Uncertainty has a significant impact on the stability and performance of engineering-to-order (ETO) production systems by affecting due date achievement, efficient resource allocation and the usage of nonregular working force. Proactive planning approaches consider information about uncertainty to protect the generated plans against future disruption. Likewise, this paper´s goal is to propose and apply a robust optimization model that addresses a tactical capacity planning problem by incorporating uncertainties regarding the duration of production processes. The model aims at solution robustness (or stability) and intends to enhance and support the decision-making process of a real-world ETO industrial setting. Monte Carlo simulation is employed to assess the simultaneous impact of multiple constraints on the robustness of the generated solutions. The computational results are adherent to the industry setting studied and show the effects of the conservatism level on the optimal values of the generated plans. To the best of our knowledge, this is the first robust optimization model for tactical capacity planning that explicitly addresses the ETO context.