Univariate and multivariate approaches for evaluating the capability of dynamic-behavior processes (case study)

The majority of classic SPC methodologies assume a steady-state (static) process behavior (i.e., the process mean and variance are constant) without the influence of the dynamic behavior (i.e., an intended or unintended drift in the process mean or variance). Traditional SPC methods have been successfully used in steady-state manufacturing processes, but these approaches are not valid for use in dynamic behavior environments. The standard assumptions in SPC are that the observed process characteristics are normally, independently and identically distributed (IID) with fixed mean μμ and standard deviation σσ when the process is in control. Due to the dynamic behavior, these assumptions are not always valid. This study provides a scientific approach for evaluating the capability of cold rolling processes (as an example of manufacturing processes that undergo many disturbances and dynamic behavior) so that quality improvement may be attained because of the good understanding of the nature of the processes. The paper proposes the appropriate procedures for evaluating the capability of the manufacturing processes, especially for those in a dynamic behavior mode, with a comparison between the univariate and multivariate capability indices.