Low-Cost Solutions to Support Flight Operations Quality Assurance

During the period 2003–2011, worldwide accident rates for business/corporate aviation were nearly four times that for commercial aviation. Flight Operations Quality Assurance (FOQA) or Flight Data Monitoring—the collection of real-time flight data for continuous safety improvement—has been routinely used by commercial aviation for over 50 years. Regulatory authorities only require airplanes over 27 tonnes maximum takeoff weight to operate an FOQA Program. Corporate/business aviation operators generally operate airplanes less than 20 tonnes using a diverse range of airplanes, which may include a single engine, twin engine, turbo-prop, and/or very light jets. This study investigates the use of low-cost, independent commercial off-the-shelf technologies using a combination of Global Positioning Systems and Attitude Heading & Referencing Systems to sense and record key flight data parameters in support of an FOQA program for corporate operators with diverse, lower-weight-category and legacy airplanes. The preliminary results suggest that such systems and Electronic Flight Information Systems (where installed) can usefully support an FOQA for lower-weight-category and legacy airplanes.

Research on Flight Operational Quality Assurance Based on ROM Concept

The flight operational quality management method based on Routine Operational Measure (ROM) can visually find hidden insecurity factors during flight operation and provide pilots directions in enhancing technologies and risk warning by completing collection off light segment data of aircrafts in routine operations and carrying out the analysis and research on flight data. Based on the theoretical analysis and the needs of airlines, this paper designs and implements the flight operational quality management system based on ROM so that the flight operational quality management of civil aviation of China can realize its critical transformation from a fleet to an individual and from ex-post to ex-ante.

A Method of Applying Flight Data to Evaluate Landing Operation Performance

Pilots’ operation has an important effect on flight safety and performance, particularly in the final landing stage when pilots need to deal with complicated operations. This study aims to determine the potential value of flight data and develop a method of evaluating a pilot’s performance during landing phase based on flight quick access recorder (QAR) data from the perspective of risk assessment. First, a Landing Operation Performance Evaluation Model was developed based on risk evaluation principles. Three landing parameters, which are touchdown distance, touchdown vertical acceleration and touchdown pitch angle, were selected as indicators to evaluate the pilots’ landing operation performance in this model. Second, the flight landing operation performance evaluation system (FLOPES) was set up based on the evaluation model. Test results showed that FLOPES can accomplish all calculation flow of operation performance evaluation. Finally, it concluded that this method is a more accurate and effective way for evaluating the landing operation performance of a flight. It could be as a practical tool for airlines to manage landing risk quantitatively and to provide a more practical support for improving training and design in aviation.Practitioner summary: This study aims to determine the potential value of flight data and to develop a method of evaluating pilot’s landing operation performance from the risk evaluation perspective. Test results showed that this method is effective and could be as a practical tool for airlines to manage landing risk and improve training.Abbreviations:QAR: Quick Access Recorder; FLOPES: Flight Landing Operation Performance Evaluation System; ICAO: International Civil Aviation Organization; IATA: International Air Transport Association; SMS: Safety Management System; CAAC: Civil Aviation Administration of China; FOQA: Flight Operations Quality Assurance; VBA: Visual Basic for Applications

Exploring ADS-B as an Alternative Data Source for Flight Data Monitoring of General Aviation

Flight data monitoring (FDM) is believed to be effective for mitigating risks of aviation accidents by the International Civil Aviation Organization (ICAO) and major airlines who implemented flight data monitoring programs. In the United States, flight data monitoring is recommended by the Federal Aviation Administration (FAA) as a voluntary safety program, namely the flight operational quality assurance (FOQA) program. However, implementing a FDM program not only requires an expensive investment on technological equipment, but also involves long-term labor costs to regularly collect and analyze flight data after flights are completed. Especially for general aviation (GA), implementation of a FDM program using on-board flight data recorder equipment may be too expensive for some owners or operators. Automatic Dependent Surveillance Broadcast Out (ADS-B Out) is a precise satellite-based surveillance system which can continually broadcast flight data. The broadcasted flight data can be received by other nearby ADS-B In capable aircraft or ground-based ADS-B receivers. This paper explores the ADS-B as an alternative data source of FDM for GA. First, the structure and content of flight data broadcast by ADS-B Out are analyzed. Then, based on the basic flight parameters, additional flight parameters are derived, and flight metrics are developed from the standpoint of flight operation analysts. Finally, the potential of ADS-B for supporting FDM is discussed.

