OCTOBER 2022 ISSUE OF THE STATISTICS DIGEST NOW AVAILABLE!
This issue includes the following:
· Mini Paper by Lynne B. Hare - Designing for Product and Process Robustness
· Feature by Rucha Deshpande - SIMCA: Soft Independent Modeling of Class Analogies
· Ellis R. Ott Scholarship Awardees for the Academic Year, 2022-23
· Memories of Gordon Clark (1934 – 2022) by Doug Hlavacek
· Columns by Donald J. Wheeler, guest author Roberto Salazar Reyna, and Jim Frost.
Mini Paper: Designing for Product and Process Robustness:
Robust Design, Taguchi Methods, Orthogonal Arrays
From the researcher’s perspective, the key question is how attain product robustness. Is it serendipity? Does it just happen? For some products, like Oreos and teabags, robustness may have been dumb luck falling into human hands years ago. But for more recent products and processes like automobiles and electronics, systematic approaches for building in robustness have been developed and improved upon.
Feature: SIMCA: Soft Independent Modeling of Class Analogies
Soft Independent Modeling of Class Analogies (SIMCA), Principal Component Analysis (PCA)
Soft Independent Modelling of Class Analogies (SIMCA) is a supervised classification technique for data modelling. As the name suggests, the technique works based on similarities and analogies between samples in the data set. The SIMCA technique is based on Principal Component Analysis (PCA) which is a preferred method of analysis when the data is highly dimensional and presents a multicollinearity problem. The dimensionality and multicollinearity challenges are often seen with in-process data obtained with process monitoring tools such as Texture analyzer, Near Infrared Spectroscope, Raman Spectroscope, or Infrared Spectroscope.