Measuring the Resilience of Additive Manufacturing Supply Chains Based on a Cloud Model: An Integrated Weighting and Cloud Centroid Method

Building resilient additive manufacturing supply chains (AM-SCs) has become a strategic imperative as firms confront increasing uncertainty, technological disruption, and volatile market environments. However, existing resilience evaluation approaches face three critical limitations: (i) insufficient integration of qualitative and quantitative indicators, (ii) biased weighting methods that rely solely on expert judgement or data-driven metrics, and (iii) limited interpretability and diagnostic value for managerial decision-making. To address these gaps, this study develops a cloud-based resilience measurement framework tailored to AM-SCs. A five-dimensional, 26-indicator evaluation system is constructed to reflect resistance, recovery, adaptability, coordination, and learning capabilities. An integrated weighting approach — combining ordinal-relation and entropy weighting — balances expert knowledge with data variability, while the cloud model generates digital features (Ex, En, He) to characterize uncertainty in mixed-type indicators. The centroid-to-ideal deviation method is introduced to quantify resilience levels and identify performance gaps. Empirical results from two AM-SC firms across two periods (2017–2019; 2020–2022) demonstrate improved resilience, with Firm B achieving a 52% reduction in weighted deviation and Firm H a 4–5% improvement. The framework offers actionable diagnostic insights and provides a reproducible approach for evidence-based resilience enhancement. Relative to prior resilience assessments that treat qualitative and quantitative indicators separately, our approach unifies them under a cloud model and yields interpretable deviation diagnostics. The results suggest targeted upgrades (supplier reliability, digital scheduling, training cadence) and offer a transferable template for AM supply-chain evaluation under data uncertainty.