Resilience in Humanitarian Supply Chains: A Review of AI-Driven Big Data Analytics Applications

Abstract

Humanitarian supply chains face unprecedented challenges due to escalating global crises, with approximately 279 million people requiring assistance in 2022 alone. The traditional reactive approaches to humanitarian logistics are becoming ineffective at managing the increasingly complex and large-scale disaster responses. In this paper, we analyzed the coupling of artificial intelligence (AI) and big data analytics (BDA) as transformative levers for improving the resilience of humanitarian supply chain operations. With a focus on disaster forecasting, resource allocation, and stakeholder coordination in humanitarian contexts, we study how AI-driven predictive modelling and BDA will change the narrative. This paper identified several significant implementation barriers, including data quality issues, infrastructure limitations, and ethical concerns related to algorithmic bias and privacy. Theoretical frameworks for using AI-BDA for humanitarian logistics exist, but examples of real-world effectiveness are limited. The paper aims to fill this research gap by evaluating some AI-BDA techniques focusing on time-series forecasting, early warning systems, logistic optimization, and real-time monitoring in humanitarian settings and offering evidence-based recommendations for implementation. The strategic integration of these technologies may be leveraged to transition from the reactive to the more proactive humanitarian response models, thereby improving aid delivery efficiency by up to 30% during a crisis. We conclude with practical guidelines for humanitarian organizations exploring the use of AI-BDA capacity building and appropriate data governance frameworks for AI-BDA adoption.