Because of the transmission of disease producing parasites to man and animals and the nuisance caused by bites from adult black flies, they are the object of large control programs in many countries in the world. Because of the extensive movement of the adult flies, these programs usually involve the treatment of large numbers of breeding sites dispersed over vast areas, and the use of commercial formulations of larvicides. In this paper we present an operational larvicide treatment strategy designed to substantially reduce the treatment costs on two levels. First, we attempt to minimize the number of injection points along rivers and the mass to apply at each, constrained by a certain level of mortality to be achieved along each river. This optimization is performed by combining a mortality model of larvae, a diffusion-transport equation that models the behavior of the product in watercourses and a shortest path model. The result allows the identification of groups of injection points through which helicopters will be routed to deploy different types of larvicide. Second, we seek to optimize the circuits traveled by the helicopters. This problem is modeled and solved by state-of-the-art computerized heuristic method. The algorithm presented will enable users to construct, for each helicopter, circuits spanning several days. These circuits account for the type of larvicide used, refueling points and overnight cities, fuel and larvicide capacities of the helicopters and the maximum flying time per day. The performance of the algorithm was successfully benchmarked on an actual three days circuit for one helicopter, issued by the Onchocerciasis Control Program (OCP) aerial operations in Western Africa. The results generate potential savings of $100, 000 for the year per helicopter. Given these substantial potential benefits, a computer package developed by the authors and embedding a first version of this procedure is presently being validated in the Onchocerciasis Control Program in Western Africa.