With the current improvement in the capabilities of mathematical simulation models, these models can be used for control applications. It opens a new methodology for controlling glass furnaces in view of optimal glass quality. The method is based on three steps: from classical theories a set of basic rules can be derived that roughly describe the relation between furnace design, glass properties and operation settings on the one hand and the glass quality on the other hand. These relations are over-simplified, but they form a good basis for designing control strategies and controllers. In the third step the control strategy is 'fine-tuned' by running the controller against a mathematical simulation model of the furnace. The result is an intelligent controller that contains a high degree of glass melting know-how.