A Probabilistic Uncertain Linguistic Decision-Making Model for Resilient Supplier Selection Based on Extended TOPSIS and BWM

Resilience is the sustainable competitive advantage of suppliers in the supply chain, and the ability of resilient suppliers to manage risk and perform better in supply than traditional suppliers in the event of disruption has driven the complexity of the current supply chain. Therefore, studying how to select a resilient supplier is necessary for establishing a supply chain with flexibility in the case of interruption. A hybrid fuzzy Multi-Criteria Group Decision-Making (MCGDM) framework is developed in this paper for Resilient Supplier Selection Problems (RSSPs). First, Probabilistic Uncertain Linguistic Term Sets (PULTSs) are introduced to deal with the subjectivity and uncertainty of experts’ assessments. Second, considering that experts may have different views on the relative importance of resilient criteria depending on their different knowledge backgrounds, the Probabilistic Uncertain Linguistic Best–Worst Method (PUL-BWM) is constructed to determine the weights of resilient criteria under different experts. In addition, given that the traditional Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) cannot handle the information metrics with negative values or reflect the correlation of information, the extended TOPSIS method based on a novel Probabilistic Uncertain Linguistic Synthetic Correlation Coefficient (PULSCC) is constructed to select the optimal resilient supplier. The novel PULSCC also overcomes the drawbacks of the existing correlation coefficient between PULTSs by considering the mean, variance, and information completeness of PULTSs. Finally, an example of resilient supplier selection in the automotive industry is performed to validate the applicability and feasibility of the proposed approach. The sensitivity and comparative analyses are conducted to demonstrate the effectiveness and superiority of the proposed framework.