High-Dimensional Process Monitoring and Fault Isolation via Variable Selection

TLDR

Process monitoring and fault isolation are critical yet challenging tasks in high‑dimensional quality control. The study proposes a variable‑selection‑based multivariate SPC method for monitoring and diagnosing faults in high‑dimensional processes. The method uses a forward‑selection algorithm to identify out‑of‑control variables and then applies a multivariate control chart to monitor them. The approach enables simultaneous fault detection and isolation, and simulations plus a real case confirm its effectiveness. Authors: Kaibo Wang ([e-mail]) and Wei Jiang ([e-mail]).

Abstract

AbstractBoth process monitoring and fault isolation are important and challenging tasks for quality control and improvement in high-dimensional processes. Under a practical assumption that not all variables would shift simultaneously, this paper proposes a variable-selection-based multivariate statistical process control (SPC) procedure for process monitoring and fault diagnosis. A forward-selection algorithm is first utilized to screen out potential out-of-control variables; a multivariate control chart is then set up to monitor suspicious variables. Therefore, detection of faulty conditions and isolation of faulty variables can be achieved in one step. Both simulation studies and a real example have shown the effectiveness of the proposed procedure.KeywordsForward SelectionLinear RegressionMultivariate Statistical Process ControlT2 Chart Additional informationNotes on contributorsKaibo WangDr. Wang is an Assistant Professor in the Department of Industrial Engineering. His email address is [e-mail] JiangJiang is a Visiting Associate Professor in the Department of Industrial Engineering & Logistics Management. His email address is [e-mail].