Çok Değişkenli Kalite Kontrolde Regresyon Düzeltmesi (Regression Adjustment in Multivariate Quality)

Abstract

One of the methods that is widespreadly used to monitor variability in a manufacturing process is construction of control charts for the variables in the stated process. If the number of variables that affect a process is more than one, then the number of constructed control charts and accordingly the cost will naturally increase. In the processes that have more than one independent variable and a dependent variable associated with these independent variables, constructing control charts related with the residuals that are obtained from the model by applying regression analysis instead of constructing control charts for each variable individually will provide significant profits in terms of time and cost. In this study, in a die-casting firm, a sample process that has seven independent variables and a dependent variable was examined, control charts were constructed in consequence of capacity analysis and the probabilities of type 1 and type 2 errors were computed.

Keywords

• Multivariate Quality Control
• Regression Adjustment
• Control Charts of Error
• Stepwise Regression
• Multivariate Process Capability.