An Alternative Approach to Dealing with the Composition Approach for Series Network Production Processes

Abstract

To deal with efficiency assessment in a two-stage production process, Despotis et al. (2016b) proposed an innovative composition approach to calculate the two-stage efficiencies, whose minimum is defined as overall efficiency of the network production process. The underlying proposed programming method in this approach is nonlinear and requires the application of a very time-consuming bi-section procedure to obtain the stage efficiencies. In this paper, we have, therefore, proposed a simple linear network DEA model to carry out the same stage efficiency assessment, which is computationally efficient, and is also readily applicable to any multi-stage production process. Additionally, unlike the Despotis et al.’s (2016b) method, our proposed alternative is extended to deal with dynamic efficiency assessment. Finally, we have considered both synthetic and real-life data to demonstrate the ready applicability of our proposed models for both static and dynamic efficiency assessments.