This work proposes a simulation-based multi objective optimization model for the operation of the product separation process in a methanol to propylene (MTP) plant, striving to enhance the system energy utilization efficiency. The formation of byproducts including an oxygenated hydrocarbon, dimethyl ether (DME), makes the product refining process highly energy-intensive, not only because that the various hydrocarbon byproducts require a long train of distillation columns but also owing to the fact that DME forms azeotrope with product propylene. Furthermore, the formation of oxygenated DME byproduct varies with the catalyst activity, which makes the system more complex. The contribution of this paper is 2-fold. First, it proposes a novel way of DME removal with responsive consumption of extractant methanol according to the DME concentration during one production period. In this way, the energy consumption of the condenser and reboiler of the methanol recovery unit can be expected to decrease by 61.5% and 37.6%, respectively. Second, a simulation-based multiobjective optimizaton framework is introduced to minimize the total heating and cooling costs and maximize the product purity, by integrating a rigorous process simulation model with an intelligent optimization algorithm. Given the characteristics of the MTP production process, three different operation scenarios, namely, the start of run, the designed operation condition, and the end of run, which correspond to different DME concentrations in the system, are covered. The approach is illustrated by its application to a real MTP plant, in which numerical results indicate that the proposed method is capable of identifying appealing operation options, and it can be utilized to support decision-making of the product separation of MTP plants.