Title: Variable selection with pre-assigned roles and cost utility analysis for process monitoring
Authors: Luan Jaupi - CNAM (France) [presenting]
Abstract: Methods for selecting a reduced number of relevant variables for process monitoring are proposed. We assume that a two-class system is used to classify the variables as primary and secondary based on different criteria. Then a double reduction of dimensionality is applied to select relevant primary variables that represent well the whole set of variables. The selection methodology uses external information and a cost-utility analysis to influence the selection process and compare one use of resources with other possible uses. 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. The advantages of the methods are illustrated with simulation results and a real application from automotive industry.