Mechanical and electrical brakes have dominated the braking industry for many years and will most likely continue to do so for the foreseeable future due to their low cost and adequate operating performance, wide range of applications, vehicle engineering, civil engineering, and biomedical engineering. Simple mechanical drum brake and magnetorheological (MR) fluid brake have presented in the current work. The main objective of this work is to increase braking torque, and to develop a new optimal design of MR fluid brake with better design and design control of the MR fluid design. To do so, four important steps have been accomplished. In the first step, a mathematical modeling of the conventional frictional brake and MR fluid brake has been developed to study and specify all design parameters. In the second step, a nondimensional, closedform analysis and a Taylor series expansion have used to examine the effects of perturbing dimensionless design parameters on the overall brakes performance. In the third step, two optimal designs for MR fluid brakes have been developed by taking advantage of sensitivity analysis and the design of experiments method also known as the Taguchi method. In the fourth step, controlling a MR fluid brake is performed by using two parallel PI controls for controlling the magnetic current and MR fluid thickness simultaneously. It was concluded that sensitivity analysis is a good method for identifying the parameters that have the greatest impact on brake performance and can be used as one method for the designer to obtain an optimal design. Four nondimensional design parameters were successfully used to describe the conventional frictional brake and seven nondimensional design parameters for MR fluid brake. Only two parameters for the conventional brake and five parameters for the MR fluid brake affect the performance and the others can be neglected. Two new designs for the MR fluid brake are presented and shown to be very simple in design, low in cost by removing a lot of additional auxiliaries for the frictional brake, and easy for control. By simultaneously controlling the MR fluid thickness and the electric current, a large range of brake torque is achieved without increasing the radial envelop for the brake, and saturation conditions in one controller are compensated for by the other controller. High angular velocities of the brake are primarily controlled by increasing the MR fluid thickness, while low angular velocities are primarily controlled by increasing the electric current. Good transient responses for regulating a constant speed (high, moderate, and low), and good stability while seeking to track a sinusoidal input have been achieved. In summary, the proposed control system for the MR fluid brake has demonstrated good controllability for the MR fluid brake.