AI-driven business analytics for sustainable supply chain management in African agribusiness: a PLS-SEM and machine learning approach
Wisdom Edem Drah, Tingyu Lu, Richard Bless EdziPurpose
This study examines the associations between artificial intelligence capabilities, supply chain performance, and sustainability outcomes in African agribusiness. Drawing on the resource-based view and dynamic capabilities theory, it investigates the direct relationships between AI capabilities and sustainability outcomes, as well as the mediating role of supply chain performance in translating AI capabilities into environmental and resource efficiency gains.
Design/methodology/approach
Survey data were collected from executives and managers across large-scale farms and agricultural multinationals operating in six African countries. The conceptual model and hypotheses were tested using Partial Least Squares Structural Equation Modeling and machine learning algorithms.
Findings
AI capabilities are positively associated with supply chain performance, which in turn corresponds with improved environmental impact, resource efficiency, sustainability reporting, overall sustainability, and general AI perception. Mediation analysis suggests that supply chain performance serves as a transmission mechanism linking AI capabilities to sustainability outcomes. Machine learning models achieved strong predictive accuracy across most sustainability dimensions.
Research limitations/implications
The cross-sectional design limits causal inference. The sample is drawn from six African countries, which may restrict generalizability. Future research should employ longitudinal designs and expand to other emerging economies.
Originality/value
This study extends the Resource-Based View and Dynamic Capabilities Theory to AI-enabled sustainable supply chain management in African agribusiness, an empirically under-researched context. It delineates both direct and indirect pathways through which AI capabilities relate to sustainability outcomes and introduces a dual-analytical approach combining PLS-SEM and machine learning for explanatory and predictive insights.