Measurement systems analysis (MSA) is typically conducted on a single quality characteristic. Multiple characteristics processes are by now so common that studies concerning measurement systems analysis cannot be restricted to the univariate framework. This paper deals with the precision of multivariate measurement systems.We discuss three multivariate MSA-approval criteria proposed in literature. We examine the problems that affect these metrics emphasizing that these issues can lead to difficulties in determining whether measurement systems will perform adequately. To help in reducing these problems we propose and discuss a statistical test for assessing measurement systems. The results suggest that the proposed test is useful for evaluating multivariate measurement systems and helps to overcome some of the critical issues concerning the existing multivariate MSA-metrics.
ON THE ASSESSMENT OF MULTIVARIATE MEASUREMENT SYSTEMS / M. Scagliarini. - ELETTRONICO. - (2011), pp. 1-6. (Intervento presentato al convegno SCo 2011 7th Conference on Statistical Computation and Complex Systems tenutosi a Padova nel September 19-21, 2011).
ON THE ASSESSMENT OF MULTIVARIATE MEASUREMENT SYSTEMS
SCAGLIARINI, MICHELE
2011
Abstract
Measurement systems analysis (MSA) is typically conducted on a single quality characteristic. Multiple characteristics processes are by now so common that studies concerning measurement systems analysis cannot be restricted to the univariate framework. This paper deals with the precision of multivariate measurement systems.We discuss three multivariate MSA-approval criteria proposed in literature. We examine the problems that affect these metrics emphasizing that these issues can lead to difficulties in determining whether measurement systems will perform adequately. To help in reducing these problems we propose and discuss a statistical test for assessing measurement systems. The results suggest that the proposed test is useful for evaluating multivariate measurement systems and helps to overcome some of the critical issues concerning the existing multivariate MSA-metrics.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
