This article proposes an approach to quantify and manage uncertainty in the conceptual design of aerial vehicles. A growing number of aerial vehicles, such as unmanned aerial vehicles (UAVs), are being wholly or partly designed with off-the-shelf components to streamline development and certification costs. The design optimization of systems typically relies on component design models in the continuous domain. In addition to the inaccuracies in the design models, the continuous domain condition is inconsistent with the finite number of components available in the market. This inconsistency yields uncertainty about the actual characteristics of the components defined by the continuous design process, hence inaccurate performance predictions at the system's level. The proposed approach is developed within the scope of the conceptual design of multirotor UAVs and covers the uncertainties related to both the design models and off-the-shelf components availability. Simple criteria are established to determine the importance of uncertainties according to the density of products in an off-the-shelf component database. The uncertainties are propagated through the system's models to evaluate the variance of the outputs and identify the most critical design parameters. A multidisciplinary design optimization framework enabling custom, off-the-shelf, and hybrid design is introduced and illustrated with a multirotor UAV case study. This case study demonstrates that switching from a custom design to an off-the-shelf selection for the most critical component, namely the propeller, results in a 69% reduction in the standard deviation of estimated hover endurance. The results obtained demonstrate that the proposed approach can contribute to the emergence of future aerial vehicles and systems that must meet demanding economic and performance targets by combining custom and off-the-shelf components in their design.