A selection method of evaluation index for uncertain information system based on synthetic algorithm

Establishment of index system plays a crucial role in the course of uncertain information system evaluation. However, there is not a quantitative means on the problem of constructing the index system for uncertain information system. In order to solve these problems, a novel synthetic algorithm is put forward based on fuzzy set theory and grey relation analysis in this paper. Fuzzy membership degree is used to ascertain the importance of index by using of fuzzy set theory. At the same time, grey relation matrix is used to analysis the relativity of index. Moreover, the rationality and the dependability of index system are validated by using of validity coefficient and dependability coefficient. On base of it, the model of index selection is set up and applied in the index selection of alarm window on main control room of the nuclear power plant. Results of the case study prove that the proposed method is reasonable and feasible.

I. Introduction

Uncertain information system evaluation involves lots of uncertain; stochastic and fuzzy factors which increase the difficulty and complexity of evaluation. However, the construction and selection of index system is extremely important in the study of uncertain information system evaluation. It is the basic and crucial components of veracious evaluation that to build the index system which is scientific, feasible and credible. It directly affects the correctness of evaluation results. However, in the course of the construction and selection of index system, the traditional methods all suffer from some illogical problem: the inconsequence of index system frame, index redundancy and index absence and so on. Therefore, there is important significance to study the index selection theory and method in uncertain information evaluation, which redounds to improve the veracity of uncertain information evaluation.