When fuel cell modules are built from individual components, the components must be assembled according to certain rules based on manufacturing tolerances. As production increases, computer implementation of selection algorithms is essential for the speedy, efficient use of the available components. Several greedy algorithms are presented which quickly produce an assembly schedule that maximizes the number of components used from a particular inventory. These algorithms use both stepwise and one-stage approaches to the larger assembly, and some ``look ahead'' to the next stage in order to further maximize the number of components used. Once these approaches yield results, a genetic algorithm can be used to further optimize the production schedule. Results are presented for real-world data which compare very favorably with procedures currently in practice.