Developing a Two-Stage Network Data Envelopment Analysis Model with Desirable and Undesirable Outputs (Case Study of Butter Manufacturing Industry)

Determining and measuring eciency is an important strategy in decision making. Due to the lack of transparency in the eciency of production stages in Butter production units, the purpose of this study is to design a data envelopment analysis model to evaluate the eciency of production stages of these units. The contribution of this research is to develop a model for evaluating the eciency of a two-stage production system with both desirable and undesirable output using the network model of Khalili and Shahmir (2015). The developed two-stage model has been conducted using accurate and certain data of twenty Butter manufacturing factories in 1399, which can evaluate eciency separately and in an integrated manner. At the end of the rst stage, there are two types of desired output. There is a desired output of breakfast cream that does not enter the second stage of production and it is supplied to market for sale, and another desirable output is the pasteurized cream that is transferred to the second stage of production. At the end of the second stage, a desirable output, which is the packaged butter and an undesirable output, which is butter dough are obtained. The eciency of all twenty units was evaluated and compared. Depending on the eciency of the whole process, there was only one ecient unit. According to the eciency of the rst stage, there was only one ecient unit out of twenty units, and according to the second stage, there are two ecient units. The conclusion is that this industry has a high ineciency in the factories located in provinces of Tehran and Alborz on their production process. Thus, it is necessary for the managers of this industry to increase eciency of their production process.