关键词:

">

"> 区间DEA；交叉效率；邻域互信息；潜在信息函数；云计算

Abstract: The traditional DEA (Data Envelopment Analysis) cross efficiency measures the efficiency by solving a model, which may can't consider the relation of competition and cooperation between DMUs (decision making units) and can’t rank to all DMUs fully. For interval DEA, three models are utilized to measure efficiency of decision making units. These three models, including aggressive, benevolent and competing cross-efficiency models, considering both competition and cooperation among DMUs. The classified model is improved in third model and put forward a method to classify the discrete data by interval neighborhood mutual information. Then, the latent information function is applied to get the weights of above mentioned three models for every decision making unit and aggregate efficiency scores of different models in accordance with the OWA operator. In addition, based on the aggregative efficiency scores, the decision making units are ranked. In the end, a numerical example about selection of cloud computing models is used to verify the rationalization and feasibility of the proposed method.

Key words: Interval DEA, cross efficiency, neighborhood mutual information, latent information function, cloud computing