A Cross-efficiency Evaluation Method Considering Decision Maker Preference Structure
In practice,managers or decision-makers usually need to make strategic decisions among similar decision-making units,such as team performance,project performance,etc.Due to the same external environ-ment,evaluation indicators,and other characteristics of similar decision-making units,it is difficult for decision-makers to make correct evaluations.Scholars have proposed and employed the cross-efficiency evaluation method to solve the above problems.However,the classic cross-efficiency of DEA pursues the absolute objectivity of the evaluation results and ignores the subjectivity of decision-makers in the evaluation process,which leads to the loss of decision-making information and weakens the validity of the evaluation results.In response to this problem,a cross-efficiency evaluation method is given considering the decision-maker prefer-ence structure,which combines with DEA and OWA operators.Specifically,the radical DEA model is used to obtain the initial cross-efficiency matrix of the decision-making units at first.Second,skewness and standard deviation are used to measure the preference structure of decision-makers,that is preference difference and prior-ity,and to modify the initial cross-efficiency matrix.Subsequently,the weights of decision-making units are obtained based on maximizing the minimum preference difference.The final cross-efficiency values of the decision-making units are received by the OWA operator to aggregate the evaluation information.The statistical data of 27 industrial robots from the study of Wang and Chin are used to test the proposed method.Further,the proposed method is compared with the traditional DEA and Chen's method.The perspective of overall ranking,the changing trend of ranking obtained by the three methods is similar;the perspective of specific ranking,the cross-efficiency difference of the proposed method is bigger than that of the other two methods.In sum,the evaluation results of the proposed method have excellent consistency,and it is more conducive for decision-makers or managers to further understand and compare the differences between decision-making units.
万方数据
维普
