Variable selection methods for process monitoring have focused mainly on the explained variance performance criteria. However, explained variance efficiency is a minimal notion of optimality and does not necessarily result in an economically desirable selected subset, as it makes no statement about the measurement cost or other engineering criteria. For many applications, it may be useful for external information to influence the selection process. For example, some variables may be easier and cheaper to carry out then others or they might be very important according to some engineering criteria. Neglecting this information in statistical process control, would be counterproductive. In this article, we propose a statistical methodology to select a reduced number of relevant variables for multivariate statistical process control that makes use of engineering and variability evaluation criteria. A double reduction of dimensionality is applied in conjunction with economic and variability selection criteria. The subset of relevant variables is selected in a manner that retains, to some extent, the structure and information carried by the full set of original variables. A real application from automotive industry will be used to illustrate the method.