A supplier risk assessment model based on hesitant fuzzy linguistic sets with possibility distributions: application in a humanitarian aid supply chain

Purpose This study aims to propose a group decision-making model designed to support the assessment of supplier-related risks. Design/methodology/approach The model uses the possibility distribution hesitant fuzzy linguistic term sets combined with the technique for order of preference by similarity to ideal solution (PDHFLTS-TOPSIS) to categorize suppliers within a segmentation matrix. While PDHFLTS is used to represent the judgments of decision makers (DMs), the TOPSIS algorithm is applied to generate the overall scores of the suppliers. A pilot application was conducted in a civil defense organization. Sensitivity analysis tests were performed to assess the impact of varying criteria weights on the results. Findings The criteria related to cost (C1), import instability (C2) and delivery time (C4) received greater weights. Suppliers A3 and A4 were classified as low risk, while A1 and A2 were categorized as intermediate risk. Practical implications Implementing the proposed model requires structured protocols for identifying risk criteria and training DMs in HFLTS. Integration into routine processes demands adapting digital systems to enable real-time processes. In addition, the use of consensus-based evaluations calls for a cultural shift toward collaborative decision making. Originality/value This is the first supplier risk assessment model that performs supplier segmentation while allowing DMs to use linguistic expressions to evaluate supplier risk levels under conditions of hesitation. It is also the first to incorporate possibility distributions associated with judgments for risk assessment and support the weighting of DM opinions based on the degree of concordance with the group, enhancing the reliability and representativeness of the final assessment.