We examine how to reduce the adverse effects of machine down time by prioritizing production equipment for attention by a limited number of maintenance personnel. The maintenance scheduling policies use system information to determine which machines are the most critical at a given time. This information includes repair times, queue lengths, due dates, available capacity, and steady-state utilizations. The effects of different staffing levels are also considered. A discrete-event simulation model of a wafer fabrication facility is used to examine the performance of the different policies. Results indicate that the choice of maintenance scheduling policy can significantly affect system performance under restrictive staffing levels. Moreover, factory performance is shown to be sensitive to small changes in staffing level.