Study On Fuzzy Multiple Attribute Decision Making Method And Its Application

Multiple attribute decision making theory and method has long been the research hotspot of decision making science.It has been widely used in military operation,recommendation system,economic management,artificial intelligence and many other fields.Due to the fuzziness of human cognition and the relative factors of decision object are quite complicated,decision information is difficult to characterize in a crisp number.In view of this,more and more scholars use fuzzy sets to model decision making information,and thus form a new research field: fuzzy multiple attribute decision making.This paper studies the multiple attribute decision making problem in intuitionistic fuzzy environment,Pythagoras fuzzy environment and hesitant fuzzy language environment.The main work and results are as follows:1.Research on intuitionistic fuzzy decision making problem at multiple angles of view.The existing quantitative information aggregation operators are mostly modeling information from one-dimensional perspective,which leads to multiple objective decision making problem cannot be solved.In view of this,this paper introduces intuitionistic fuzzy power Heronian mean operator and intuitionistic fuzzy power geometric Heronian mean operator.These new operators can not only capture the relevant information between the attribute variables,but also excavate the integrity of the decision information by implementing the crossover operation of the Heronian mean operator and the relative closeness of the power average operator.On this basis,the paper proposes a method to solve the problem of intuitionistic fuzzy multiple attribute decision making,and it is used to solve the case of recommendation of electronic products and asset acquisition of listed companies.The experimental results show the feasibility and effectiveness of this method.Modern decision making problems often require optimal selection under multiple objectives.The multiple objective decision making operator proposed in this paper provides a theoretical reference for solving such problems.2.Research on Pythagorean fuzzy decision making problems under bounded rationality.In view of multipule attribute group decision making problem in which the attribute values are Pythagorean fuzzy numbers and the criteria weights are unknown,amethod based on D-S theory and Pythagorean fuzzy hybrid weighted MSM(PFHWMSM)operator is proposed.Firstly,the expert's fuzzy measures and weights are obtained from decision information matrix.Then,the PFHWMSM operator is introduced to aggregate the attribute information,and obtain the comprehensive evaluation information of expert.Then,using the evidence combination method to infuse the comprehensive evaluation information,the comprehensive evidential information is obtained.Moreover,the trust interval of alternatives can be built,which is used for decision making.Finally,the application of the decision method in green supplier selection is considered.The PFHWMSM operator proposed in this paper can embody the dominant characteristics of aggregated objects,and overcome the defects of the existing weighted MSM operators which can not satisfy idempotency and degeneracy.The proposed decision making method integrates operator and evidence combination and other tools,so that the decision making model can fully reflect objective information and subjective psychological factors.Thus more reliable decision results can be obtained.It also provides a demonstration for the comprehensive use of different types of decision making tools.3.Research on Pythagorean power fuzzy decision making problems.The existing Pythagorean fuzzy decision making methods cannot deal with complicated exponential calculation which the exponents are PFNs and the bases are crisp numbers.As the complement of the existing operational rules,this paper defines two Pythagorean fuzzy exponential calculation rules to handle crisp numbers and interval numbers separately.In addition,we put forward the novel concept of dual Pythagorean fuzzy number.On these bases,we propose PFN weighted exponential aggregation operator and dual PFN weighted exponential aggregation operator.Then,an approach based on these exponential aggregation operators is developed for multi criteria decision making problems.Finally,this paper illustrates the applicability of the proposed approach through a convenient example.This study greatly expands the application scope of Pythagoras fuzzy sets and provides a feasible path for the representation of uncertain attribute weight information.4.Research on hesitant fuzzy linguistic decision making problems.Hesitant fuzzy linguistic term set is a flexible qualitative modeling tool.The main advantage is that it can better depict the psychological characteristics of decision makers.We first define a generalized weighted power averaging operator to fuse hesitant fuzzy linguisticinformation.The operator is an implicit function,so it is more universal than the existing generalized power weighted average operator.In addition,in order to avoid the randomness of the subjective weighting method,we propose the hesitant fuzzy linguistic generalized power ordered weighted averaging operator for the case where attribute weights are completely unknown based on the complete order of hesitant fuzzy linguistic terms.Most of the hesitant fuzzy linguistic ordering operators in the literature are constructed based on the partial order of terms,so the ordering weighted average operators proposed in this paper can be applied to a wider range.In addition,on the basis of the analysis of the defects of the existing weighted Bonferroni mean operator,the hesitant fuzzy linguistic reducible weighted Bonferroni mean operator and the generalized hesitant fuzzy linguistic reducible weighted Bonferroni mean operator are proposed,which can be used to aggregate the hesitant language information under different situations.New Bonferroni mean operators have good properties such as idempotent,degeneracy,boundedness and so on.Moreover,based on Frank T-norm and T-conorm,we defined the Frank operation laws for hesitant fuzzy linguistic.And then,hesitant fuzzy linguistic Frank weighted arithmetic and geometric aggregation operators are proposed.New operators are the generalization of HFLWA and HFLWG operators.The defined Frank aggregation operators contain a parameter which can reflect the decision maker emotion.This makes them more flexible in the process of information fusion.Finally,based on the above hesitant fuzzy linguistic operators,a series of decision making methods are proposed to solve decision problem in the hesitant fuzzy language environment.From the perspective of decision maker's risk preference and prospect expectation,this paper analyzes these decision making methods and provides suggestions for decision makers.The research enriches hesitant fuzzy linguistic multiple attribute decision making theory,and provides a reference for extending other numerical aggregators to hesitant fuzzy linguistic environment and solving multiple attribute decision making problems.In addition,the proposed multiple attribute decision making models is simple and has popularization value.