Research on mining the causal factors of ship collision accidents based on grounded theory

In recent years, ship collision accidents have been widely concerned due to their high frequency of occurrence and serious losses. However, the analysis of the accident cause, especially the human factors, often lacks the corresponding data support. Therefore, this paper introduces the grounded theory (GT) to conduct the text mining of the accident report, aiming at taking accident as a reference to find out the causes of the ship collision accidents, and obtain the structured framework, identify and evaluate key factors through structural and quantitative methods. Secondly, by comparing HFACS-MA and HFACS-COLL, HFACS-COLL is improved and the results are structured to construct a HFACS framework for ship collision accidents. Finally, the statistical analysis method is used to analyze the correlation and accident-causing factors at all levels. The results show that the causal factors in ship collisions can be represented by a top-down, coarse-grained to fine-grained multi-layered HFACS hierarchy and there are correlations between external factors, organizational influences, unsafe supervision, preconditions for unsafe acts, and unsafe acts. External factors are the potential cause of ship collision accidents and the potential inducement of organizational influences. Negative organizational influences will lead to the genesis of unsafe supervision, form the preconditions of unsafe acts, and eventually lead to the occurrence of unsafe acts. To conclude, the HFACS framework of stand-on vessels and give-way vessels in ship collision accidents was obtained by GT text mining respectively. Combined with WBA and statistical analysis methods, the causes of accidents and the correlation between factors were excavated.