Biopharmaceutical products will be important drivers of improving medical treatments and the standard of living in the 21th century. The development and commercialization of biosimilar monoclonal antibodies is a major milestone to maximize patient benefit, as these products deliver equivalent clinical effect at lower costs than the originator product. One of the major challenges in biosimilar development is to adjust the quality profile of the drug substance with technological control strategies in a tight range which is ultimately determined by the characteristics of the originator product. This goal is in accordance with the quality by design (QbD) paradigm, which demands the thorough understanding of interactions between process parameters and product quality attributes to assure consistent process output. This thesis focuses on the implementation of QbD tools into biosimilar process development aiming to understand links between process parameters and product quality attributes in a fed-batch recombinant CHO process producing a monoclonal antibody. In order to achieve this targeted knowledge, four essential steps have been identified and accomplished. First, standard risk assessment tools were tailored to address the above discussed unique characteristics of biosimilar development. These novel tools allowed a risk-based investigation of the complex interactions between critical process parameters (CPPs) and critical quality attributes (CQAs). Thereby, physiological features of the production cells were identified to have a vast impact on these interactions. Accordingly, the second step was to develop workflows for the quantification of physiological variables on the level of cell metabolism and to investigate the effect of multiple CPPs on these variables. Thereafter, novel control strategies were developed to steer physiological features such as the metabolic switch to lactate uptake as well as specific productivity in the fed-batch process. The control of physiological variables enabled the combination of standard QbD tools with the physiological approach. Consequently, the final step was to involve the controlled physiological features as input and output variables in design of experiment (DoE), multivariate data analysis (MVDA) and process analytical technology (PAT) tools. The essential novelty of the presented work is the combination of QbD tools with the quantification of physiological variables in cell culture process development. This approach enables the generation of enhanced process understanding and the development of a novel control strategy to adjust a CQA of the product. The anticipated benefit of the presented workflow over conventional QbD approaches is the identification of novel CQA control strategies based on sound process understanding, an aspect especially relevant for biosimilar development.