Multivariate Statistical Process Control of Electrostatic Separation Processes

Multivariate control charts are commonly used for monitoring processes the quality of which is determined by two or more correlated output variables. The aim of this paper is to point out the effectiveness of multicriterion control charts for supervising the variability of the outcome of an electrostatic separation process. The experiments were carried out on samples of chopped electric cable wastes, similar to those currently processed by the recycling industry. The two output variables considered in the study were the masses of product recovered in the middling and conductive compartments of the collector. When the separation process was under control, the two variables were correlated, and T2-type control charts could be established. Two out-of-control situations were simulated. The multivariable control charts monitoring the process location and spread were able to detect these situations, though each output variable taken independently remained within the control limits.