Effect of HFACS and non-HFACS-related factors on fatalities in general aviation accidents using neural networks.

Dahai Liu, Tiffany Nickens, Leon Hardy, Albert Boquet

Research output: Contribution to journalArticlepeer-review

Abstract

This study applied a backpropagation artificial neural network approach to investigate both the Human Factors Analysis and Classification System (HFACS)-related unsafe act tiers of factors and other non-HFACS factors in an attempt to recognize patterns for general aviation accident fatalities. Data were obtained from the HFACS database and extracted from the National Transportation Safety Board database from 1990 to 2002. Multiple neural network models were created and the best fit model was selected based on a sequence of criteria. A sensitivity analysis was performed on the validated model to rank the factors that lead to general aviation fatalities. Results are discussed and practical implications are given.

Original languageAmerican English
JournalDefault journal
StatePublished - Jan 1 2013

Cite this