Research On Anti-efficiency Network DEA Model And Its Application

Data Envelopment Analysis(DEA)is a mathematical programming method that can be used to evaluate the same type of decision-making unit with multiple inputs and multiple outputs to find the best decision-making unit.In recent years,DEA methods have been widely used.In the evaluation of various production activities,and achieved important results,it is because it can evaluate the decision-making unit and provide useful information for decision makers.However,traditional DEA can freely choose the optimal weight,which is a kind of attention.In order to obtain maximum efficiency,the evaluation process will weaken the influence of the insufficient aspects of the decision-making unit.This method of free choice of weights can effectively identify inefficient decision-making units,but it reduces the evaluation ability of DEA.In order to solve this problem,a large number of scholars have studied the anti-efficiency model and have been widely applied in practice.The research on the inverse efficiency model is still in a simple single stage.Constructing the inverse efficiency model under the complex network structure and constructing new indicators can effectively improve the evaluation ability of the original model.The traditional DEA method treats each decision-making unit as a “black box”and ignores the internal organizational structure,which affects the evaluation results.In the actual production process,the internal structure of many decision-making units is a network system rather than a “black box”.For example,the operation process of a commercial bank can be regarded as a serial network structure,and the forest production system is a typical parallel structure.In this paper,the series and parallel connection in the network DEA is taken as an example to construct the inverse efficiency model and the parallel structure under the series structure.The inverse efficiency model is used to decompose the efficiency of the inverse efficiency model under the series structure.Then,according to the evaluation purpose of the decision maker,three efficiency indicators are constructed based on the efficiency and the inverse efficiency.In the empirical stage,the model is applied to the evaluation of China's commercial banks and forest production systems in a certain region,and the sample data is compared in the traditional DEA model,the inverse efficiency DEA model and the re-constructed three efficiency evaluation indicators.The results were analyzed and compared.It was found that the combination of theinverse efficiency index can effectively improve the evaluation ability of the original model.The efficiency decomposition of the inverse efficiency model model under the series structure can also effectively identify the inefficient stage,showing the model and efficiency index.Rationality and feasibility.Expanded the evaluation theory of network DEA.