Bayesian Sequential Control Charts for Monitoring Multivariate Processes

The control charts are an effective tool to enhance quality in industrial sectors. To make full use of the sample' information in different stages and consider the parameter uncertainty risk in statistical process control, this paper introduces a reference prior distribution for the parameters in quality models, and constructs control with the warning limits and control limits in terms of the quality variables' predictive distributions as well as the relationship between the multivariate student t distribution and F distribution, monitoring the variables change in processes. When the current stage is under statistical control, the parametric posterior distribution is considered to be their priori distribution in the next stage, by which an sequential Bayesian multivariate control approach is established.