Investigation of occurrence likelihood of human errors in shipping operations

Human errors are one of the most significant contributory factors in ship accidents. This study aims to explore the quantitative relationship between the occurrence likelihood of human errors and external factors including the environmental characteristics, accident characteristics and ship characteristics. A multinomial logistic regression model has been developed to reflect the relationship between these factors and the occurrence likelihood of human errors using 14 years’ ship accident records from Fujian water areas. Model results show that the season of spring, poor visibility and night are more likely to be associated with high occurrence likelihood of negligence and judgment/operation errors. One important finding is that mooring and no strong-wind condition are two circumstances that would highly increase occurrence likelihood of all kinds of human errors. Fishing boats and engineering ships/sand dredgers are the primary ship types which have relatively higher occurrence likelihood of negligence errors than other ship types.