Authors: Azadeh Adibi; Connie M. Borror; Douglas C. Montgomery
Addresses: School of Computing, Informatics, and Decision Systems Engineering, Arizona State University, 699 S Mill Ave., Tempe, AZ 85281, USA ' School of Mathematical and Natural Sciences, Arizona State University at the West Campus, P.O. Box 37100 Phoenix, AZ 85069, USA ' School of Computing, Informatics, and Decision Systems Engineering, Arizona State University, 699 S Mill Ave, Tempe, AZ 85281, USA
Abstract: In this study, a p-value-based method for monitoring polynomial and nonlinear profiles in Phase II process monitoring is proposed. Performance of the proposed method is evaluated using the average run length criterion under different shifts in the model parameters. In this approach, the p-values are calculated for all subgroups within a sample. If any p-value is less than a prespecified threshold, the chart signals out of control. The main advantage of the proposed method is its ease of implementation in practice. Moreover, in this method, only one control chart is needed for routine monitoring of the model parameters. Only if an out-of-control signal is observed, then individual monitoring of the regression model parameters is needed. Performance of the proposed approach is compared to the T² method. Results of a simulation study on the proposed p-value approach are provided.
Keywords: profile monitoring; Phase II process monitoring; average run length; ARL; polynomial models; nonlinear modelling; p-value; control charts; SPC; statistical process control; simulation.
International Journal of Quality Engineering and Technology, 2015 Vol.5 No.2, pp.101 - 113
Available online: 09 Sep 2015 *Full-text access for editors Access for subscribers Purchase this article Comment on this article