Agent-based modeling for mosquito control

Abstract

Malaria infection remains a major health challenge in many parts of the world, especially in sub-Saharan Africa. Although intervention strategies abound, the efficiency of control measures emanates from understanding the host-seeking behaviors of the malaria vector species. Discrete agent-based modeling was employed to demonstrate the protective efficacy of Insecticidal treated nets on Anopheles gambiae and Anopheles arabiensis by considering the treatment measure alongside the control case; with just a physical net barrier. The aforementioned modeling approach is indeed suitable in demonstrating the dissimilar host-seeking behavioral tendencies of An. gambiae and An. arabiensis when confronted with Insecticidal treated nets. The models are calibrated against several real experimental hut trial datasets by employing the adaptive Markov chain Monte Carlo. Both the control and treatment case models are able to replicate the datasets considered, virtually within the reported error bounds. Moreover, the datasets employed are able to properly identify the sampled parameters in both cases, and as such, reveal relatively low uncertainties in these model parameters.