Browsing by Author "Barry, David J."
Now showing 1 - 2 of 2
Results Per Page
Sort Options
Item Open Access Estimating runway veer-off risk using a Bayesian network with flight data(Elsevier, 2020-05-23) Barry, David J.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.Item Open Access Using machine learning methods in airline flight data monitoring to generate new operational safety knowledge from existing data(Elsevier, 2019-01-15) Oehling, Julian; Barry, David J.The aim of this work is to investigate the possibility of using machine learning (ML) methods in order to generate novel, safety-relevant knowledge from existing flight data. Airlines routinely generate vast amounts of flight data from routine monitoring, but the concept of extracting safety knowledge from this data is still based on detecting exceedances of expert-defined thresholds. This system is conceptually unable to detect novel occurrences for which no such filters exist. ML techniques are able to close this gap. This paper first reviews the literature to select an appropriate ML method. A form of unsupervised learning called “Local Outlier Probability” is selected. Next, an appropriate feature space is developed and implemented in the flight data monitoring system of a supporting airline to generate the dataset. This dataset is cleaned and the outlier calculation performed. The results are statistically analysed. Furthermore, the top outliers are reviewed by the airline’s review pilots in the same way as the traditional exceedance events. Last, the severities and safety relevance of both types of events are compared. This work successfully shows that the chosen approach is able to reduce the number of undetected safety-relevant occurrences by finding novel occurrence types which were undetected by a contemporary and mature flight data monitoring system. This research builds on recent literature by developing a novel method which can be scaled to work in an airline production environment with large datasets, as demonstrated by the efficient analysis of 1.2 million flights.