We present an application of optimal process synthesis approach to heterogeneously catalysed gas-phase reaction systems. It enables the systematic identification of optimal conceptual process designs, the exploration of the relationships between design complexity and performance and the execution of subsequent synthesis stages that enrich the reaction models to incorporate detailed representations of phenomena so that the process designs can be evolved into optimal schemes that resemble reality closely. The technology is applied to evolve a process for the production of acetic acid via ethane oxidation, an industrially relevant process. Throughout the multi-level design cycle, information on the optimal operating envelopes is generated and can be fed back to the kinetics development team to guide additional experiments so as to ensure that kinetic models match the optimal process in which the catalyst is to be used. The evolution takes the form of an iterative process performed in multiple stages.