AI in supply chain: Techniques, applications, real-world cases and benefits under SCOR framework

Abstract

This article focuses on practical perspectives of Artificial Intelligence (AI) applications in Supply Chain Management by exploring commonly used AI techniques, use cases and benefits of applying AI in Supply Chain Management with real-world examples from multinational corporations like DHL, IBM, Walmart, Amazon, Google, among others. The findings are grouped according to the four stages of the SCOR (Supply Chain Operations Reference) framework, i.e plan, source, make, deliver, to facilitate visualization. We find that AI techniques including Neural Networks, Genetic Algorithms, Support Vector Machines, Reinforcement Learning, Fuzzy Logic, and Natural Language Processing are applied to enhance supply chain efficiencies, lower costs, increase profits, improve customer satisfaction, save operational time, reduce potential disruption, better suppliers/customers relationships, improve product quality, enhance safety, and shorten lead times… These stem from nine benefit groups, namely PLAN (demand forecasting, inventory optimization, supply risk mitigation), SOURCE (procurement, supplier selection), MAKE (product quality assurance, smart warehouse management, predictive maintenance), DELIVER (route optimization, dynamic pricing, and last mile delivery, and customer service). Limitations and future research directions are discussed.