The so-called lumped reaction networks are extensively used to model complex processes such as hydrocracking. Despite this, studies on the further applicability of these networks during a reactor scale-up and design are notably sparser. The application of a lumped reaction network to solve such problems requires dealing with a wide range of uncertainties, for example, reaction kinetics, the heat of reaction, or pseudocomponent densities. In this work, the design procedure of a trickle-bed hydrocracking reactor with multiple catalyst layers is carried out using a few-step lumped reaction network. The uncertain parameters are considered in a stochastic objective function using uniform probability distributions. Moreover, we extend this approach to catalyst deactivation as well, pointing out that this phenomenon can also be interpreted as a form of uncertainty, instead of estimating the activity using more complex and resource-intensive dynamic simulations. The results obtained by the application of the stochastic design method are compared to the performance of the conventional model-based design as well. An improved test of robustness is applied to evaluate the performance of the reactors under various uncertain conditions. The results indicate that the application of the suggested methods can simplify the structure of the hydrocracking reactor. For example, a fewer number of catalyst layers will be required while retaining the robustness of the reactor at the same time.