Contemporary approach in earthquake engineering tends to dissipate some part of seismic energy by additional fuses. Owing to its easy-to-produce and plug-and-play characteristics, energy dissipative steel cushion (SC) can be a prominent candidate to serve this purpose. Since its closed form design equations are already available in the literature, seismic performance and energy dissipative characteristics of SC might be improved by optimal sizing. Hence, distinct mathematical optimization techniques, namely sequential quadratic programming, gradient-based method, and Lagrange multiplier method, are employed. Results of the optimization techniques are evaluated through experimentally verified finite element analyses. Consequently, some geometric dimension ratios are provided for the optimal sizing of SCs. Comparisons between the optimization studies yielded that the gradient-based method requires fewer function evaluations to converge while the Lagrange multiplier method with a Hessian produces more accurate results.