We study the integrated location–transportation problem under uncertain demand, addressing the case of a pharmaceutical logistics network in Brazil. We propose a mathematical model with multi-time scales to make decisions accordingly with their timing. The model considers practical features, such as fleet sizing, safety constraints in cargo transportation, and tax issues. We address uncertainty in decision-making by developing a robust counterpart and present solution methods based on Fix-and-Optimize heuristics. We perform computational experiments using real data from a partner company and evaluate the impact of uncertainty on the problem. The heuristics outperform the MIP model by reducing the average of logistics costs and taxes by 40%. Demand uncertainty and variability significantly influence the problem decisions. Also, the robust model reduces expected solution costs. Thus, these models and solution methods can support the decision-making process on location–transportation problems in Logistics Network Planning in contexts similar to the pharmaceutical industry in Brazil.