A Natural Language Processing Model for Analyzing Aviation Safety Event Reports: A Subset of Results
Document Type
Article
Publication Date
5-31-2023
Abstract
Many civil aviation authorities, operators, and manufacturers utilize voluntary safety reporting programs (VSRPs) to understand risk within their operations. Insights from these first-hand accounts can lead to significant safety and efficiency improvements. Subject matter experts often read and analyze these reports by labeling factors of interest to derive safety insights. The resources required for this analysis can limit the insights an organization can obtain from their VSRP data. A novel machine learning model was developed and trained on over 50,000 rows of manually labeled aviation VSRP data. This model uses machine learning and natural language processing (NLP) to automate the task of labeling aviation safety reporting data and codifying report narratives according to a structured list of human factors topics. This paper presents a subset of interim model results and discusses the implications of using NLP to identify reports citing human factors topics from aviation VSRP data.
Repository Citation
Hinson, R. J.,
Bynum, E.,
Kinsella, A.,
Berry, K.,
& Sawyer, M.
(2023). A Natural Language Processing Model for Analyzing Aviation Safety Event Reports: A Subset of Results. 22nd International Symposium on Aviation Psychology, 27.
https://corescholar.libraries.wright.edu/isap_2023/23
