A ONE-SIDED SINGLE SCREENING-PROCEDURE BASED ON INDIVIDUAL-UNIT MISCLASSIFICATION ERROR

A procedure is proposed for screening applications in quality control that eliminates overselection (namely, the condition where the proport ion of accepted screened items is greater than the proportion of confo rming items in the original population) and controls for individual un it misclassification error (IME), namely, the probability of an accept ed/selected item being below specification. Screening involves selecti ng items whose performance is within specifications based on observing one or more correlated screening variables in lieu of observing the p erformance variable directly. Overselection and an uncontrollably high IME can occur when (1) there is a high correlation between the perfor mance and screening variables and (2) the proportion of conforming ite ms among accepted items (or its complement, average outgoing quality ( AOQ)), is prespecified as in, for example, Owen et al. (1975). We inst ead propose a procedure that prespecifies a maximum tolerable IME (ema xa) for accepted items, which eliminates overselection and assures tha t the IME of each outgoing item as well as the AOQ are both equal to o r less than emaxa. The quality of each individual item as well as AOQ is thereby explicitly controlled. Only the use of standard normal tabl es is required to implement our proposed procedure.