Enhancing supply chain resilience with artificial intelligence a bibliometric analysis and systematic literature review

Abstract

Among many innovative technical solutions, Artificial Intelligence (AI) is considered a promising approach for fostering resilient supply chains. This study aims to investigate how AI enhances Supply Chain Resilience (SCRES) by identifying AI applications in supply chain contexts, evaluating their impacts on SCRES, and uncovering gaps and emerging trends for future research. We adopted the PRISMA procedure to collect appropriate papers and then employed VOS Viewer to uncover key insights within scholarly literature. After that, selected papers underwent systematic content analysis to synthesize key concepts and applications as well as valuable aspects of AI application in fostering SCRES. The AI-SCRES relationship is an emerging field, with a notable increase in publications, particularly driven by unprecedented events like the COVID-19 pandemic. Findings showed 11 AI techniques frequently mentioned in the literature for SCRES enhancement. Bayesian networks emerged as the most discussed and mature, followed by artificial neural networks and genetic algorithms. These technologies are predominantly used for risk prediction, automated reasoning, optimization, and decision support. While AI offers substantial benefits such as enhanced decision support and demand forecasting, it also brings challenges like the need for highly skilled personnel, investment costs, and data-related risks.