Application of Artificial Intelligence and Ant Colony Algorithm in the Simulation of Malaria Prevention and Control

Abstract

Introduction: Malaria remains one of the leading causes of morbidity in rural areas of Angola, especially in Bié Province, where logistical constraints hinder the effectiveness of prevention and control strategies. In response, integrative approaches are needed to link community-based nursing care with computational optimization tools.
Objective: To evaluate the application of the Ant Colony Optimization (ACO) algorithm, as an artificial intelligence technique to improve malaria prevention and control strategies.
Methods: A computational simulation was developed in Python (v3.10) using networkx, pymoo and matplotlib libraries. The model integrated three variables: distribution of nursing staff, frequency of home visits, and educational activities. Two scenarios were compared (random planning vs. ACO-optimized planning) in a simulated population of 120 people across 10 rural communities.
Results: The ACO-optimized planning reduced the total travel time from 39.2 to 27.0 hours (−31.1 %), increased home visit coverage from 73 to 110 people (+50.7 %) and improved care in priority areas from 46 % to 82.5 %. Educational interventions increased adherence to antimalarial treatment by 18 % to 35 %.
Conclusions: The implementation of the ACO algorithm proved effective in enhancing coverage, optimizing resource allocation, and strengthening the community-based response to malaria, contributing to the development of sustainable and replicable strategies in other highly endemic settings.
Keywords: malaria; territorial planning of community nursing; medical informatics applied to public health; health intervention management; direct community intervention; optimization algorithms; logistics efficiency.