Artificial intelligence in Zambian education: A systematic literature review and conceptual framework for sustainable resource allocation

Abstract

Zambian educational institutions face persistent challenges in allocating material, financial, and human resources fairly and efficiently. These challenges include uneven access to infrastructure and learning materials, staffing constraints, weak information systems, and limited capacity for data-informed planning. Recent scholarship suggests that artificial intelligence (AI) can support forecasting, decision support, monitoring, and optimisation, while also highlighting risks associated with data quality, infrastructure, algorithmic bias, privacy, and unequal institutional capacity. This article presents a systematic literature review (SLR) examining the relevance of AI to sustainable educational resource allocation, with particular attention to the Zambian context. Searches were conducted across Scopus, Web of Science, the Education Resources Information Center (ERIC), and Google Scholar for peer-reviewed journal articles published between 2021 and 2026 and were structured and reported with reference to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines. The final coded corpus comprised 180 studies. The synthesis identifies three principal application areas, predictive analytics, teacher allocation, and infrastructure monitoring, alongside supporting domains such as financial planning and policy simulation. Drawing on the Resource-Based View, Dynamic Capabilities Theory, and AI and machine-learning perspectives, the review develops a conceptual framework that integrates AI-enabled allocation with governance, human oversight, contextual adaptation, and accountability. The review indicates substantial potential for AI-supported resource management, but the evidence does not justify assuming automatic gains in efficiency or equity. For Zambia, implementation is likely to depend on reliable data, connectivity, institutional capability, professional development, appropriate governance, and locally adapted models. The proposed framework therefore provides a basis for subsequent empirical validation rather than a claim of demonstrated implementation effectiveness.