Study On Measurement System Capability Analysis Based On Simple Linear Profile Data

Measurement system capability analysis is important to quality control. Measurement system capability analysis is one of the modern quality management tools and the frequency of its using is becoming more and more highly. The ultimate purpose of measurement system capability analysis is to determine whether the capability of the system is enough or not, and to determine whether the measurement system is acceptable or not.Whether in the practice of production or in scientific research, measurement system analysis have been widely used, it is an essential step of the research about the design of experiment and the process capability. It is also an important tool to improve the quality of the product or service in enterprise. Much of the literature for measurement system capability analysis focuses on univariate measurement systems. However, in modern industry, the quality of some products require multiple properties to reflect the quality, the measurement values is related to a set of multidimensional data, analysis method of univariate in this environment measurement systems is generally not applicable, so more and more enterprises begin to pay attention to multivariate measurement system capability analysi, scientific research in this field also increased gradually.However, the existing methods are not suitable when the quality of a product is characterized by a profile. In this paper, an approach is proposed when the measured results of objects can be characterized by simple linear profiles. Since the measurement error affects the estimation of the slope and the intercept, the measurement system capability analysis for simple linear profiles can be regarded as a two-response problem. MANOVA is applied to estimate the variance-component matrices of measurement error and other sources of variation. Then two criteria which are the development of the multivariate criteria precision to tolerance ratio and percent R\&R are introduced to assess the simple linear profile measurement system capability. In this paper, the proposed method is applied to a case which demonstrates how to implement it.