Level: Intermediate
【Presenter】
Jeremy Ash, JMP Analytics Software Tester, SAS Institute Inc.
[Announcement summary]
The Model-Based Multivariate Control Chart (MDMVCC) platform creates control charts based on Principal Component Analysis (PCA models) and PLS models. This control chart can be used for failure detection and diagnosis with multidimensional data. Here, we will demonstrate monitoring by MDMVCC based on the PLS model using the Tennessee Eastman process (simulated chemical industry process). In this simulation, quality and process variables are measured as the chemical reactor produces a liquid product from a gaseous reactant.
First, perform a failure diagnosis in the offline state. In this case, you will alternate between multivariate control charts, univariate control charts, and process diagnostic reports, which is very easy to do on the MDMVCC platform. Next, we will connect the JMP to an external database and demonstrate online monitoring with the MDMVCC platform. Since product quality variables have a time delay until all measurement results are available, failure detection is also generally delayed. In MDMVCC based on the PLS model, quality variable variability is monitored as a function of process variables, but process variables are generally available relatively quickly, which is useful for early failure detection.
[Presenter Profile]
Jeremy Ash holds a degree in bioinformatics from North Carolina State University and is currently working as an analytics software tester at JMP. In my dissertation, I wrote about computational methods in chemoinformatics, chemometrics, and bioinformatics. He holds a Master's degree in Statistics from North Carolina State University and a Bachelor of Science degree in Biology from the University of Texas at Austin.
