Search

The safety of crude oil tankers is always a focus under the rapid development of oil trade and maritime transportation. The purpose of this study is to investigate the critical factors that lead to crew casualties and address the hot spot distribution characteristics by reviewing accident data over the last 26 years. A Zero-Inflated Negative Binomial (ZINB) regression model is introduced to evaluate the extent of eight selected factors on crew mortality. Three others regression models (Poisson, Negative Binomial, and Zero-Inflated Poisson regression model) are chosen for model comparison to demonstrate the superiority of the ZINB model. The results showed that the ZINB model outperforms the other regression models in terms of fitting performance. The leading contributors of the crude oil tanker accident casualties were found as foundered and fire/explosion, followed by war loss/hostilities, collision, weather conditions, classification society, hull/machinery damage, and ship size. Furthermore, the Kernel Density Estimation (KDE) methodology is adopted to identify hot spots and their trend evolution features of critical accident factors. Crude oil tanker accidents were observed more likely to occur in the waters around the United Kingdom, Gulf of Mexico, Strait of Malacca & Singapore Strait, Mediterranean Sea, China, Japan & Korea, and the Gulf Area, which may be due to the dense shipping routes, poor geographical conditions, and adverse weather conditions. The results of this study provide conducive suggestions for those policy-makers to develop appropriate and effective strategies for preventing crude oil tanker accidents. [ABSTRACT FROM AUTHOR]