An interval efficiency evaluation model for air pollution management based on indicators integration and different perspectives

摘要

Efficiency evaluation of air pollution management is one of China's most popular political concerns. Many studies have been devoted to developing an effective efficiency evaluation model for air pollution management. However, three challenges must be addressed and solved. First, previous studies treated air pollution data as real numbers in efficiency evaluation. This ignored the fluidity and variability of air pollutants. Second, the indicators used in previous studies were selected based on experts' knowledge. Accordingly, there was information loss in efficiency evaluation. Last, many factors can affect the efficiency evaluation of air pollution management. Yet, previous studies neglected to evaluate efficiencies from different perspectives. To address the challenges above, the interval data was first used as basic data to meet the flow characteristics of air pollution management. Next, the interval evidential reasoning (IER) model and the interval data envelopment analysis (IDEA) model were employed. Both models are suitable for interval data. They were introduced to propose a new efficiency evaluation model: the IER-IDEA model. The main aims of the proposed model are twofold: 1) include interval indicator integration with a new weight calculation method; and 2) evaluate interval efficiency from three different perspectives. To demonstrate the effectiveness of the proposed IER-IDEA model, a case study was performed in 29 Chinese provinces. Its purpose was to assess efficiency evaluation of air pollution management. Experimental results demonstrated that the IER-IDEA model obtains desired efficiencies of air pollution management under the consideration of indicator integrity, interval uncertainty, and different perspectives. Moreover, it can effectively distinguish regional differences in the efficiency of air pollution management compared to some existing efficiency evaluation models. (C) 2019 Elsevier Ltd. All rights reserved.