Improving weighting in sustainable supplier selection with the uncertainty priority analysis on criteria importance

Abstract

Optimisation in decision-making problems under uncertain environments is critical for successfully identifying the "best alternative", given the decision-makers preferences, subjective judgments and opinions. In sustainable supplier selection problems, the different levels of uncertainty in the weighting criteria evaluation may affect the ranking of the best alternative. This work proposes a novel method to improve criteria weighting in sustainable supplier selection, integrating the analysis of uncertainty level in expert group decision-making. In the methodology, the uncertainty level framework is applied to analyse and represent the uncertainty dimensions in the criteria weighting process. Then, the Fuzzy Logic is used to elicit criteria importance and evaluate uncertainty dimensions. Furthermore, the Failure Mode and Effect Analysis technique is utilised to quantify and prioritise uncertainty level in weighting criteria. Finally, the criteria weighting improvement in supplier selection is conducted. Its implementation in a household manufacturer case study in Brazil, Environment criteria have the highest level of uncertainty in the weighting process. The impreciseness type and incompleteness evaluation issue caused by the lack of information affects the quality of the decision. Product quality, environmental costs and customer satisfaction are the most improved criteria influencing expert group decision-making under uncertainty. The originality of this work relies on providing corporate decision-makers with a methodology that analysis uncertainty level and improves the weighting process for an optimal supplier selection. It also provides actionable insights: a decision-making procedure for criteria weighting improvement, experts' obstacles for an effective weighting process, and a detailed uncertainty analysis which serves as input for multi-criteria to support supplier selection decisions and minimise risks in the supply chain.