Haojie Ren, Nan Chen, Zhaojun Wang
Process monitoring and fault diagnosis using profile data remains an important problem in statistical process control (SPC). It is particularly challenging with multichannel profiles. This article proposes a novel scheme for Phase-II monitoring of multichannel profiles. The proposed method integrates the exponentially weighted moving average (EWMA) scheme with multichannel functional principal component analysis (MFPCA) to obtain the charting statistics. Their control limits can be effectively computed by the Markov chain method. Moreover, we provide a natural diagnostic procedure to locate the possible change point once the chart is out-of-control. Our proposed method is demonstrated to be effective and efficient through simulation results and an industrial case study from a multioperation forging process.
© 2001-2024 Fundación Dialnet · Todos los derechos reservados