This work is focused on applying dynamic simulation models to optimize the performance of an industrial process for the selective oxidation of ethylene to ethylene oxide in an externally cooled fixed bed reactor. The chemical reactions involved in the synthesis of ethylene oxide are highly exothermic, thereby making the internal temperature control of this unit a challenging task. The objective is to identify and test the critical variables to maximize the production of ethylene oxide product in an existing full-scale production facility without compromising plant safety, personnel hazards and equipment integrity. The sets of nonlinear kinetic rates of the epoxidation of ethylene (and other side reactions that affect the process) to ethylene oxide over the silver catalyst are strongly coupled with the heat and mass transport, and the process governing equations. The model incorporates both catalyst deactivation and the effect of the moderating agent 1, 2-dichloroethene (DCE). The model predictions were validated against plant data obtained from an industrial ethylene oxide reactor. The model was used for process sensitivity analyses and for the optimization of the process. In particular, the thermal stability, feed composition and heat integration aspects were found to be of crucial importance for the process.