Research On Human Factors Of Ship Collision Accidents Based On HFACS And BN

The vast majority of cargo transportation in international trade is accomplished by ships sailing at sea.In recent years,with the increasing attention of the world to the ocean,the development of maritime transportation has ushered in a new upsurge.However,frequent ship collision accidents are still major problems for people.In addition,studies show that most ship collision accidents are related to human factors.Nowadays,most of the research work on human factors in ship collision accidents focuses on the unsafe acts and the preconditions for unsafe acts involved in the process of ship collision,and less attention is paid to the underlying causes(such as factors of organizational influences)affecting unsafe acts.In addition,it is a feature of existing studies to put a large number of ship collision accidents together for analysis and to find the causal chain of accidents,but it is ignored that unsafe acts in accidents of different severity may have different inducing paths.Given these above,the main work of this thesis is as follows:(1)Combined the Analytic Hierarchy Process and the Fuzzy Comprehensive Evaluation Method to classify ship collision accidents according to their severity,and 194 ship collision accident reports collected were classified into three grades:very serious accidents,serious accidents,and less serious accidents.And the very serious accidents and serious accidents were merged into group 1 for a small number of samples and less serious accidents as group 2.(2)Introduced the Human Factor Analysis and Classification System(HFACS)and corrected the four levels of HFACS such as organizational impact,unsafe supervision,preconditions for unsafe acts and unsafe acts appropriately according to the characteristics of ship collision accidents,then the human factors of ship collision accidents in two groups were classified and analyzed.After that,the Chi-square test and Odds Ratio analysis of the factors between adjacent levels in the HFACS of two groups of accidents were carried out using SPSS,then obtained the correlation between the factors of adjacent levels.(3)Finally,according to the factors’correlation relationship obtained by SPSS analysis,the Bayesian Network(BN)of two groups of accidents were constructed,and the influence relationship between the upper level’s factors and the unsafe act factors were analyzed,and the most likely inducing paths of the single factor and the whole factors in the unsafe act level of two groups of accidents were obtained.Through comparative analysis,it was found that the factors at the level of unsafe act had the greatest impact on the occurrence of ship collision accidents,while the key factors such as improper emergency measures,imperfect emergency plans and so on would cause accident results worse.In order to reduce the accident prevention cost,prevent the accident results worse,the suggestions of implementing intervention measures from the upper levels of HFACS and strengthening the control of key factors were put forward.