Two-Phase Stochastic Problem for Nebulizer Vehicle Routing

Dengue poses a major public health challenge in Brazil, and controlling the Aedes aegypti mosquito through fogging vehicles is one of the main strategies to curb its spread. This work proposes a two-stage Stochastic Programming approach for planning the routes of these vehicles, extending the Orienteering Arc Routing Problem (OARP) by treating the benefits of fogging, proportional to the number of reported cases, as random variables. The first-stage decision defines the current route, while the second-stage decisions react to each future scenario. We present an Integer Linear Programming formulation, a constructive heuristic, and an analysis of the EVPI and VSS indicators. Computational experiments with real data from Alto Santo and Limoeiro do Norte (Ceará, Brazil) show that combining the bounds from the recourse problem, EEV, and WS reduces the average gap from 54.98% to 30.47%.