When designing logicallinguistic controller, it is the most important to ensure a semantic order relationship between linguistic elements. HA (Hedge Algebra) [1] [2], [3] is an algebraic structure based on a linguistic rule whose semantics assure order relations. Based on HA, there are many applications that have been developed such as fuzzy database,classification, control, …[1], [3]. Because HA allows building a model calculation for LRBS-based controllers, it can beapplied in the field of control and automation [3]. According to the HA, linguistic terms are computed based on their semantic value which is always order and suitable for constructing computational models in approximation problems. The HA controllers will have the advantages of the fuzzy control, while promoting the advantages of computation based on the semantic validity of the language and the problem of intuitive thinking. In this study, we present a method of designing the controller based on the approach of Hedge Algebras (HA) that mentioning the optimization of the controller parameters by Genetic Algorithm (GA). To illustrate the effectiveness of the design method we apply the problem of slab temperature control with assuming the mathematical model of the slab is as the transfer function model.