A0646
Title: Robust statistical process monitoring of multivariate profiles
Authors: Christian Capezza - University of Naples Federico II (Italy) [presenting]
Fabio Centofanti - Universita di Napoli Federico II (Italy)
Antonio Lepore - Universita di Napoli Federico II (Italy)
Biagio Palumbo - University of Naples Federico II (Italy)
Abstract: In modern Industry 4.0 quality control applications, manufacturing processes generate large amounts of data, which often include outliers adversely affecting traditional control chart methods, especially in high-dimensional contexts. To face these challenges, the research introduces a novel framework, named robust multivariate functional control chart (RoMFCC), specifically designed to monitor multivariate functional quality characteristics while neutralizing the influence of both functional casewise and componentwise outliers. This innovation is important in multivariate profile monitoring, where outliers can affect an entire vector of functions or only a few components. The RoMFCC framework contains four main elements: a functional filter to detect functional component-wise outliers, a robust imputation of missing components in multivariate functional data, a robust dimension reduction that deals with functional case-wise outliers, and a procedure for prospective process monitoring. The RoMFCC's superior performance is assessed through a wide Monte Carlo simulation in comparison to competing monitoring schemes that have already appeared in the literature. The practical applicability of the RoMFCC is then demonstrated in monitoring a resistance spot welding process in automotive body-in-white manufacturing. The RoMFCC is implemented in the R package fun charts, which are available on CRAN.
