Nansukusa, YudayaAsikuru, SalamaKalyankolo, UmaruNafuna, Ritah2026-06-202026-06-202026-06-15Nansukusa, Y., Asikuru, S., Kalyankolo, U., & Nafuna, R. (2026). Modeling aflatoxin risk dynamics in Uganda’s groundnut value chain: A System dynamics decision support approach. Scientific Journal of Computer Science, 2(2), 265-275.3110-3170https://dir.muni.ac.ug/handle/20.500.12260/1000This research on aflatoxin risk modeling helps achieve SDGs 2, 3, and 8 by supporting food safety, public health, and economic growth. It contributes to transforming Uganda’s agri-food system by guiding agricultural research, post-harvest handling, storage, and quality standards. The results show that improving post-harvest practices is the best long-term way to control aflatoxin, highlighting the need for ongoing investment rather than quick fixes. The model is a practical tool for policy-makers, suggesting that post-harvest infrastructure should be a priority, interventions should have dedicated budgets, and resource allocation should use flexible, data-driven approaches.Aflatoxin contamination remains a persistent threat to food safety, public health, and trade in Uganda’s groundnut value chain, where a large share of household and market samples exceed national and international safety limits. Despite sustained investment in awareness campaigns, improved storage, and biocontrol products, contamination remains high and unevenly controlled, in part because interventions are typically evaluated in isolation and are rarely supported by dynamic tools that capture the feedback, delays, and trade-offs linking climate, farmer behaviour, institutional support, and markets. This study develops and analyses a System Dynamics (SD) decision-support model of aflatoxin risk in the groundnut value chain, framed within an Information Systems view of simulation-based decision support. Causal loop diagrams constructed in Vensim PLE and a stock-and-flow model implemented in STELLA Architect represent the reinforcing and balancing feedback structures governing contamination, including the “Shifting the Burden” archetype. Scenario simulations and a one-at-a-time sensitivity analysis show that symptomatic measures such as awareness campaigns deliver only temporary relief, whereas post-harvest practice quality emerges as the highest-leverage parameter; a realistic mixed-policy scenario that combines moderate investment across practices, awareness, and storage technology drives contamination below regulatory thresholds within the simulated horizon. These findings indicate that durable mitigation in low-resource settings depends on sustained structural investment rather than reactive fixes, and they demonstrate how SD modelling can guide adaptive, evidence-based food-safety policy.enSystem Dynamics modellingAflatoxin contaminationGroundnut value chainInformation SystemsFood-safety decision supportArticle