AI and big data boost humanitarian logistics in Ghana and South Africa

Artificial intelligence (AI) and big data analytics (BDA) significantly improve humanitarian supply chain resilience in developing countries, according to a new study. Titled “Enhancing Humanitarian Supply Chain Resilience: Evaluating Artificial Intelligence and Big Data Analytics in Two Nations” and published in the journal Logistics, the research assesses AI-BDA applications in Ghana and South Africa, two countries facing frequent disaster-related disruptions.

Authored by Emmanuel Ahatsi and Oludolapo Akanni Olanrewaju of Durban University of Technology, the study employs a large-scale quantitative approach to identify how specific AI-BDA techniques influence disaster response logistics. Analyzing data from 200 humanitarian supply chain professionals, the study evaluates four core tools: time-series forecasting (TSF), early warning systems (EWSs), logistics optimization (LO), and real-time monitoring (RTM). It offers critical insights for humanitarian organizations, policymakers, and global development agencies working to modernize supply infrastructure in crisis-prone regions.

Which AI and big data techniques are most common in humanitarian logistics?

Humanitarian supply chains in Ghana and South Africa are increasingly turning to AI and BDA to address long-standing issues of inefficiency, resource misallocation, and slow disaster response. Among the AI-BDA techniques, logistics optimization and time-series forecasting emerged as the most commonly implemented. Multi-objective optimization models for resource allocation and demand prediction systems received high adoption scores, indicating widespread recognition of their practical value.

However, early warning systems, despite their known effectiveness in forecasting floods and positioning relief supplies, showed relatively limited implementation. Similarly, real-time monitoring technologies for tracking relief goods and personnel remained underutilized, largely due to infrastructure limitations and high costs of integration. These findings reveal a gap between the known utility of these tools and the actual deployment capacities within many humanitarian operations.

The researchers found that while organizations are beginning to integrate AI-driven forecasting models and optimization algorithms, several advanced technologies, such as deep learning-based EWSs and blockchain-enhanced RTM, remain largely out of reach for resource-constrained humanitarian actors.

How do these technologies impact supply chain resilience?

The study demonstrates a statistically significant positive impact of all four AI-BDA techniques on humanitarian supply chain resilience (HSCR). Regression analysis confirmed that time-series forecasting and logistics optimization have the strongest effects, improving decision-making accuracy, inventory management, and distribution speed. These tools enable organizations to better anticipate demand, preposition critical supplies, and rapidly adapt to on-the-ground changes during emergencies.

Early warning systems also contribute meaningfully by enhancing preparedness through accurate disaster risk prediction, although their effect was slightly lower, possibly due to incomplete data integration and algorithm complexity. Real-time monitoring technologies showed solid performance in increasing operational visibility, particularly when combined with mobile and IoT solutions, though technical and budgetary barriers continue to hinder widespread adoption.

Moreover, the study highlights the role of technology readiness and organization size as important moderating factors. Larger organizations with advanced digital infrastructure and a culture of innovation are more likely to successfully implement and benefit from AI-BDA systems. Conversely, smaller and less digitally mature agencies may struggle with adoption, limiting the overall impact on HSCR.

What strategic actions are needed to improve adoption and impact?

The research recommends targeted interventions to accelerate the effective use of AI and BDA in humanitarian logistics:

• Capacity Building: Humanitarian organizations should invest in training programs to build technical competencies among supply chain managers, data scientists, and IT staff. Partnerships with universities and tech firms can facilitate the creation of AI-literate teams capable of sustaining long-term innovation.

• Infrastructure Development: Governments and international donors must prioritize the development of digital infrastructure, especially in rural or disaster-prone regions, to support the integration of real-time monitoring and early warning systems.

• Data Quality and Governance: Ensuring access to clean, reliable, and interoperable data is essential for the successful deployment of AI and BDA tools. Establishing ethical frameworks to address concerns around privacy, bias, and algorithm transparency is also critical.

• Multi-Stakeholder Collaboration: Cross-sector partnerships involving humanitarian agencies, tech providers, and affected communities can facilitate context-specific, ethically grounded implementations of AI and BDA.

The study also calls for further research to include diverse humanitarian contexts and longitudinal studies that track the evolving impact of digital innovations over time. These efforts would help refine existing models and tailor solutions to different socio-political and environmental settings.

To sum up, for nations like Ghana and South Africa, where the stakes of disaster preparedness are high, investing in AI-BDA capabilities is no longer optional - it is essential.