B1590
Title: A bivariate semiparametric chart based on the number of observations lying between the control limits
Authors: Markos Koutras - University of Piraeus (Greece)
Elisavet Sofikitou - University of Piraeus - University of Piraeus Research Center (Greece) [presenting]
Abstract: A new bivariate, Shewhart-type semiparametric control chart is intorduced which takes into account not only the location of a single pair of order statistics, but also the number of specific observations of the test sample that lie between appropriately chosen control limits of the historical sample. The major advantage of this chart is the fact that the $FAR$ and the $ARL_{in}$ values are not affected by the choice of the marginal distributions; although these quantities are typically affected by the dependence structure of the monitored characteristics, they are practically almost the same when different copulas are used. Expressions for the operating characteristic function of the new control chart and the FAR are obtained. In addition, an exact formula is provided for the alarm rate when the bivariate observations follow a specific bivariate distribution. Finally, a detailed numerical study is carried out to assess the performance of the new control chart. The work has been partly supported by the University of Piraeus Research Center.
Journal