Research On The Law Of Loss Of Life At Sea And Marine Individual Risk Acceptance Criteria

The paper researched on the law of formation and distribution of loss of life at sea of the current stage on the understanding of the main contradiction in the study of marine traffic safety. It was very great to guide the study and practice directions of marine traffic safety, and it was very important to improve the protective ability of the safety of life at sea. This dissertation firstly reviews the present analyzes its potential problems and makes clear of relative targets, ideas and tasks. Focus on the theme of the law of loss of life at sea, forcast and measure, the paper carrys on its research from the following aspects.First of all, the paper studied the law of the loss of life at sea, time, form and fatality rate. Quantitative analysis of the general characteristics of loss of life at sea.Secondly, the paper studied the nature of accidents of loss of life at sea from the perspective of multidimensional association rules based on the Apriori algorithm. For the existing problems of the study of laws of loss of life at sea and the insufficient safety technology, it used Apriori algorithm to make the multidimensional association rules research on the nature of accidents of loss of life at sea, discovering the important association rules between marine traffic accidents and operation accidents on board. The main type of accidents causing the death at sea is the operation accidents rather than those widely concerned by the public such as shipwrecks, marine traffic accidents and missing. It also reveals the close relationship between marine traffic safety and operation safety on board. In addition, The paper put forward that the operation safety into the categories of the study of marine traffic safety and increase the research levels so as to supplement and enrich the theory of marine traffic safety.Thirdly, it studied the combined prediction model of loss of life at sea based on ARIMA and S VM. Preventing loss of life at sea has always been the focus of the study and practice of marine traffic safety. For the problem that the existing predictive method is unable to achieve the ideal prediction accuracy, it put forward a kind of a combination prediction model based on of ARIMA model and SVM method. It firstly adopted ARIMA model to deal with the linear part, and then used the SVM model to deal with the nonlinear part. It is verified that, compared with the prediction of the single model, the combined model is more precise. This combined prediction model is a kind of effective prediction model of loss of life at sea.Finally, it studied the Marine Individual Risk Acceptance Criteria (MIRAC) based on Pareto Optimality. In view of the current situation where the widely accepted risk criteria have not been set up in the marine traffic area. On the basis of defining the concept of MIRAC and the principle of threshold determination, the MIRAC threshold was determined based on Pareto Optimality.