An optimal design of multivariate control charts in the presence of multiple assignable causes

摘要

In quality surveillance, it is often necessary to measure the quality of a product on more than one characteristic. Also, processes subjected to only one assignable cause are rare in practice. Hence there is need for a good technique for designing a control chart that takes account of all the relevant characteristics and different assignable causes. We present an economic model for designing a matched single-cause model from the original multiple-cause model. The results of the final computations show that, under the model, the multivariate control chart is generally more efficient than the X̄-chart.