This paper describes the new algorithm for optimal distribution network (DN) reconfiguration through integration of cycle break algorithm (CBA) and genetic algorithm (GA). Although different approaches utilizing GA have been developed so far, they usually had issues with poor convergence rate, long computational time and were often applicable only to the small scale DNs. Unlike these approaches, the algorithm described in this paper brings a number of uniqueness and improvements that allow its application to the DN of real size with high degree of topology complexity. Optimal DN reconfiguration is formulated for the two different objective functions: minimization of total power/energy losses and minimization of network loading index. In doing so, the algorithm maintains radial structure of distribution network through entire process and assures fulfilment of various physical and operational network constraints.