DAU Library

학위논문 동서

Fuzzy Fault Tree에 의한 人間信賴性 解析

김정만

부산: 동아대학교 1996. -

97장; 27cm. -

학위논문(박사)- 동아대학교 대학원 기계공학과 `96년8월

영문초록 : The uncertainties and ambiguities involved in many factors that affect human reliability have to be represented as a quantitative form. In the conventional HRA(human reliability analysis), the probability -based theory is used to evaluate the effect of these uncertainties and ambiguities but the actual human reliability should be different from that. In other words, human reliability cannot be based on the probability-based theory but on the fuzzy set theory. This is largely due to the expression of actual error causes such as bad physical condition, high psychological stress level and unsuitable aptitude, etc.. The probability-based HRA makes use of error rates, however, it is difficult to collect data enough to estimate these error rates, and the estimation is dependent only on engineering judgement. And, the basic human error rates are modified by experts to consider the influence of these factors on reliability, however, the influence is not easily expressed quantitatively, because the relation between human reliability and each of these factors is not clear. In this paper, after finding the relation between human errors occurred in subjective uncertainty and their causes which are quantified, a rational method of HRA that is capable of interpreting more clearly the structure of human reliability is proposed. And the error possibility which is a fuzzy set on the interval [0, 1] is used to represent human reliability as a membership function. By the use of fuzzy reasoning that plays an important role to clarify the relation between human reliability and human error, a method is proposed to derive error possibility from the subjective estimation of not only the basic human error rates but also many factors that affect human reliability. When it is supposed that many basic and intermediate events are hierarchically connected to the top event through the Fault Tree structure, by applying fuzzy operation to FTA(fault tree analysis), it is possible to simplify the operation by use of logic disjunction and logic conjunction to a structural function and the structure of human reliability can be represented as a membership function. In order to evaluate the influence of human error on the man-machine system, the fuzzy measure and fuzzy integral proposed by Sugeno are used. And it is shown that the human reliability of the top event can be expressed by linguistic variables using the concept of pattern recognition. Finally, a numerical example of HRA of man-machine system using the Fuzzy Fault Tree is given.

바로가기