Comparing multiple process overall yields from multiple manufacturing lines. Issue 23 (2nd December 2017)

• Record Type: Journal Article
• Authors: Lin, Chen-Ju; Shen, Yi-Ling
• Is Part Of: Communications in statistics. Volume 46:Issue 23(2017)
• Page Start: 11764
• Page End: 11775
• Publication Date: 2017-12-02
• Subjects: Pre-amplifier -- Process selection -- Simultaneous confidence intervals.; Mathematical statistics -- Periodicals; Mathematics; Statistics; 519.2
• DOI: 10.1080/03610926.2017.1280169
• Languages: English
• ISSNs: 0361-0926

Abstract

ABSTRACT: A manufacturing process often uses multiple manufacturing lines to produce the same product for the productivity purpose. The capabilities of multiple manufacturing lines are often non-identical in practice. The task of process evaluation needs to measure the overall yield. This research compares the overall yield among I processes from K manufacturing lines, I ⩾ 2, K ⩾ 2. An unconstrained multiple comparisons with the best (UMCB) method based on the overall yield index is proposed to determine the best process with the highest overall yield. The simultaneous confidence intervals of each difference from the unknown best process in terms of the overall yield index are provided at a confidence level of at least 1 −α. To demonstrate the effectiveness of the UMCB method, the method was applied to choose the best process producing the pre-amplifier of microelectromechanical systems (MEMS) sensors in the production environment with multiple manufacturing lines.