Two-sided Matching Decision Making Methods Based On Fuzzy Preference Relations With Self-confidence

Two-sided matching decision problem refers to that matching objects in one set need to be matched with matching objects in the other set,which exists widely in human beings’ daily life,such as marriage matching problems,employee-job matching problems and matching problems in venture investment activities.In recent years,the researches on two-sided matching problem have attracted a lot of attention from scholars all over the world and abundant research results have been achieved.However,in actual two-sided matching decision making problems,matching objects might express their preference information in the form of fuzzy preference relations with self-confidence(FPRs-SC).What’s more,matching objects might have disappointment and elation towards matching result.Based on above analysis,this thesis focuses on two-sided matching based on FPRs-SC with consideration of matching objects’ psychological behaviors.The main research work of this thesis is summarized as follows.(1)Methods to derive priority weight vectors and algorithms to improve consistency levels of FPRs-SC are investigated.For acceptable FPRs-SC,a logarithmic least squares model is built and solved to derive its priority weight vector,based on which consistency level of an FPR-SC is measured by calculating deviation between an FPR-SC and its priority weight vector.For FPRs-SC with unacceptable consistency levels,two algorithms are put forward to improve their consistency levels.For unacceptable FPRs-SC,an algorithm is put forward to derive priority weight vectors,as well as measure and improve consistency levels.(2)For two-sided matching decision making problems based on FPRs-SC,this thesis put forward decision making methods with consideration of matching objects’ psychological behaviors.First,based on methods to derive priority weight vectors and algorithms to improve consistency levels of FPRs-SC,calculation of original satisfaction degrees of matching objects is proposed.Second,this thesis studies the disappointment and elation of matching objects towards matching result based on disappointment theory and develops formulae to calculate disappointment and elation values,based on which calculation of adjusted satisfaction degrees is presented.Third,stable matching conditions are taken into consideration and a stable two-sided matching model is established aiming at maximizing total satisfaction degrees of both sides.The optimal stable matching result can be derived by solving this model.(3)This thesis demonstrate the proposed models and methods by a venture investment twosided matching problem and matching results under different matching conditions are compared and analyzed.Results demonstrate that stable matching conditions and self-confidence levels have an effect on matching results.The proposed models and algorithms not only can enrich the theory and methods of twosided matching decision making,but also be applied to deal with practical two-sided matching decision making problems.It is of certain significance and value in both theory levels and application levels.