Artificial intelligence in supply chain decision-making: a systematic review of models, applications, and implementation challenges

Abstract

This systematic review examines the application of Artificial Intelligence (AI) in supply chain management from 2010 to 2024. Using the PRISMA approach, it analyzes 142 peer-reviewed studies covering machine learning, deep learning, natural language processing, reinforcement learning, and expert systems across areas such as demand forecasting, inventory management, supplier selection, logistics, and disruption response. The findings show that AI improves decision-making, predictive accuracy, and real-time responsiveness, but adoption is challenged by poor data quality, lack of explainability, legacy system integration, cybersecurity risks, and shortages of skilled professionals. The review also highlights concerns about data bias, transparency, and algorithmic accountability. It recommends future research focusing on interpretable AI, cross-functional integration, and supply chain resilience.