AI-Driven Big Data Analytics for Supply Chain Resilience

Abstract

Over recent years, natural disasters, the COVID-19 pandemic, and geopolitical tensions have highlighted vulnerabilities within global supply chains. Breakdowns in production, logistics, and distribution have highlighted a need for resilience – the ability to prepare for, respond to, and recover from disruptions. Artificial Intelligence (AI) and data science offer one pathway to resilience, helping to improve demand forecasts, optimize inventory policies, manage supplier and customer ecosystems, predict events, and provide asset and risk intelligence. Nevertheless, despite partnering to deliver the next generation of ground-breaking intelligent systems, traditional AI techniques cannot learn nothing without data. Supply chains can be considered Big Data Ecosystems, as vast quantities of internal and external data are sourced from multiple systems and tiers, and flow in all directions.

By extending big data concepts with empirical research, evidence is provided to support the development of a modern architecture and testable theoretical framework for AI-augmented data analytics. It argues that without addressing fundamental data issues, the application of cutting-edge AI techniques will be limited in scale and impact, concentrated on the forecasting and synthetic production approach, rather than everything that resembled production problems through the Supply Chain Management (SCM) area. Furthermore, only a small part of the potential value creation hidden within big data will finally be realized. An assessment of industry case studies and empirical applications illustrated that the framework can be deployed across multiple sectors. However, in order to be successfully implemented, it must be properly completed by the supporting data infrastructure.