Process development and design in a strictly regulated environment need effi cient methods to generate product quality as early as possible, and to prove the reproducibility and reliability of process operation at the production scale afterwards. For pharmaceuticals, regulatory authorities like US Food and Drug Administration ( FDA ) and European Medicines Agency (EMA) have demanded Quality by Design ( QbD ) approaches in order to fi le information -based decisions. This chapter describes the state -of -the -art of QbD by statistical design of experiments. and actual developments to improve the predictive power and accuracy in combination with a reduction of later efforts by physicochemical -based process modeling. The benefi ts are pointed out exemplifi ed by an examplea typical hydrophobic interaction chromatography step in monoclonal antibody ( mAb ) downstream processing.A model has to be generated that is suffi ciently predictive within a given process operation parameter range with a known suffi cient accuracy. Therefore, a physicochemical -based mathematical process model is derived including nonidealities like fl uid dynamics by axial dispersion, mass transfer kinetics by fi lm and pore diffusion, and multicomponent equilibrium phase by interferences.The model parameters are determined experimentally at the laboratory scale (about a few milliliters) in order to gain an acceptable experimental error of lower then ± 5%. The model ' s accuracy is defi ned with the aid of Monte -Carlo simulations, which are described in detail. Afterwards, the model is validated at the laboratory scale and production scale with a set of different operation parameters to determine the predictive range and accuracy. Finally, some parameter studies are described in order to illustrate the method, discuss benefi ts, and point out limitations.