Statistical process monitoring of between-part and within-part variations using independent component analysis

The data collected from in-process measurement usually contain useful information about the nature of the source of process variability. In this paper, each variation source is assumed to generate a different spatial variation pattern in the quality characteristics measurements. The variation source might also reveal interesting temporal pattern over the data sample. The spatial variation pattern and temporal pattern caused by a variation source may turn out to be the observed within- and between-part variations in the monitoring of product measurements. The study reported in this paper aimed at applying independent component analysis (ICA) to monitor within- and between-part variations. Various monitoring statistics obtained from ICA are used to construct the control procedure. The average run length (ARL) is used to measure the abnormalities detection performance. An extensive comparison based on simulation study indicates that the ICA-based control charts perform better than conventional control charts in terms of ARL. The paper contributes to the monitoring of within- and between-part variations.