Research On Multi-attribute Decision Making Methods Based On Z-number

Multi-attribute decision making(MADM)is to select the best alternatives or sort alternatives under the restriction of multiple attributes.Its development has benefited many areas of society.The fuzziness,inaccuracy and incompleteness of evaluation information in multi-attribute decision making problems often have great influence on the decision result.Therefore,in multi-attribute decision making,people need a fuzzy language,Z-number,which can describe practical problems more accurately and carefully and close to human language.As a fuzzy language that can describe information reliability,Z-number is a powerful tool to represent evaluation information in multi-attribute decision making.It consists of a fuzzy limit A and a reliability measure B,which are linked by a underlying probability distribution,resulting in a more complex structure than ordinary fuzzy numbers.And its calculation is also a complex process.Considering the wide application prospect of Z-number in practical problems,it is of great research value and practical significance to explore the effective calculation method of Z-number.The specific research content of this paper is as follows:1.Aiming at the problem of information distortion in the process of Z-number calculation,exp-TODIM model based on Z-wasserstein distance is studied.The Z-wasserstein distance measurement method is used to measure the distance between two Z-numbers.A new decision model is constructed by combining Z-wasserstein distance with exponential TODIM method(exp-TODIM).The model is less affected by parameter changes and reduces information loss to a certain extent.The feasibility,stability and advantages of this method are demonstrated by model application,sensitivity analysis and comparative analysis.2.Aiming at the problem of multi-attribute decision making(MADM)with unknown attribute weight,the evaluation method of distance from average solution(EDAS)under Z-number environment is studied.Criterion importance through inter-criteria correlation(CRITIC)method is used to determine attribute weight,and Z-number centroid point is taken to deblur Z-number.It makes the calculation process more simple and avoids subjective prediction.The feasibility,effectiveness and practicability of the proposed method are verified by example analysis and comparative analysis.