In this paper, we discuss the problem of quality control with an unrel
iable machine which produces defects at a rate of LAMBDA0 per unit whe
n in-control and a rate of LAMBDA1 when out-of-control (where LAMBDA1
> LAMBA0). Every h time periods, we sample n units, count the number o
f defects, and (using a process based on a Shewart c-chart) test the h
ypothesis that the machine is in control by comparing the total number
of defects to an upper control limit (UCL). More important, we introd
uce the concept that a buffer inventory which immediately follows the
unreliable machine may reduce expected total costs. This buffer serves
to delay the movement of items from the unreliable machine to the nex
t stage of the production process. In this way, we can isolate and rep
air most defective items before they are embedded in a product downstr
eam or sold to customers where repair is more costly. To search for th
e optimal control policy, we find bounds for n, h, and UCL; given valu
es for these variables, we show how the optimal buffer size can be det
ermined directly. Numerical results illustrate the magnitude of potent
ial savings.