In this study, as an attempt to incorporate Safety-II, which focuses on the continuation of daily stable operations, into safety management, we extracted efforts or actions that are practiced at the sharp end to make things go right (Good Jobs), from safety reports describing events where an accident or a disaster had been successfully avoided or overcome. In particular, we used and analyzed the Voluntary Information Contributory to Enhancement of the Safety (VOICES),which is one of the safety reports utilized widely in Japanese aviation and open to the public for access. We tried to construct a model for extracting Good Jobs from the safety reports in VOICES by text mining. Furthermore, we applied this model to over 1,500 reports, and tried to examine the characteristics of Good Jobs in aviation through cluster analysis. From the results, we were able to break down and deepen our understanding of specific actions regarding the four resilience abilities (anticipating, monitoring, responding, and learning) which are human factor capabilities required in Safety-II.
View full abstract
