As long as the design of a fuzzy logic controller depends mainly on rule base it is considered the core of the fuzzy logic system, and is built by designers based on their experience and/or some experiments. Rules reduction is on the most important issues in the field of fuzzy control. This thesis introduces a Genetic-based method for rule base reduction according to Pittsburgh approach. With the specific structure of the chromosome, and the adequate fitness function, the proposed method with Genetic Algorithm produces a fuzzy rule base with small number of rules. Two examples were conducted to examine the applicability of the design methodology; the first is for an inverted pendulum as Single Input Single Output system and the second is for a two link mechanical arm of robot system as Multi Input Multi Output. The obtained results show the applicability of the presented methodology. Comparison to other design approach with respect to some well known dynamic performance characteristics like overshoot, rise time, settling time and steady state error (S.S.E) are also presented. Finally, the robustness of the proposed controller is tested for a continuous step disturbance tracking input of different levels and it shows good tracking ability.