Accurate test limits using the bootstrap

During inspection, mass-produced items are compared to given specification limits. The presence of measurement errors forces to set test limits which are slightly more strict. It is important to make this choice accurately: test limits which are too strict cause unnecessary loss of yield, whereas choices which are too liberal lead to consumer losses exceeding prescribed bounds. From earlier work, such accurate test limits are available. Using rather elaborate corrections, these limits have also been adapted to the often more realistic case where (some of) the underlying parameters need to be estimated. In this paper it is investigated how well such corrections can be achieved using the bootstrap method instead.