Abstract

Process capability indices (PCI) have been widely used in the manufacturing industry for providing numerical measures on process precision, process accuracy and process performance in single characteristic and multiple characteristics, such as multi-characteristic process capability analysis (MCPCA). However, the parameters of the manufacturing processes are usually unknown and need to be estimated from the collected samples. Therefore, it is preferable to obtain an interval estimate, for which we can assert with a reasonable degree of certainty that it contains the true index value. This paper proposes a reliable approach to construct confidence intervals of the index values, and plot the corresponding joint confidence regions on multi-characteristics quality performance measurement diagram for monitoring and controlling the performance of all process characteristics simultaneously. Finally, an application example of a super twisted nematic liquid crystal display (STN-LCD) process is presented to demonstrate how the proposed approach can be applied to real applications.

Key words: Capability zones, joint confidence regions, multi-characteristic process capability analysis (MCPCA), process yield index.