| The relative efficiency of an enterprise is an important indicator for evaluating the business situation,and the development of feasible efficiency evaluation and improvement programs is of great significance for the development of the enterprise.In reality,there are numerous uncertainties: for example,the subjective preferences of decision makers will affect the evaluation results;in addition,imprecise objective data also lead to uncertainty in the measurement results.Therefore,this paper investigates the problem of efficiency evaluation and improvement based on interval data envelopment analysis(DEA)under uncertainties.In this paper,DEA efficiency analysis is divided into three stages: construction of DEA efficiency measurement model,decision making units(DMU)efficiency ranking,and DMU efficiency improvement.For these three problems,the corresponding solutions are proposed based on the interval DEA method and the method integrated with other theories,and the main work is summarized as follows:(1)A non-radial interval DEA model with slacks-based measure is proposed.To solve the disadvantage of underestimation of efficiency in the traditional SBM model,an interval DEA model with slacks-based measure is proposed to calculate the efficiency values at the farthest and the nearest distance from the strongly effective frontier respectively,then the efficiency interval is obtained.Furthermore,the method is extended to interval data context to solve the problem of efficiency measurement under imprecise data.(2)A full ranking method of decision making units by combining DEA and Anlytic Hierarchy Process(AHP)is proposed.In order to solve the drawback that the traditional DEA model can only judge the effectiveness of DMUs,but cannot fully rank DMU.The DEA efficiency measurement results are compared in pairs to obtain the interval multiplicative preference relation and then generate the consistent interval multiplicative preference relation to achieve the full ranking of DMUs.(3)An interval DEA efficiency improvement method based on target setting is proposed.In order to solve the problem that it is difficult to define the efficiency improvement criteria in interval DEA,a bi-level DEA efficiency improvement model was proposed.In the upper model,the efficiency improvement target is set according to the actual situation,while the lower model ensures the optimality of DEA model weight selection.Then,the efficiency improvement model under other improvement objectives is proposed by combining the decision maker preferences.Finally,the solution method is given for the proposed bi-level DEA model.This study proposes corresponding theoretical solutions to the three problems existing in interval DEA efficiency analysis,and further applies the proposed methods to the case of tourist hotels in Taipei,to solve three realistic problems of interval efficiency measurement,business performance ranking and performance improvement,respectively,which has theoretical and practical value. |