Developing a new chance-constrained DEA model for suppliers selection in the presence of undesirable outputs

Supplier selection is a significant and widely studied theme since it has a significant influence on purchasing management in supply chain. Slacksbased measure - undesirable output (SBM-undesirable output) model is one of the new models in data envelopment analysis (DEA). In many real-world applications, data are often stochastic. A successful approach to address uncertainty in data is to replace deterministic data via random variables, leading to chance-constrained DEA. In this paper, a SBM-undesirable output model is developed to assist the decision makers to determine the most appropriate suppliers in the presence of both undesirable factors and stochastic data, and also its deterministic equivalent which is a non-linear programme is derived. Furthermore, it is shown that the deterministic equivalent of the stochastic SBM-undesirable output model can be converted into a quadratic programme. In addition, sensitivity analysis of the SBM-undesirable output model is discussed with respect to changes on parameters. A case study demonstrates the application of the proposed model.