Danica Heller-Krippendorf develops concepts and approaches optimizing the applicability of MVA on data sets from an industrial context. They enable more time-efficient MVA of the respective ToF SIMS data. Priority is given to two main aspects by the author: First, the focus is on strategies for a more time-efficient collection of the input data. This includes the optimal selection of the number of replicate measurements, the selection of input data and guidelines for the selection appropriate data preprocessing methods. Second, strategies for more efficient analysis of MVA results are presented. Contents

• Advantages of Correlation Loadings for MVA of ToF-SIMS Spectra

• Required Number of Replicate Measurements in an Industrial Context

• Selection of Suitable Data Pre-processing in Root Cause Analysis Including the Selection of an Efficient Peak List, Scaling, Normalization and Centering

• New Presentation of PCA Results in Order to Simplify Data AnalysisTarget Groups

• Scientists and students in the field of surface analysis, data evaluation, ToF-SIMS and methods of multivariate data analysis

• Practitioners, especially in industry, in the field of surface and error analysisAbout the AuthorDanica Heller-Krippendorf did her research and dissertation at the University of Siegen, Germany, in collaboration with a German analytical service company. Now she is engineer in analytics at a DAX company.

Autorentext
Danica Heller****-Krippendorf did her research and dissertation at the University of Siegen, Germany, in collaboration with a German analytical service company. Now she is engineer in analytics at a DAX company.

Inhalt
Advantages of Correlation Loadings for MVA of ToF-SIMS Spectra.- Required Number of Replicate Measurements in an Industrial Context.- Selection of Suitable Data Pre-processing in Root Cause Analysis Including the Selection of an Efficient Peak List, Scaling, Normalization and Centering.- New Presentation of PCA Results in Order to Simplify Data Analysis.