ASSESSMENT OF VESSEL COLLISION RISK FACTORS USING GRADUAL REGRESSION

Abstract

The purpose of this article is to develop a method of gradual regression for assessing ship collision risk factors, which allows for a more accurate assessment of the probability of collision and the development of more effective measures to reduce the risk of collision by including in the model only those collision risk factors that have a statistically significant relationship with the probability of collision. Ship collisions are one of the most common incidents at sea. They can lead to serious consequences such as injuries, loss of life and pollution. It is important to conduct a collision risk assessment to prevent them. A ship collision risk assessment involves identifying factors that could lead to a collision and assessing the likelihood of such collisions. In recent years, intelligent methods of collision risk assessment have become increasingly common. They allow for a more efficient and accurate risk assessment than traditional methods. This paper reviews existing research in the field of collision risk assessment. The task of assessing collision risk factors was formally defined and the data used was pre-processed. The main result of the work is the developed methodology of incremental regression and its applicability for assessing collision risk factors. The paper identifies and selects risk factors that may affect the probability of a ship collision and analyses the relationship between these factors. The results of the experiment showed that the most statistically significant risk factors of a collision are the distance between ships, course difference, ship speed, visibility, and weather conditions. The distance between ships had the greatest impact on the probability of a collision, followed by the difference in course. Vessel speed, visibility and weather also had an impact, but to a lesser extent. These results highlight the need to consider these factors when managing a vessel to reduce the risk of collision. As suggestions for future research, it is recommended to expand the data sample, analyse collision data for a longer period, analyse collision data in different conditions, and develop measures to reduce the risk of ship collision based on risk analysis.