| As a kind of universal social activity in human life,decision-making has been widely used in various fields,such as finance,management,control,manufacture,and other industries.Because of the complexity and diversity of decision-making problems,multi-attribute decision-making has attracted wide attention.In multi-attribute decision-making problems,due to the complexity of the decision environment,fuzzy information often appears in decision-making,they are difficult to describe with precise numbers.There are many common ways to describe uncertain information,such as interval numbers,fuzzy numbers,language variables,semantic variables,etc.An effective method is to give the upper and lower bounds of information and express fuzzy information by interval number.This also constitutes an important multi-attribute decision-making problem,that is,interval-valued multi-attribute decision-making.At present,the research on interval-value multi-attribute decision-making usually adopts sorting.Although the accuracy of decision-making results is high,it is necessary to sort all objects in the process of sorting objects.That is,the algorithm of sorting needs to traverse all objects,and then give the relationship between objects according to the results.In this process,it will bring higher computational and time complexity,especially in large-scale decision-making,but it will inevitably consume a lot of manpower and material resources.Because of the high cost of ranking,it is worth studying to find a fast,efficient decision-making method.The three-way decision originated from rough set theory.The core idea of the three-way decision is to divide the whole into three parts,and take different decision-making behaviors and strategies for different parts,and evaluate the corresponding strategies according to actual needs.This transformation from "one" to "three" can be regarded as a whole to part,coarse-grained to fine-grained,macro-micro process.The process of applying strategy is to solve the problems in each region,so as to improve the decision efficiency and reach a certain expected state.As a way to deal with uncertain problems,the three-way decision theory can solve the defects of cost and high complexity in multi-attribute decision-making,which makes it possible to use three-way decisions to solve multi-attribute decision-making(MAMD)problems.In this paper,based on the interval value information system,the multi-attribute decision-making method under the interval value information system is studied.(1)Based on the evaluation value of the information system is interval number,a method of comparing interval numbers by using the connection number is proposed.After comparing the size relation of interval numbers,the paper defines the advantages and disadvantages of the object.Then,the fuzzy advantage relation of the object is constructed,and a comparison method of the superiority degree between objects is given.(2)As a popular measure,distance has been widely used in decision-making.In order to better reflect the comparison results,after the comparison method of the superiority degree between objects is proposed,this paper changes the degree into the distance,and then establishes the calculation formula of the advantage amplitude.all the objects are sorted through this formula,and the order relation of the object set is obtained.(3)In order to accelerate the process of multi-attribute decision-making and reduce the cost of decision-making,the three-way decision is introduced into interval-value multi-attribute decision-making,and a three-way decision(FDD-3wd)based on fuzzy dominance degree is constructed.Based on the superiority degree between objects,this paper gives a method to determine the loss function based on the relative advantage of the object;then,by means of the weight relative probability matrix,the average value of each line is used to show the comprehensive superiority degree of the object to obtain the conditional probability.Finally,the conditional probability and threshold value are compared to determine the area of each object. |
