Artificial Intelligence in Supply Chain Management: State of Art and Future Research Directions

Abstract

Artificial Intelligence (AI) has had a profound effect on supply chain management, offering better forecasting, resource planning, inventory management, and operational decisions. Rarely, though, operational data are uniform and aggregated from various parts of the supply chain, and may suffer from inconsistencies which impact prediction accuracy and confidence in decisions-making. This paper seeks to broadly review the claim of AI in the area of supply chain and introduce a Reliability-Aware Multi-Source Feature Fusion (RAMSF) framework to enhance intelligent decision generation. The framework aims to combine the heterogonous data from the supply chain via their structured pre-processing, source representation of features, feature fusion between two or more data sources, predictive modeling using AI, and verification of the prediction reliability, in order to produce a consistent operational decision-making. All the feature integration methods neglect one or multiple aspects of reliability verification and representative feature construction which are important for RAMSF. The framework aims to enhance prediction accuracy, computational speed, and resource utilization, and to ensure consistency in the managerial process in various supply chain scenarios. Moreover, the study covers ongoing research trends, prevailing challenges, and potential opportunities for future research, which serve as a solid basis for creating effective AI-based supply chain management solutions. The suggested system achieved a Decision Reliability Index (DRI) of 96.18%, signifying a highly reliable and consistent generation of supply chain decisions.