The near miss reports are utilized for preventing accidents and incidents in various industrial fields. The authors analyze countermeasures against both actual near misses and hypothetical ones which crew expected to likely occur.
First, 28,932 countermeasures reported from Japanese domestic tankers are categorized into 11 thinking steps based in 4 step/M (Minimum encounter, Minimum probability, Multiple detection and Minimum damage) by text mining technique.
Secondly, the data is analyzed by Principal Component Analysis; it is shown that the degree of difficulty in each countermeasure taken on tanker is corresponding to the first principal component and spare time up to taking countermeasure is corresponding to the second one. It is clarified that these two components can be used as index to detect vulnerability of countermeasures taken on tankers.
View full abstract