Research on the Classification of the Relationships Among the Same Layer Elements in Assurance Case Structure for Evaluation

Abstract

The use of assurance cases in certification raises the question of assurance argument sufficiency and the issue of confidence (or uncertainty) in the argument’s claims. Some researchers propose to model confidence quantitatively and to calculate confidence in argument conclusions. However, most of the existing assurance evaluation techniques focus on top-bottom decomposition and assume that the elements of the same level are independent of each other. The lack of information on the relationships among supporting elements of the same level may easily lead to deviations from estimation expectations. In this paper, a modified approach for evaluation of confidence in assurance cases is proposed. Firstly, in order to eliminate the deviation from the independence hypothesis, we discuss the relationships of the supporting elements for confidence evaluation and propose a simple classification. Then we compare the different confidence results under the independence assumption and the correlation assumption using the same confidence evaluation method, and discuss the causes of differences. Finally, we discuss several future work.