This document presents a study that uses an Imperialistic Competitive Algorithm (ICA) to optimize both real power loss and voltage stability limit in power transmission systems. The algorithm is applied to a New England power system model with 39 buses. It simultaneously optimizes transformer tap positions, UPFC location and injection voltage. Simulation results show the ICA approach reduces real power loss and improves voltage stability compared to a prior Bacteria Foraging Algorithm approach. The dual objectives of minimizing losses and maximizing stability are achieved through the evolutionary optimization of the ICA.