Hybrid separations combining distillations and crystallisations have a significant potential for process intensification. To address the large number of degrees of freedom in the design of hybrid separations, a three-step approach is utilised. However, it can only be applied if all parameters for the rigorous modelling of crystallisation and cost functions are known a priori, which is often not the case. In this paper, we propose a four-step design method which can be applied in early process development stages when not all model parameters are available. In the first step, different process variants are generated. In the second step, the variants are evaluated using rigorous models, wherein the unknown model parameters are varied to quantify their influence on the process performance. If hybrid separations appear to be compatible, experiments are performed to determine the unknown parameters in the third step. In the last step, an optimisation is performed to find the optimal process, when necessary in dependence of unknown cost parameters. The developed tools and the feasibility of the approach are illustrated with the separation of a binary mixture of long-chain isomeric aldehydes. (C) 2012 Elsevier B.V. All rights reserved.