Hi-tech manufacturing process selection problem for very low fraction of defectives with multiple characteristics

Hi-tech manufacturing process selection is the problem of comparing two processes and selecting the one that has a very low fraction of defectives (VLFD). In this paper, we consider process selection by C^T_pk, which is an extension of the widely-used popular index Cpk to cases with multiple characteristics. We propose an analytical approach is proposed to tackle the process selection problem. Hypothesis testing using two procedures to compare the two processes is developed. Critical values of the test are obtained to determine the selection decisions. Sample sizes required for designated selection power and confidence levels are analyzed. The results provide useful information to practitioners. We present an applied example that compares two light emitting diode (LED) production processes to illustrate the practicality of the proposed approach to a real problem in a factory.