Machine Learning in Supply Chain Management: A Systematic Literature Review

The supply chain ecosystem is currently benefiting from a great dynamic resulting from the digitalization of organizations and trades. For all the stakeholders in the area, this is a real breakthrough, and machine learning is at the core of this revolution. It has profoundly revolutionized companies in many aspects including the evolution of communication methods, the automation of many processes, the growing importance of information systems, etc. With shrinking margins and more demanding customers, supply chain management in increasingly becoming a source of competitive advantage. Its management and optimization requires a factual to Supply Chain decision making at strategique, tactical and operational levels. In this context and data rich environment, machine learning approaches and techniques find numerous useful applications for supply chain decision making. Today, companies have no choice but to apply Machine Learning solutions in almost every part of their processes. This fact seems even clearer in markets where competition is fierce. While Machine Learning does not redefine the enterprise, it is certainly a powerful asset for both marketing and process optimization purposes. It is so ingrained in the strategies of companies that now most of them rely heavily on it for all processes from creation, to product quality control, to public relations. In recent years, a series of practical applications of machine learning (ML) for supply chain decisions have been introduced. By interconnecting the ML methods applied for SC decision making, this paper identifies current SC applications and indicates potential research gaps. In this paper, we examine the general usability of machine learning techniques to assist with supply chain decisions. The main objective of this research is therefore to study how machine learning techniques can be integrated into the range of tools available to supply chain decision makers in order to take advantage of the increasing volume of data generated in the supply chain.