Data envelopment analysis of two‐stage processes: an alternative (non‐conventional) approach

Network data envelopment analysis (NDEA) is an extension of standard DEA that models the efficiency assessment decision-making units (DMUs) by considering their internal structure. While in DEA, DMU regarded as a single process, NDEA, viewed network interconnected sub-processes (stages, divisions), where flow intermediate products (measures) essential assessment. In prevalent conventional methodological approach to are assumed distinct entities with inputs and outputs. Thus, each sub-process has its own production possibility set (PPS), which can be derived axiomatically from assumptions using minimum extrapolation principle. The PPS overall system defined composition individual PPSs. comprises all methods, divisional efficiencies computed jointly mathematical program. A fundamental property connecting efficient if only divisions efficient. regardless method used, there cases none observed DMUs rendered corroborated real-world case studies. This main issue we discuss this paper motivation propose alternative, non-conventional, address it frame two-stage processes. We consider process two perspectives depending on role measures: producer consumer intermediates. As our based satisfies basic desirable properties. NDEA property, both efficient, met. frontier explicitly DMUs, inefficient projected system. convergent procedure presented for purpose. proposed equivalently expressed multiplier forms due strict primal-dual correspondence operate under constant variable returns-to-scale assumptions. use 22 automotive manufacturers fiscal year 2019 example illustrate approach. Comparison other methods also provided.