In a radial distribution system, component failure leads to service interruption and subsequent disconnection of a few load points. An abnormal variation in the voltage levels and high power loss gives rise to reliability issues. In this work, feeder reconfiguration is performed to improve reliability, minimize voltage deviation and power loss. A Binary Particle Swarm Optimization based search function handles the reconfiguration. The goal of this search function is to identify the optimal tie switches which minimize the multi-objective function. The search algorithm based on Binary Particle Swarm Optimization is utilized to handle the reconfiguration which identifies the optimal tie switches for minimizing the multi-objective function. The effect of the various algorith parameters on the convergence rate is studied. The proposed algorithm has been tested on Standard IEEE 33 and 69 test systems and the obtained results show a boost in the system reliability.