Erythropoietin (EPO) is one of the most important hormones in the human body, due to its effective pharmaceutical performance, recombinant human EPO (rhEPO) is often produced by Chinese Hamster Ovary (CHO) cells as host cells through recombinant DNA technology on an industrial scale. In this study, the reported genome-scale metabolic network of CHO cells was upgraded to integrate EPO production pathways using the INIT algorithm in the RAVEN Toolbox. After quality analysis for the reconstructed model, performance of the model was examined under two different culture conditions provided within the literature. Such analysis were implemented through Flux Balance Analysis (FBA) and Multi-objective Analysis techniques and the results highlighted the effectiveness of these culture conditions. To enhance the efficiency of rhEPO production, analysis of essential genes and reactions, sensitivity of essential amino acid supplementation and flexibility of amino acid uptakes was also performed through a series of standard in silico techniques in constraint-based analysis.