Decision-making,based on information acquisition and experience accumulation,is the process of analyzing various factors and finally making decisions by certain means and tools.Individuals are the main body of decision-making activities,characterized with different professional knowledge backgrounds and decision preferences.With economic development and social progress,individual decision-making cannot cope with the increasingly complex decision-making environment any more.Therefore,it is necessary to apply collective wisdom to give full play to the advantages of group decision-making(GDM)then contribute to social production and life activities.Due to the ambiguity thinking and the limited cognition of human beings,decision makers(DMs)are more willing to adopt linguistic information to express their own opinions.The subjective expression of linguistic opinions requires a set of targeted methods to solve the problems encountered in the process of GDM under the linguistic environment.And meanwhile,in view of the incomplete rationality of the DMs,the role of the psychological behavior of the DMs in the decision-making process cannot be ignored.This paper integrates prospect theory,regret theory,and overconfidence theory to analyze the problems encountered in the consensus reaching process(CRP)and alternative selection process of GDM under linguistic environment,with solutions proposed.The main work focuses on the following five aspects:(1)The impact of regret-aversion psychology of middle-level decision-makers on decision-making results under the tri-level decision-making structure is studied.Confronted with critical decision-making problems,DMs at different levels are involved.As they have different interests at stake,their decision-making preferences are various.The middle-level DMs hope that they can refer to the information provided by the low-level DMs and their opinions can be adopted by the upper-level DMs.Therefore,it is necessary to take their regret-aversion psychology into consideration in the process of decision-making.Furthermore,linguistic information processing methods under different granularities and different information categories are offered.Finally,a GDM method for multi-granular linguistic information under a tri-level decision-making structure is proposed,based on regret theory.(2)The impact of the DMs’ consideration of attribute expectations on the decision results using linguistic distribution assessment information is studied.The expression of linguistic distribution evaluation information can well retain the original opinions and the corresponding probability information.However,when there is incomplete partial probability information,the new concept of stochastic linguistic terms is proposed to integrate all types of distributed linguistic information and transform them into stochastic linguistic terms.Meanwhile,the complete linguistic distribution assessment information is inferred,the known complete or incomplete linguistic distribution information applied.Considering that the DMs’ expectations on attribute values will make them compare the evaluation value and the expected values,a GDM method with linguistic distribution assessment information is therefore proposed based on prospect theory.(3)The asymmetry of unit adjustment cost for adjusting linguistic opinions during CRP from the perspective of prospect theory is studied.Psychological behaviors of DMs in the process of consensus reaching and alternative selection are analyzed in the basis of prospect theory.As far as the adjustment process of opinions is concerned,considering that the linguistic opinions before and after adjustment may belong to different semantic expression categories,a consensus model with the minimum asymmetric adjustment cost is constructed.For the selection process,DMs would take original opinions as reference points and compare them with the adjusted opinions so as to impact the decision results.Taking this into consideration,a GDM method based on prospect theory driven by minimum asymmetric adjustment cost consensus model is proposed by taking the comprehensive prospect value as the basis for alternative selection.(4)The GDM problem under two-dimensional linguistic environment with the expression of reliability degree is studied.The operation rules between two-dimensional linguistic expressions are discussed from two perspectives including general and stochastic perspectives.The applicable situations of the two processing methods and their respective advantages and disadvantages are illustrated.Besides,to avoid interferences caused by excessive subjective information,a weight model with reverse thinking is proposed to find the weight information making different alternatives be an optimal choice.Finally,considering the regret-aversion psychology of DMs in the decision-making process considered,a GDM method under the two-dimensional linguistic environment is proposed based on regret theory.(5)The impact of decision-makers’ overconfidence psychological behavior on the CRP with linguistic preference information in large-scale GDM is studied.A clustering method for large-scale DMs and a two-stage minimum adjustment consensus model are proposed,which are aiming at minimizing the number of opinions that need to be adjusted.Since the DMs’ overconfidence psychological behavior will lead to overestimated reliability degree when giving linguistic preference relations,and,there is almost no attention to the measurement of reliability degree of adjusted opinions in existing research,based on two opinions mentioned above,an objective reliability degree measurement method of adjusted opinions is provided,and a GDM method based on overconfidence theory driven by minimum adjustment consensus model is proposed. |