Research On Key Technology Of Formal Safety Assessment

As one of the most international industries, shipping industry serves more than 90 percent of global trade by carrying large quantities of cargo. Also it is one of the industries with highest risk. In recent 10 years, the total amount of marine accident decreased. However, there is still a high ratio of serious accidents which cause huge losses. Therefore, safety of shipping has always been the focus of people's attention.Formal Safety Assessment (FSA) is a structured and systematic methodology, aimed at enhancing maritime safety, including protection of life, health, the marine environment and property, by using risk analysis and cost benefit assessment. With the promotion from International Maritime Organization (IMO), FSA has been widely used in ship design and building, vessel and navigation environment assessment, human element analysis, forecasting and preventing of marine accident and its potential risk, etc. Establishment of international convention, regulation, and rule also need applying FSA. Researches are focused on FSA.Theories, methods, and models of frame and 5 steps of FSA, including identification of hazards, risk analysis, risk control options, cost benefit assessment, and recommendations for decision-making, are discussed in this paper. Based on their main points and problems, key problems with wide and significant impact are concluded as the foundation of following researches. Data preparation, concordance of expert judgment, and risk contribution tree method are defined as three key technologies to be improved.For improving the data preparation of FSA, two kinds of data are defined, static data and dynamic data. Collection methods of data are discussed. Key points of improving data preparation are found. To improve the static data preparation, easier and more practical methods of AIS and RADAR data processing are generated considering the actual circumstance in China. Popularity and practicability are enhanced, and dependence on specific facility is reduced. To improve the dynamic data preparation, marine accident database is established after determining the basic data of accident, search and rescue, and oil pollution prevention.For improving the concordance of expert judgment, the sensitivity of concordance coefficient is analyzed based on the expert ranking method and concordance coefficient recommended by IMO. Proper number of experts for a working group is determined. Fuzzy cluster analysis is used to calculate similitude degree of experts' judgment for categorizing when the concordance coefficient is low. According to the category, experts can be reorganized to do the judgment again. Multi-attribute decision making and group decision making methods are introduced into FSA. Analytic Hierarchy Process (AHP) is chosen as the most appropriate method to improve concordance of expert judgment. Experts' judgment are adjusted by subject way and automatic way to meet the concordance standard.For improving the risk contribution tree method, risk contribution tree is divided into fault tree and event tree. Current methods of Fault Tree Analysis (FTA) and Event Tree Analysis (ETA) are discussed. Risk contribution tree is simplified into fault tree as the event tree can be converted to fault tree. Binary Decision Diagram (BDD) is introduced into FSA. Construction methods of BDD from fault tree are presented and compared. Ordering methods of basic event is discussed. Efficiency and accuracy of top event probability calculating are compared between BDD and conventional FTA method for illustrating the improvement of risk contribution tree by BDD.