Abstract: A hybrid genetic algorithm (HGA) composed of the standard genetic algorithm (SGA) and unidirectional searching algorithm (USA) is proposed, taking their advantages of time saving, high efficiency and powerfully local searching capability of USA, as well as the global searching capability of genetic algorithm (GA). The results by exemplification show that this method can not only be used directly to work out the optimum structural design with discrete variables to constrain both stress and cross-section area but deal with the discrete structural optimization featured with multi-loading, multi-constraints and multi-variables. HGA, as an efficient optimal method having advantages of both SGA and USA without their disadvantages, is superior to either one.