This thesis is concerned with the design of variable sampling interval multivariate control charts with fixed times in which samples are taken at specified equally spaced fixed time points, and additional samples are taken between these fixed times if preceding sample statistics falls close to the control limit.
The proposed control charts investigate all quality characteristics at fixed sampling times and specified number of quality characteristics selected by regression-adjusted variables at additional sampling times. A Markov chain approach is used to derive the formulas for evaluating the proposed charts. The performances of the proposed charts are studied under various correlation structures and process mean shifts, and they are compared with those of the conventional fixed sampling interval(FSI) charts and variable sampling interval(VSI) charts. Guidelines are given for selecting design parameters for practical applications.
