Maritime Accidents and Human Performance: the Statistical Trail

A multi-year project has been undertaken by the American Bureau of Shipping to identify publicly available databases of marine accidents, review the database structures, and analyze the contents. The main objective of the project is to better understand the role of the human in accident causation and consequence mitigation. This knowledge can be used to determine where the American Bureau of Shipping (ABS) and ship designers and operators might direct their efforts with regard to rulemaking, establishing design criteria and standards, planning operations, or directing future ship design or research and development efforts. The technical task objectives of the project are to identify maritime accident and near miss databases and where possible: access the databases and assess the data content and organization; analyze the data for trends and causal patterns related to human performance, error, and accident causation; identify research topics and projects based on accident data and stated or inferred causal factors; support development of accident investigation methods that better recognize human error, produce a yearly report to update marine industry trends regarding incident causation. Findings of this project noted that human error continues to be the dominant factor in maritime accidents and that among all human error types classified in numerous databases and libraries of accident reports, failures of situation awareness and situation assessment overwhelmingly predominate. According to the data analysis, approximately 50% of maritime accidents are initiated by human error, while another 30% of maritime accidents occur due to failures of humans to avoid an accident. In other words, in 30% of maritime accidents, conditions that should have been adequately countered by humans were not.