Title: Qualitative classification of descent phases in commercial flight data

Authors: Edward Smart, Honghai Liu, Chris Jesse, David Brown

Addresses: Institute of Industrial Research, University of Portsmouth, PO1 3QL, England, UK. ' Institute of Industrial Research, University of Portsmouth, PO1 3QL, England, UK. ' Institute of Industrial Research, University of Portsmouth, PO1 3QL, England, UK. ' Institute of Industrial Research, University of Portsmouth, PO1 3QL, England, UK

Abstract: Flight data from commercial aircraft in the descent is analysed using one-class classification techniques to identify possible unstable approaches. The method considers snapshots of flight parameters at certain heights in the descent and identifies any abnormalities. Combination rules are used to analyse the flight as a whole and it is found that it is possible to detect abnormal flights with a good degree of accuracy.

Keywords: flight safety; novelty detection; Gaussian mixture; MOG; minimum spanning tree; combination rules; computational intelligence; flight data; commercial aircraft; aircraft descent; qualitative classification; descent abnormalities; abnormal flights.

DOI: 10.1504/IJCISTUDIES.2009.025337

International Journal of Computational Intelligence Studies, 2009 Vol.1 No.1, pp.37 - 49

Published online: 19 May 2009 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article