作者: Murat Caner Testik , George C. Runger
DOI: 10.1080/07408170600692176
关键词: Statistical theory 、 Multivariate statistics 、 Variable (computer science) 、 Control chart 、 Computer science 、 Set (abstract data type) 、 Common cause and special cause 、 One sided 、 Quadratic programming 、 Mathematical optimization
摘要: Process knowledge can be exploited to improve the performance of control charts and it is not unusual know that a specific variable shifts above or below its mean under an assignable cause. In such case, one-sided chart common. The available statistical theory for tests used provide reasonable compromise numerical procedure design implement multivariate solutions. Although simulation in analysis, direct estimate through simulation. Instead, weights are estimated these easily set desired on-target average run length. Furthermore, interesting quadratic programming solution incorporated into analysis. Then results extended partial case where only some (not all) variables known shift one direction charts. A modern method blend algorithms practical solution. W...