Answers are grounded in this paper and cite their sources.
Virtual industry-education collaboration platform procurement requires combining judgments from stakeholders whose expertise varies by criterion while retaining bounded uncertainty in both criterion importance and vendor performance. This study develops a grey-epistemic-trust (GET) weighting rule and integrates it with the established permutation-based Grey–QUALIFLEX ranking procedure. GET operationalizes epistemic reliability using three observable signals: criterion-specific domain-expertise rating, experience-based credibility proxy, and interval-width confidence proxy. The conjunctive product determines criterion-specific decision-maker influence. The empirical case contains six vendors, 18 criteria, and eight decision-makers. Interval grey importance assessments were elicited directly on a 0–10 scale; vendor performance was rated individually on a 0–100 scale and aggregated into group interval grey performance values using the observed minimum and maximum ratings after restoring canonical vendor identities. The analysis identifies the fourth vendor ( \(A_4\) ) as the top-ranked alternative and the fifth vendor ( \(A_5\) ) as the lowest-ranked alternative, with \(Net(A_4)=2.431\) and \(Φ (π ^*)=5.210\) . Weighting ablation shows that the empirical ordering is stable across GET, equal, midpoint-mean, grey-geometric, and domain-expertise-only weights. To evaluate decision quality beyond rank correlation, a fixed-seed 1000-replication controlled experiment with known latent weights and oracle rankings was added. Under a reliability-aligned data-generating regime, GET reduced mean absolute weight-recovery error from 0.0037 (midpoint mean) to 0.0030 and increased oracle top-choice agreement from 93.5 to 95.0%; under a reliability-neutral regime, midpoint averaging performed better, showing that GET’s advantage is conditional rather than universal. This study contributes a transparent reliability-weighting mechanism, a reproducible grey outranking integration, and an empirical vendor-selection application.
No takes yet. Share an insight, caveat, or question.
Chen et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: