Load shedding is considered as a last alternative to avoid the cascaded tripping and blackout in power systems during generation contingencies. It is essential to optimize the amount of load to be shed in order to prevent excessive load shedding. To minimize load shedding, this paper proposes the implementation of nature inspired optimization algorithm known as glowworm swarm optimization (GSO) algorithm. The optimal solution of steady state load shedding is carried out by squaring the difference between the connected and supplied power (active and reactive). The proposed algorithm is tested on IEEE 14, 30, 57, 118 and Northern Regional Power Grid (NRPG)-(India) 246 bus test systems. The viability of the proposed method in terms of solution quality and convergence properties is compared with the conventional methods, namely, projected augmented Lagrangian method (PALM), gradient technique based on Kuhn-Tucker theorem (GTBKTT) and second order gradient technique (SOGT). © 2014 Production and hosting by Elsevier B.V. on behalf of Ain Shams University.