Microgrids are known as clusters of distributed energy resources serving a group of distributed loads in grid-connected and isolated grid modes. Nowadays, the concept of microgrids has become a key subject in the smart grid area, demanding a systematic procedure for their optimal construction. According to the IEEE Std 1547.4, large distribution systems can be clustered into a number of microgrids to facilitate powerful control and operation infrastructure in future distribution systems. However, clustering large systems into a set of microgrids with high reliability and security is not reported in current literature. To fill-out this gap, this paper presents a systematic and optimized approach for designing microgrids taking into account system reliability- and supply-security-related aspects. The optimum design considers sustained and temporary faults, for system reliability via a combined probabilistic reliability index, and real and reactive power balance, for supply security. The loads are assumed to be variable and different distributed generation (DG) technologies are considered. Conceptual design, problem formulation and solution algorithms are presented in this paper. The well-known PG&E 69-bus distribution system is selected as the test system. The effect of optimization coefficients on the design and the robustness of the algorithm are investigated using sensitivity studies.