Measuring and predicting supply chain resilience through AI-based technology: evidence from the Chinese manufacturing industry

Purpose This study aims to enhance supply chain resilience (SCR) by measuring and predicting this complex concept, thereby achieving a more resilient B2B marketing performance. Design/methodology/approach This study meticulously designs a two-stage research framework grounded in artificial intelligence. It constructs an indicator system to measure SCR and designs a data collection framework to facilitate innovative acquisition of knowledge. Subsequently, an intelligent model is proposed to predict the SCR, which also identifies the significance of the key indicators. Findings Quantitative measurements of SCR not only reveal the heterogeneity across diverse sub-industries and industries with varying manufacturing technology levels but also identify five enterprise clusters, thereby enabling enterprises to better understand their unique characteristics. Comparative experiments with classical models demonstrate the superior performance of the proposed intelligent model. The model also identifies the importance of seven indicators that provide specific operational pathways for enterprises to enhance their SCR. These findings can help enterprises strengthen their supply chain capabilities and improve their B2B marketing performance. Originality/value This study achieves a scientific measurement of the complex concept of SCR through theoretical integration and innovative data collection schemes. It further investigates SCR prediction using AI technologies, an area scarcely addressed in the existing literature. This is a prominent theoretical innovation. Focusing on the manufacturing industry, this study provides targeted suggestions for enhancing SCR and ensuring its application value in the B2B market.