A density-based statistical process control scheme for high-dimensional and mixed-type observations

Statistical Process Control (SPC) techniques are useful tools for detecting changes in process variables. The structure of process variables has become increasingly complex as a result of increasingly complex technologies. The number of variables is usually large and categorical variables may appear alongside continuous variables. Such observations are considered to be high-dimensional and mixed-type observations. Conventional SPC techniques may lose their accuracy and efficiency in detecting changes in a process with high-dimensional and mixed-type observations. This article presents a density-based SPC approach, which is derived from a Local Outlier Factor (LOF) scheme, as a solution to this problem. The parameters in an LOF scheme are investigated and a procedure to design a corresponding control chart is presented. The good performance of the proposed control scheme is demonstrated via numerical simulation.