This research presents the way of improving the performance of the distribution network byrnimproving the voltage profile and reduces the power loss by integrating D-STATCOM to Kotebe K3rndistribution feeder. Kotebe distribution substation is one of the distribution substations found inrnAddis Ababa, Ethiopia feeding areas around Kotebe, Wesen, CMC, Figa and Gurdshola. Amongrnthose feeders, Kotebe K3 has been selected as test system due to its longest rout length, large loadsrnand high permanent power interruption rate. The selected feeder has been modeled in MATLABrncomputational tool and load flow analysis has been simulated using Newton-Raphson method. It hasrnbeen observed that the Newton-Raphson load flow simulation of Kotebe K3 gives a power loss ofrn1.7351 MW and 0.7080 MVAR respectively and 41 buses below the expected voltage range. rnThis research uses a hybrid of Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) forrnoptimal placement and sizing of D-STATCOM for power loss reduction and improvement of voltagernprofile. The GA-PSO optimization technique considers both real and reactive power losses. Therncombined sensitivity factor uses both real and reactive sensitivity factors to identifying the candidaternbuses for D-STATCOM allocation. The combined sensitivity factor identifies bus 22,23,24,25 and 26rnas the most five sensitive buses, which are used for placement of D-STATCOM. The GA-PSOrnoptimized result shows that bus 22 is the most effective bus for placing D-STATCOM in terms ofrnreducing the power loss at relatively small size and keeps the bus voltage under the acceptable range. rnThe simulation result of the integrated D-STATCOM on bus 22 reduces the real power loss fromrn1.7351 MW to 1.0948 MW, which is by 36.90%, and the reactive power loss from 0.7080 MVAR torn0.4453 MVAR, which is by 37.11%, while maintaining all bus voltages in the range between 0.95rnand 1.05pu. Furthermore, a comparison analysis shows that GA-PSO method gives the greatestrnreduction in both real and reactive power loss compared to GA and PSO methodologies.