Discovering Anomalous Aviation Safety Events Using Scalable Data Mining Algorithms

Matthews, Bryan; Das, Santanu; Bhaduri, Kanishka; Das, Kunal; Martin, Roscoe C.; Oza, Nikunj C.

doi:10.2514/1.i010080

Cited by 47 publications

(19 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Moreover, how to characterize the temporal structure during various flight phases remains unresolved. Most recently, Matthews et al summarize the knowledge discovery pipeline for aviation data using these algorithms discussed above [10].…”

Section: Introductionmentioning

confidence: 99%

Analysis of Flight Data Using Clustering Techniques for Detecting Abnormal Operations

Das

Hansman

et al. 2015

Journal of Aerospace Information Systems

Self Cite

132

View full text Add to dashboard Cite

The airline industry is moving toward proactive risk management, which aims to identify and mitigate risks before accidents occur. However, existing methods for such efforts are limited. They rely on predefined criteria to identify risks, leaving emergent issues undetected. This paper presents a new method-ClusterAD-Flightwhich can support domain experts in detecting anomalies and associated risks from routine airline operations. The new method, enabled by data from the flight data recorder, applies clustering techniques to detect abnormal flights of unique data patterns. Compared with existing methods, the new method no longer requires predefined criteria or domain knowledge. Tests were conducted using two sets of operational data consisted of 365 B777 flights and 25519 A320 flights. The performance of ClusterAD-Flight was compared with those of Multiple Kernel Anomaly Detection (MKAD), another data-driven anomaly detection algorithm in recent years, and with Exceedance Detection (ED), the current method employed by the airline industry. Results showed that both ClusterAD-Flight and MKAD were able to identify operationally significant anomalies, surpassing the capability of ED. ClusterAD-Flight performed better with continuous parameters, while MKAD was more sensitive towards discrete parameters. Nomenclature v = high-dimensional vector to represent a flight x i j = value of the i th flight parameter at sample time j

show abstract

Section: Introductionmentioning

confidence: 99%

Analysis of Flight Data Using Clustering Techniques for Detecting Abnormal Operations

Das

Hansman

et al. 2015

Journal of Aerospace Information Systems

Self Cite

132

View full text Add to dashboard Cite

show abstract

“…It could be shown that in conjunction with expert knowledge these approaches did produce meaningful results, some of them clearly safety-relevant. Matthews et al (2013) investigated the use of different data mining algorithms to identify safety-relevant occurrences of different varieties in up to 19,243 flights. They found that in collaboration with review pilots it was possible to detect novel threats through this data mining approach.…”

Section: Machine Learning In Flight Data Monitoringmentioning

confidence: 99%

Using machine learning methods in airline flight data monitoring to generate new operational safety knowledge from existing data

Oehling

Barry

2019

Safety Science

View full text Add to dashboard Cite

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.

show abstract

“…One-class SVM [39] is then used to construct a separating hyperplane to detect anomalies. This method was applied to the FOQA dataset [33] and showed high accuracy in discovering operationally significant aviation safety events.…”

Section: Related Workmentioning

confidence: 99%

Vector Autoregressive Model-Based Anomaly Detection in Aviation Systems

Melnyk

Matthews

Valizadegan

et al. 2016

Journal of Aerospace Information Systems

Self Cite

View full text Add to dashboard Cite

Detecting anomalies in datasets, where each data object is a multivariate time series (MTS), possibly of different length for each data object, is emerging as a key problem in certain domains. We consider the problem in the context of aviation safety, where data objects are flights of various durations, and the MTS corresponds to sensor readings. The goal then is to detect anomalous flight segments, due to mechanical, environmental, or human factors. In this paper, we present a general framework for anomaly detection in such settings, by representing each MTS using a vector autoregressive exogenous (VARX) model, constructing a distance matrix among the objects based on their respective VARX models, and finally detecting anomalies based on the object dissimilarities. The framework is scalable, due to the inherent parallel nature of most computations, and can be used to perform online anomaly detection. Experimental results on a real flight dataset illustrate that the framework can detect different types of multivariate anomalies along with the key parameters involved.

show abstract

Discovering Anomalous Aviation Safety Events Using Scalable Data Mining Algorithms

Cited by 47 publications

References 17 publications

Analysis of Flight Data Using Clustering Techniques for Detecting Abnormal Operations

Analysis of Flight Data Using Clustering Techniques for Detecting Abnormal Operations

Using machine learning methods in airline flight data monitoring to generate new operational safety knowledge from existing data

Vector Autoregressive Model-Based Anomaly Detection in Aviation Systems

Contact Info

Product

Resources

About