Optimization of a redundancy allocation problem designed specifically for systems with k-out-of-n subsystems has been the motive of this work. The main motive is to select the components and redundancy levels to maximize system reliability given system-level constraints. With cost and weight as constraints as 3 out of 5 subsystems has been optimized in the present study for maximizing the reliability of a mechanical system. Component reliabilities in every stage are treated as unknowns in this model. Evolutionary and Generalised Reduced Gradient method (GRG nonlinear Solver) in Excel has been applied here to determine the component reliabilities, stage reliabilities to maximize system reliability. A comparative study has been done using the methodology in the present paper with the methods already used by the other authors for the similar systems.