Asymmetric monitoring of multivariate data with nonlinear dynamics

We consider asymmetric change detection by generalizing the one-sided MEWMA control chart. In particular, we revise and extend the one-sided MEWMA algorithm to cope with mixed alternatives where some coordinates are allowed to increase and others may change in any direction. Moreover, in order to cope with nonlinear serially correlated data, we introduce one-sided MEWMA control charts based on semi-parametric stationary block bootstrap. The motivating application is related to massive brake-disk production for the automotive industry. The example considers seven geometrical and dimensional parameters and shows how the proposed method discriminates between geometrical deformation and dimensional shifts.