GRAFIK PENGENDALI MULTIVARIATE EXPONENTIALLY WEIGHTED MOVING COVARIANCE MATRIX (MEWMC) DENGAN DIAGNOSIS PERGESERAN MENGGUNAKAN REGRESSION-ADJUSTED VARIABLES; MULTIVARIATE EXPONENTIALLY WEIGHTED MOVING COVARIANCE MATRIX (MEWMC) CONTROL CHART WITH SHIFT DIAGNOSIS USING REGRESSION-ADJUSTED VARIABLES

Iudina, Manda W, Herni Utami

2015 | Skripsi | FMIPA

Multivariate Exponentially Weighted Moving Average (EWMA) is one of the multivariate control charts are commonly used to detect small shifts that occurred in the mean vector. The shift in the process can occur not only on the mean but also on the variability of the process. Therefore needed a method that can be used to detect the variability shift. Multivariate exponentially Weighted Moving Covariance Matrix (MEWMC) is a multivariate control chart which is a modification of the MEWMA control chart but addressed to detect small shifts in the covariance matrix. MEWMC control chart analysis uses multistandardized data and MEWMC assumes that knowing mean vector and covariance matrix from the predecessor in control process. An important aspect that should not be missed from a multivariate control chart analysis is the diagnosis after the signal out of control. Regression-adjusted variables is used to diagnose the cause of the out of control process on Multivariate EWMC control chart. MEWMC control chart can detect small shifts in the covariance matrix for individual observations, while MEWMA chart indicates that the process under controlled conditions. By using regression-adjusted variables, the source causes a shift in the covariance matrix can be known certainty.

Kata Kunci : Multivariate control chart; Multivariate EWMC control chart; Regression-adjusted variables