An optimal design strategy based on genetic algorithms (GA) is proposed for nonlinear hysteretic control devices that prevent pounding damage and achieve the best results in seismic response mitigation of two adjacent structures. An integrated fuzzy controller is used in order to provide the interactive relationships between damper forces and input voltages for MR dampers based on the modified Bouc-Wen model. Furthermore, Linear Quadratic Regulator (LQR) and H2/LQG (Linear Quadratic Gaussian) controllers based on clipped voltage law (CVL) are also used to compare the results obtained by fuzzy controller. This study employs the main objectives of the optimal design that are not only to reduce the seismic responses but also to minimize the total cost of the damper system. A set of Pareto optimal solutions is also conducted with the corresponding results obtained from the optimal surface of Pareto solutions in this study. As a result, decreasing the number of dampers does necessarily increase the efficiency of the system. In fact, reducing the number of dampers for the dynamic response of the system can contribute more than increasing the number of dampers.