Expert Evaluation Scale Based On Probability Distribution And Its Application

The scoring habits of experts,which we call the expert-evaluation-scale(EES),are an important reason of the difference of the preference values for the alternatives.We show that many inaccurate calculation results of consensus and ranking results may occur in group decision making(GDM)problems without considering the EES.In order to remove the impact of different EES,we propose the definition of EES and the uniform-expert-evaluation-scale(UEES)strategy.Then we prove the influence of the selection of UEES on two most commonly used consistency measures and two frequently used aggregation operators.Besides,in some real-world GDM problems,a decision maker may hope that the final scores of alternatives are subject to his/her desire distribution.Then,we present a novel ranking method based on the selecting strategy of UEES to obtain the decision maker's desire final scores.Finally,we use a numerical example to clarify the effectiveness of the proposed method.A simulation experiment is used to analysis the sensitivity of our method to the amount of data.The:simulation experiment shows that the proposed method is more suitable for large sample cases(n?30)but the accuracy of the proposed method is still at a high level under the small sample cases(n<30).