Multiple Attribute Decision Making Research Based On Attribute Weights And Conversion Relations

Fuzzy Multi-Attribute Decision-Making(FMADM)problem is becoming an important research field of operational research and modern decision science.According to what they know about the state of things and the effect of action,people make decisions usually by using certain theories and methods or uncertain theories and methods.For different decision-making situations,fuzzy set theory,intuitionistic fuzzy set theory,interval number theory can be used to deal with the uncertain information of corresponding types.Although some achievements have been made on the research of uncertain theories,the researches on theory and application are still not deep enough and there still exist many issues which need to be improved and further studied.The objective weighting methods,subjective weighting methods,objective and subjective weighting methods,including single weighting methods and combination weighting methods are very important methods used to determine attribute weights.It is an urgent problem to be solved that how to balance the all aspects in combination weighting methods.Sometime,the decision maker will give decision information by using different fuzzy numbers,so it is important to consider the conversion relationship between fuzzy numbers and the corresponding decision method.The article mostly studies some methods about weighting and ranking options.The article proposes some weighting methods and ranking methods,and the main research works are as follows:1)Research on application of combination weighting methods in FMADA problems.We analyze the shortcomings of single weighting methods about entropy and deviation,and introduce singe balance factor and double balance factors to deal with the weighting methods in multi-attribute decision-making problems.At last,we rank the options by means of close coefficient and correlation coefficient on options and optimum solution.2)Research on application of vector group of attribute weights in FMADM problems.Concerning multi-attribute decision-making problems,whose decision information are expressed by intuitionistic fuzzy numbers,we firstly get attribute weights by means of minimizing the deviations of attribute values and preference values of decision makers and maximizing the deviation between attribute values respectively.And then we get the vector group of attribute weights.At last,we rank the options by means of correlation coefficient on options and ideal solution.3)Research on unary interval number and its application in FMADM problems.We replace intuitionistic fuzzy number with nuary interval number while introducing risk factor which reflects the decision maker's attitude to the risk.We get the attribute weights by using the deviation method and replace unary interval number with connection number.At last,we rank the options according to the weighted comprehensive values coming from complex operation of connection number.4)Binary interval number and its application in FMADA problems are studied in this thesis.We improve the decision making method on unary interval number,and replace intuitionistic fuzzy number with binary interval number.ByUsing the transformation method,we consider not only the effect of membership and non membership degree,but also the effect of hesitancy degree on decision making process.We replace the binary interval number with binary connection number then calculate the weights.Concerning the risk factor,we rank the options based on two different methods respectively.The corresponding practical examples are illustrated to show their effectiveness and feasibilities.In this paper,the main innovation points are as follows:1)Research on application of combination weighting methods in FMADA problems.We pay attention to digging for informations.We consider not only the importance of information,but also the relationship between data.In order to balance the relationship between two aspects,we introduce the balance factor.2)We establish optimization models for solving the optimal weights,and get the vector group of the attribute weights.Compared to the other decision-making methods,more attention is paid to the independence of options in the vector group of decision method.And we point out that even the same attributes of different options play different part in decision making problems,so the weights of every options shoud be different.3)Converting intuitionistic fuzzy numbers into unary interval number.This transformation formula converts "point estimation" into "line estimation'',so that the formula can describe the fuzziness of things more accurately.By establishing optimization models,we solve the optimal attribute weights and rank the options.4)Converting intuitionistic fuzzy numbers into binary interval number.This transformation formula converts "point estimation" into "line estimation",as well as considers the effect of membership,non membership,hesitancy degree in decision making,so it can describe the fuzziness of things and the intention of decision maker more accurately.By establishing optimization models,we solve the optimal attribute weights and rank the options.