Ranking sustainable suppliers using congestion approach of data envelopment analysis

Abstract Selecting sustainable suppliers is one of the important tasks of decision makers to survive in today's competitive businesses. One of the uses of data envelopment analysis (DEA) is to provide improvement solutions (benchmarks) for inefficient decision making units (DMUs). In classical DEA models, improvement solutions are merely proposed by one of the triple approaches including input-oriented, output-oriented, and hybrid approaches. However, based on the congestion approach of the “theory of economies of scale”, in some cases, the benchmarks derived from the triple approaches are not sufficiently precise. In “theory of economies of scale”, sometimes with an increase in inputs we see more increment in outputs (Wei and Yan, 2004). This means that the triple approaches do not consider all contributing factors of inefficiencies of DMUs. In other words, the triple approaches overlook the shortage of congested inputs as an inefficiency reason of DMUs. Focusing solely on triple approaches cannot always be correct given that sometimes inefficiency reason for some DMUs is due to a shortage of inputs. The objective of this paper, accordingly, is to consider congestion approach (congested inputs) in evaluations of DMUs. To do so, we propose a novel DEA model to show that an increase in congested inputs may lead to higher outputs/efficiency. By so doing, the reliability of improvement solutions are increased. Here, a DEA model is developed to assess the sustainability of suppliers. Also, the congestion of DMUs is detected in pairwise comparisons. A case study is presented to demonstrate the applicability of the presented model.