The Comprehensive Analysis Of Seamen's Collision Avoidance Behaviors And Relative Factors Based On Navigation Simulator Training

Collisions between ships are the main part of marine traffic accidents, most of which are caused by operators. Therefore, it is an effective way in decreasing collison accidents that the study of seamen's behavior during collision avoidance(CA) should be intensified and put on emphasis. Studies in this area can be seen in documentaries home and abroad, which are retrospected and summed up comprehensively. Quantities of CA samples are obtained during seamen training on nautical simulator. Based on these studies and samples, three methodologies, i.e. statistics analysis, Multiple Regress Analysis(MRA) and Principal Component Analysis(PCA), are employed. Overall analyses are exhibited of the influence of 5 factors in seamen's CA action, i.e. the time and the magnitude of course alternation, the relative direction of target ship, the velocity ratio of two ships and the target ship's DCPA. And the relativities of the 5 factors are also analyzed. The conclusions presented in this thesis are useful in seamen's CA decision-making, training on the simulator, and teaching and research of CA courses. Unprecidentedly, the use of the idea of PCA in CA research is put forward, and shows a good performance. It is of special significance in the field of the study of CA. Especially, several mathematical methodologies are employed, and the comprehensive researches are carried out to the same problem, in order to compare and test each other, and provide the references for the solutions of the similar ones.