A New Strategy To Rank Fuzzy Number Based On The Condorcet's Jury Theorem And The PageRank Algorithm And Its Application To Intelligent Decision-making For Enterprises

Decision-making,a kind of activity that often occurs in business management,is a process in which enterprise managers,operators and government officers make their decisions after analyzing,calculating and judging the factors affecting the realization of the goals,with aims to achieve specific goals.It is performed according to some important elements such as objective possibilities,certain information and experience,and generally,professional tools,skills and methods are indispensable.Because of the complex and changeable social environment and intensified market competitiveness,enterprise managers,operators and government officers have to confront more and more complicated decision-making problems.Among them,how to compare and rank various alternatives is the core challenge that needs to be tackled in enterprise intelligent decision-making.Generally speaking,in the evaluation of alternatives,there is no situation where an alternative is fully dominant in all aspects.It is usually superior in some aspects,while weaker in others.At the same time,it is often difficult for decision makers to give an accurate evaluation of these alternatives.Therefore,decision makers often tend to employ uncertain values such as "very good,good,ordinary,poor,very poor" to measure evaluation information.However,traditional decision-making methods can only deal with precise values.Thus it often seems inadequate in the face of this kind of decision-making problems with uncertain information.For this reason,people attempt to depend on the experience and wisdom of decision-makers to handle multi-attribute decision-making problem with uncertain information,but this often generates deviation,which influences the efficiency of enterprise production and operation.In the objective world,the uncertainty contained in events which are difficult to give a precise definition is called fuzziness.An important characteristic of human intelligence is that its thought and action are accompanied by vagueness.In order to solve the problem of intelligent decision-making under fuzzy conditions,the intelligent decision-making method based on fuzzy set theory emerges,which aims to simulate human thinking by computer.In decision analysis,if the evaluation information of decision makers is described by fuzzy numbers,then the comparison and selection of alternatives can be summed up as ranking fuzzy numbers.Although many scholars have conducted in-depth research on ranking fuzzy numbers,and realized a lot of research results,the existing ranking methods have more or less defects due to the semi-ordered structure of fuzzy number.Therefore,ranking fuzzy numbers has always been an important issue in fuzzy set theory.Considering that ranking fuzzy number is an important research content in intelligent decision-making science,and the process of making decision of an enterprise can be considered as a process of ranking various alternatives,this paper introduces fuzzy number ranking method into enterprise intelligent decision-making process,and puts forward a fuzzy number ranking strategy based on the Condorcet's jury theorem and the Page Rank algorithm.The proposed ranking strategy first maps each alternative to fuzzy numbers,and then integrates several fuzzy number ranking methods by Condorcet's jury theorem in a majority manner voting into a frame to rank fuzzy numbers in pairs.Finally,the classical algorithm in the field of machine learning-Page Rank algorithm is applied to calculate the ranking results of all those fuzzy numbers as the final ranking results of each alternative,so as to help enterprises make scientific decisions.The main innovation points of this paper are as follows: for the first time,the Condorcet's jury theorem in the field of social science,the Page Rank algorithm in the field of machine learning,and the fuzzy number ranking method in the field of fuzzy mathematics are organically combined to propose a new framework of ranking fuzzy number.In this framework,any ranking method that contributes to improve the ranking accuracy can be added,and meanwhile,any ranking method that is inappropriate can be excluded.Through tuning,the ranking accuracy of the strategy still has huge space for improvement.Simultaneously,due to the introduction of Page Rank algorithm,it can also have prominent computational efficiency in the extreme environment of massive data.Five application examples of ranking fuzzy number in Chapter 4 manifest that this strategy can not only effectively compensate for the shortcomings of employing single fuzzy number ranking method,but also greatly enhance the accuracy.Chapter5 lists two supplier selection cases and a property service project evaluation case to illustrate the efficiency of the proposed fuzzy number ranking method.