Calculation of Luenberger productivity index in two-stage network with the presence of undesirable intermediate output and uncontrollable input data

In recent literature, many studies have been performed on DEA performance evaluation of systems with two-stage network structure. One of the topics of interest to researchers is the study of the progress or regress of production over separate time periods in the two-stage network structure. The Luenberger productivity index is one of the techniques introduced in this field. However, the calculation of output or progress in the two-step network in the presence of undesirable intermediate products and uncontrollable input data has received less attention. The main purpose of this paper is to investigate the impact of the presence of undesirable intermediate products and uncontrollable input data in evaluating the performance of two-stage series-structured network systems when discussing the progress or regress of production systems during different periods of activity. For this purpose, in this paper, we introduce a new directional distance function and propose a method to solve this model that is not problematic in the adjacent time periods. Then, we present a three-stage algorithm to calculate the Luenberger productivity index in a two-stage network with the presence of these data. Finally, with an example, we show the applicability of the proposed algorithm.