Microgrids are small-scale power systems that include local generation and storage units to serve their loads. The effectiveness of such systems depends on both sizing and operations, that need to be efficient to minimize costs while ensuring reliable power delivery. In this paper, we build a stand-alone microgrid while considering not only electric power, but also cooling, heating, and hydrogen consumption. A unit commitment algorithm, formulated as a mixed integer linear programming problem, is used to determine the best operation strategy of the system. A genetic algorithm is used to search for the best size of each component. Results show the feasibility of the proposed sizing method.