Most researches of facility layout planning (FLP) have aimed at finding a layout with which evaluation indices based on distance are minimized. Because temporal efficiency has not been considered in this stage but in post stages, the resultant temporal efficiency may not be optimal enough. The authors have developed an FLP method considering temporal efficiency, in which facility layout is optimized using genetic algorithm (GA), and have enhanced it so that robustness against changes in production environment can be taken into consideration. However, the enhanced method involves a large computational load, since numerous production scenarios need to be considered. This paper provides a method for reducing computational load in the robust FLP based on the sampling approach where each layout plan is evaluated with only a limited number of production scenarios in the optimization process by GA. Numerical experiments showed the potential of the proposed method to efficient robust FLP considering temporal efficiency.