Data-Driven Modeling and Analysis Framework for Cockpit Human–Machine Interaction Issues

As the level of automation within an aircraft increases, the interactions between the pilot and autopilot play a crucial role in its proper operation. Issues with human–machine interactions have been cited as one of the main causes behind many aviation accidents. The complex nature of a pilot’s interactions with an autopilot system makes it challenging to identify all possible actions that may lead to an human–machine interaction-related incident. In this paper, a data-driven analysis tool is proposed to detect and categorize potential human–machine interaction issues in large-scale flight operational quality assurance datasets. The proposed tool is developed using a multilevel clustering framework, where a set of basic clustering techniques are combined with a consensus-based approach to group human–machine interaction events and create a data-driven model from the flight operational quality assurance data. The proposed framework is able to effectively compress a large dataset into a small set of representative clusters within a data-driven model, enabling subject matter experts to effectively investigate potential human–machine interaction issues.

HFOQA 이벤트 분석과 SMS를 통한 국내 회전익항공기 체크리스트 개발

According to various researches, one of main causes of flight accidents is influence of human factors. It is important to collect and analyze event data in advance to minimize and prevent human errors by developing safety checklist. As a result international aviation industry developed Flight Operations Quality Assurance for commercial airliners, and, upon its proven effectiveness, it is now highly recommended to adopt for helicopter operations. Consequently, S government organization instituted Helicopter Flight Operations Quality Assurance system in 2012. Hence, this study conducted regression analysis of S organization's flight event frequency and applied ICAO SMS Matrix to categorize the severity of the events according to international standards. Based on the analysis, this research derived fundamental checklist factors for helicopter, which can be a proactive measure for safe operation.

Examination of excessive fuel consumption for transport jet aircraft based on fuzzy-logic models of flight data

Because the amount of fuel usage is huge, and exhaust emissions of commercial transports is regulated under the Kyoto Protocol, how to reduce fuel consumption is an urgent problem for all airlines. The airlines urgently need a practical method to reduce fuel consumptions within constraints to accomplish the mission. The objective for the present paper is to use a model-based method and flight data to enhance fuel efficiency for commercial transports. The fuzzy-logic modeling (FLM) technique will be employed to establish the reference lift-to-drag (L/D) model. The model is utilized to predict the deficiencies of lift-to-drag ratio through sensitivity analysis to determine the relative contributions from influencing flight variables for the excessive fuel consumption. This method is based on the flight data in quick access recorder (QAR) available in the Flight Operations Quality Assurance (FOQA) program. The process of identifying how flight variables reduce the lift-to-drag ratio of a twin-jet transport aircraft in cruise flight will be presented. The sensitivity derivatives for slow-varying and fast-varying influence variables are analyzed. The factors contributing to the degradation of aerodynamic efficiency (i.e. L/D) in operations are evaluated. The excessive fuel consumption due to the effects of deficiencies will be estimated in the present paper. If all the deficiencies can be improved, the fuel efficiency enhancement will be achieved.

Optimizing the Event Set for Collegiate Aviation FOQA Programs

Flight Operations Quality Assurance or FOQA is a proven tool in the effort to enhance aviation safety. Employed by air carriers as early as the 1960’s, FOQA allows aviation operations and safety managers to objectively monitor how their aircraft are being operated. This data can then be translated into informed decisions to improve the safety and efficiency of the overall operation. While FOQA has proven itself in the world of airlines and other commercial aviation ventures, these flight data monitoring programs have largely been absent in the area where the vast majority of flight operations occur, general aviation. Advancements in technology allow those in general aviation management positions the ability to apply the same techniques in general aviation. This study attempts to further the body of knowledge of Flight Operations Quality Assurance (FOQA) programs by examining the unique requirements of a FOQA program adapted to the university flight-training environment. The methodology employed are qualitative in nature employing a Delphi study to gather data from a group of subject matter experts on both FOQA and general aviation flight instruction. Qualitative observations gathered by the researcher from the direct observation of flight instruction will be used to supplement the data gathered from the Delphi Study. Through the analysis of these two data sets, this study determines what events and parameters should be monitored in a collegiate FOQA program.

