A fuzzy MCDM-based approach to monitoring the vaccine supply chain sustainability using objective data and subjective opinions

Purpose This study aims to propose an integrated framework for monitoring vaccine supply chain (VSC) sustainability across multiple regions by combining subjective expert judgement and objective performance data. It also examines how regional rankings change when the relative emphasis on these two evidence sources varies. Design/methodology/approach A VSC sustainability evaluation index was developed using sustainability criteria and associated performance indicators across economic, social and environmental dimensions. Subjective weights were derived using spherical fuzzy DEMATEL to capture causal interrelationships among criteria under uncertain expert judgement. Objective weights were obtained using CRITIC to identify criteria with high regional variation and non-redundant information. The two weights were combined through a sensitivity factor, and VIKOR was applied to generate compromise rankings of regions. Findings The framework was applied to 18 blocks in an Indian district. SF-DEMATEL identified cost optimisation, sustainable financing and training as the most structurally influential criteria. CRITIC identified sustainable financing, workforce requirement and energy and emission as the most informative criteria for differentiating regional performance. Under balanced weighting, Namkum and Sonahatu emerged as the strongest-performing blocks, while Chanho and Itki appeared among the weakest. Sensitivity analysis showed that some blocks remained stable across weighting conditions, whereas others were more sensitive to expert-derived causal judgement. Practical implications The framework helps public health managers identify whether a region requires expert-led systemic diagnosis, PI-level corrective action or broader multi-criteria strengthening. Originality/value This study develops a VSC sustainability assessment framework integrating causal expert weights, objective regional performance information and sensitivity-based ranking diagnosis.