Application of Artificial Intelligence in Supply Chain Management: A Review on Strengths and Weaknesses of Predictive Modeling Techniques

Abstract

This study examines and summarizes the body for studies on supply chain management (SCM) applications of artificial intelligence (AI) predictive modeling techniques. It does this by examining 55 pertinent publications that were published between July 2024 and 2015. Sources for the literature included reputable research resources such as Google Scholar, IEEE Xplore, Web of Science, Science Direct, and Scopus. With an emphasis on machine learning algorithms, deep learning frameworks, and conventional statistical techniques, the paper demonstrates the wide variety of predictive modelling techniques used in the field. The results show that the use of machine learning techniques is becoming more common, and these techniques have been shown to offer a great deal of promise for enhancing SCM decision-making and forecasting accuracy. Deep learning frameworks are becoming increasingly potent instruments as well, especially for handling big datasets and identifying intricate linkages in the supply chain. This review provides a thorough resource for researchers and practitioners who wish to understand the advantages and disadvantages of various prediction modeling techniques in AI for SCM. It also highlights some noteworthy weaknesses, such as issues with data quality, model interpretability, and the requirement for domain-specific knowledge. Lastly, the synthesis of findings shows that although AI-driven prediction models can improve efficiency and responsiveness in SCM, their successful implementation requires careful consideration of organizational context and operational constraints.