Discovering Anomalous Aviation Safety Events Using Scalable Data Mining Algorithms

The worldwide civilian aviation system is one of the most complex dynamical systems created. Most modern commercial aircraft have onboard flight data recorders that record several hundred discrete and continuous parameters at approximately 1 Hz for the entire duration of the flight. These data contain information about the flight control systems, actuators, engines, landing gear, avionics, and pilot commands. In this paper, recent advances in the development of a novel knowledge discovery process consisting of a suite of data mining techniques for identifying precursors to aviation safety incidents are discussed. The data mining techniques include scalable multiple-kernel learning for large-scale distributed anomaly detection. A novel multivariate time-series search algorithm is used to search for signatures of discovered anomalies on massive datasets. The process can identify operationally significant events due to environmental, mechanical, and human factors issues in the high-dimensional flight operations quality assurance data. All discovered anomalies are validated by a team of independent domain experts. This novel automated knowledge discovery process is aimed at complementing the state-of-the-art human-generated exceedance-based analysis that fails to discover previously unknown aviation safety incidents. In this paper, the discovery pipeline, the methods used, and some of the significant anomalies detected on real-world commercial aviation data are discussed.

Modern Airline Pilots' Quandary: Standard Operating Procedures—to Comply or Not to Comply

Modern airline pilots are tasked every flight with the safe and efficient operation of highly automated airliners in today’s complicated global and economic environments. Airlines have developed standard operating procedures (SOP) for normal, abnormal, and emergency operations. These procedures serve as a script for crews to follow. These procedures are designed by airlines to ensure that aircraft are operated in the (1) most safe, (2) most efficient, and (3) most on-time manner. For the most part pilots will comply with SOP, but when they (1) don9t agree with SOP, (2) don9t understand SOP or the risks associated with not complying with SOP, or (3) don9t feel adequately trained to know what SOP is, it is difficult to motivate them to comply. Airlines have the means to measure compliance through Flight Operations Quality Assurance (FOQA) and Line Operations Safety Audit (LOSA). The purpose of this research is to determine if increased understanding, knowledge and awareness of the risk of noncompliance with SOP increase airline pilots’ compliance with SOP. This research explores data from line checks at a major US airline that was gathered in pursuit of understanding what drives SOP compliance. Baseline data was gathered and analyzed to determine the top 12 noncompliant items. The airline provided training during the Human Factors module in each pilots recurrent training on Pilot Intentional Non Compliance (PINC). The training including developing pilots’ understanding that while most Aviation Safety Action Program (ASAP) reports grant pilots immunity from legal action, if a violation is labeled PINC, ASAP protections do not apply. Further line checks were conducted after the pilots received the PINC training. The top 12 noncompliant items from the pre-PINC training group were compared to the same 12 items in the post-PINC training group. Significant improvement in SOP compliance was found in six of the 12 items tested. The results established that training pilots on the risk of PINC did significantly increase SOP compliance.

Performance Diagnosis for Turbojet Engines Based on Flight Data

The purpose of the current study is to improve the flight data analysis in the existing program of flight operations quality assurance (FOQA) of the airlines. The goal is to detect any potential problems related to engine health. The exhaust gas temperature (EGT) is the primary parameter in performance monitoring and diagnosis. Performance reference models are obtained through fuzzy-logic modeling. As the first example in the present paper, the potential problems and abnormal conditions can be diagnosed after a comparative analysis with the reference model for a four-engine jet freighter. This paper also uses a twin-jet passenger transport as the second example to illustrate the concept of relative efficiency for engine components. The sensitivity derivatives of EGT with respect to the operational variables indicate that the performance of these two engines has been declining because of aging.

Safety Risk Analysis in Flight Operations Quality Assurance

Flight data recorded by Quick Access Recorder (QAR) is gradually used to analyze pilot's operations quality, especially for the QAR exceedances. In this paper, we use the values of flight parameters to construct the state-space of pilots’ operations. A flight is seen as a point in it. Basing on the distributions of normal flights and flights triggering an exceedance, risk analysis is carried out to divide this state-space into dangerous and safety regions. A multi-objective programming is established to analyze the safety risk in pilot's operations and we use Particle Swarm Optimization (PSO) to find the boundaries of regions-of-interest. The feasibility of this method is validated through experiments with real flight data.