In Colombia, Aedes aegypti is resistant to most used insecticides. Due to the slow development of resistance to Bacillus thuringiensis var. israelensis (Bti) as well as its high specificity and environmental safety, the use of this larvicide becomes an alternative in the management of this vector. In this work, we evaluated experimentally and describe by a mathematical model the dyamics of control of natural populations of A. aegypti using Bti. The susceptibility profile to Bti was determined through dose-response bioassays with larvae collected in Armenia (Quindío, Colombia). In addition, in order to estimate the susceptibility profile, an analysis was carried out using the mortality data obtained from the four localities analyzed. The mortality data were used to estimate the lethal concentrations (LC50 and 95) from each locality. Using these LC, hypothetical simulations of vector behavior were carried out, obtained from a mathematical model that describes the population dynamics, using successive applications of Bti at different time intervals. The dose-response bioassays indicate tha the analyzed vector populations are susceptible to Bti since they present a biological response similar to the one obtained from A. aegypti Rockefeller strain. Computer simulations using sustained periodic inspection indicate Bti is effective for the vector control. Nevertheless, its long-term efficiency depends on relation between the LC and the frequency of application. It is concluded that the sustained application of Bti represents a longterm viable alternative for the control of A. aegypti populations.