A joint-inputs Network DEA approach to production and pollution-generating technologies

Many real-world production processes generate pollution as an undesirable but inevitable by-product. Identifying, measuring and removing existing inefficiencies can simultaneously increase productivity and benefit the environment. Recently, a pollution generating production technology has been proposed with two distinct processes: a production process and a pollution generation process. This paper uses the novel concept of joint inputs to formulate a parallel-processes Network DEA approach that uses a non-oriented, Network Slacks-Based Measure of efficiency (NSBM) model to assess the efficiency of the overall system and of the individual processes. No previous Network DEA approach had considered joint inputs. In addition, the proposed Network DEA approach improves over existing methods since, among other interesting features, it uses Linear Programming and it identifies all input and output inefficiencies. The proposed approach is rather general and can thus be applied to all production processes involving pollution generation. Numerical results for two datasets, one involving 92 coal-fired power plants and the other 23 EU and OECD countries, are presented.