Explainable Artificial Intelligence for Supplier Cost-Risk Assessment and Margin Optimization in Global Manufacturing Supply Chains

This study has addressed the problem of limited transparency in artificial intelligence-supported supplier assessment, where complex predictive models may identify cost risks and recommend sourcing actions without sufficiently explaining the factors responsible for those outputs. The purpose of the study has been to quantitatively examine how Explainable Artificial Intelligence capabilities are associated with Supplier Cost-Risk and Margin Optimization Performance in global manufacturing supply chains. A quantitative, cross-sectional, case-study-based research design has been employed across selected manufacturing and enterprise supply-chain environments using procurement analytics, supplier evaluation systems, AI-supported decision tools, and advanced digital sourcing platforms. Using purposive sampling, 320 questionnaires have been distributed to procurement, sourcing, supply-chain, purchasing, supplier relationship, financial, operations, and AI or analytics professionals, of which 301 have been returned and 289 have been retained as valid responses, producing a 90.3% usable response rate. The study has examined Explainable Supplier Cost-Risk Identification, AI-Enabled Cost Driver Transparency, Explainable Supplier Risk Prioritization and Decision Support, and Explainable Margin Scenario Analytics as predictors of Supplier Cost-Risk and Margin Optimization Performance. Data have been analyzed using descriptive statistics, Cronbach’s alpha, construct-validity tests, Pearson correlation, and multiple regression. Reliability coefficients have ranged from .84 to .91, with KMO = .89 and Bartlett’s Test significant at p < .001. Supplier Cost-Risk and Margin Optimization Performance has recorded M = 4.18, SD = 0.53. All four predictors have demonstrated significant positive correlations with performance, led by Explainable Supplier Cost-Risk Identification, r = .75, followed by AI-Enabled Cost Driver Transparency, r = .72, Explainable Supplier Risk Prioritization and Decision Support, r = .69, and Explainable Margin Scenario Analytics, r = .66, all p < .001. The regression model has explained 71.2% of performance variance, R² = .712, adjusted R² = .708, F (4, 284) = 175.50, p < .001, with significant standardized effects of β = .31, .28, .24, and .21, respectively. These findings indicate that explainable AI can strengthen supplier cost visibility, risk prioritization, sourcing decision quality, and margin-oriented management by making AI-supported assessments more interpretable and managerially actionable.