A cloud model-based probabilistic linguistic multi-attribute decision method and its application to public evacuation in nuclear accidents

Abstract

Probabilistic linguistic term set can effectively express the evaluation preferences of decision makers, and cloud model can combine fuzziness and randomness, describe the correlation between fuzziness and randomness with numerical features, and form a mutual mapping between qualitative and quantitative. For the multi-attribute decision problem with unknown expert weights and attribute weights under the probabilistic linguistic term set environment, this paper proposes a cloud model-based multi-attribute decision method. Firstly, golden section method is used to transform probabilistic linguistic information into digital features of clouds, and the definition of the probabilistic linguistic cloud and calculation formula of the distance between the probabilistic linguistic clouds are given. Secondly, the expert weights and attribute weights are calculated based on this distance formula, and the alternatives are ranked using the technique for order preference by similarity to ideal solution (TOPSIS). Finally, the proposed decision-making method is applied to the decision of public evacuation in nuclear accidents, and the feasibility of the proposed method is verified by comparative analysis.