Introduction to Chemometrics

1
RECENT TRENDS IN QUALITY ASSURANCE
TECHNIQUESCHEMOMETRICS

• MS.SHEVANTE T.B
• M.PHARM(QAT)
• V.I.P.E.R
• GUIDED BY-DR.MR.GADHAVE M.V

2

• CONTENTS
• INTRODUCTION
• VARIOUS TECHNIQUES
• AREA OF APPLICATION
• SPECTROSCOPISTS REQUIREMENTS FOR CHEMOMETRICS
• PROCESS ANALYSIS
• MULTIVARIATE DATA ANALYSIS
• ADVANTAGES OF MULTIVARIANT DATA ANALYSIS
• PROCESS ANALYTICAL TECHNOLOGY (PAT) AND QUALITY
  BY DESIGN (QBD)
• USES OF MULTIVARIATE ANALYSIS METHODS
• CONCLUSION
• REFERENCES

3

• INTRODUCTION
• Introduced by Svante Wold 4 and Bruce R.
  Kowalski in the early 1970s
• A reasonable definition of chemometrics remains
  as how do we get chemical relevant information
  out of measured chemical data, how do we
  represent and display this information, and how
  do we get such information into data? as
  mentioned by Wold .
• Chemometricians have applied the well-known
  approaches of multivariate calibration, chemical
  resolution, and pattern recognition for
  analytical studies.
• Chemometrics is the use of mathematical and
  statistical methods to improve the understanding
  of chemical information and to correlate quality
  parameters or physical properties to analytical
  instrument data.

4

• WHITE, BLACK, AND GRAY SYSTEMS
• Mixture samples commonly encountered in
  analytical chemistry fall into three categories,
  which are known collectively as the
  white--grayblack multi-component system.

SYSTEM DESCRIPTION
White All spectra of the component chemical species in the sample, as well as the impurities are available.
Black no a priori information regarding the chemical composition
Grey No complete knowledge is available for the chemical composition or spectral information
5

• Various techniques
• Partial Least Squares (PLS) ,
• Soft Independent Modeling of Class Analogy
  (SIMCA)
• Methods Based on Factor Analysis,
• Principal-component Regression (PCR)
• Target Factor Analysis (TFA)
• Evolving Factor Analysis (EFA)
• Rank Annihilation Factor Analysis (RAFA)
• Window Factor Analysis (WFA)
• Heuristic Evolving Latent Projection (HELP)
• Artificial Neural Network.
• Multiplicative scatter Correction.

6

• PARTIAL LEAST SQUARES REGRESSION (PLS REGRESSION)
• It is a statistical method that bears some
  relation to principal component regression
  instead of finding hyperplanes of maximum
  variance between the response and independent
  variables, it finds a linear regression model by
  projecting the predicted variables and the
  observable variables to a new space.
• SOFT INDEPENDENT MODELLING BY CLASS ANALOGY
  (SIMCA)
• It is a statistical method for supervised
  classification of data.
• PRINCIPAL COMPONENT REGRESSION (PCR)
• It is a regression analysis technique that is
  based on Principal component analysis(PCA).
• Principal component analysis (PCA) is a
  statistical procedure that uses an orthogonal
  transformation to convert a set of observations
  of possibly correlated variables into a set of
  values of linerly uncorrelated variables called
  principal components.
• FACTOR ANALYSIS
• It is a statistical method used to describe
  variability among observed, correlated variables
  in terms of a potentially lower number of
  unobserved variables called factors.

7

• RANK ANNIHILATION FACTOR ANALYSIS (RAFA)
• It is used to analyze difference spectra of
  kinetic-spectrophotometric data. Annihilation of
  the contribution of one chemical component from
  the original data matrix is a general method in
  RAFA
• WINDOW FACTOR ANALYSIS (WFA)
• It is a self-modeling method for extracting the
  concentration profiles of individual components
  from evolutionary processes such as flow
  injection, chromatography, titrations and
  reaction kinetics.
• HEURISTIC EVOLVING LATENT PROJECTION(HELP)
• It is new method to resolve two way bilinear
  multi component data into spectra chromatograms
  of pure constituents.
• EVOLVING FACTOR ANALYSIS (EFA)
• It is a recently developed method for a
  completely model-free resolution of overlapping
  peaks into concentration profiles and absorption
  spectra

8
Mathematics
Mathematics
Organic
Organic
Chemistry
Chemistry
Statistics
Statistics
Biology
Biology
Analytical
Analytical
Computing
Industrial
Applications
Computing
Industrial
Applications
CHEMOMETRICS
CHEMOMETRICS
Chemistry
Chemistry
among others
among others
p
Theoretical
Theoretical
Pharmaceuticals
Engineering
Engineering
and Physical
and Physical
Chemistry
Chemistry
9

• AREA OF APPLICATION
• Spectroscopy analyzing spectroscopic data
• Demonstrates the basic principles underlying the
  use of common experimental, chemometric, and
  statistical tools.
• Emphasis has been given to problem-solving
  applications and the proper use and
  interpretation of data used for scientific
  research.
• Useful for analysts in their daily problem
  solving, as well as detailed insights into
  subjects often considered difficult to thoroughly
  grasp by non-specialists.
• Provides mathematical proofs and derivations for
  the student or rigorously-minded specialist.
• Multivariate analysis.

10

• Chemometric analysis for spectroscopy
  mathematical and statistical methods to improve
  the understanding of chemical information and to
  correlate quality parameters or physical
  properties to analytical instrument data
• Patterns in the data are modeled these models
  can then be routinely applied to future data in
  order to predict the same quality parameters.
• The result of the chemometrics approach is
  gaining efficiency in assessing product quality.
  It can lead to more efficient laboratory
  practices or automated quality control systems.
  The only requirements are an appropriate
  instrument and software to interpret the patterns
  in the data.
• Efficient ways to solve the calibration
• Problem for analysis of spectral data
• Spectroscopists use software packages for data
  analysis, modeling,classification and prediction
• An essential part in the modern chemical and
  biomedical industries

11

• The science of chemometrics gives spectroscopists
  many efficient ways to solve the calibration
  problem for analysis of spectral data.
• Chemometrics can be used to enhance methods
  development and make routine use of statistical
  models for data analysis.
• Spectroscopists use software packages like The
  Unscrambler for spectroscopic data analysis,
  modeling, classification and prediction to meet
  process monitoring and quality assurance needs.
• SPECTROSCOPISTS REQUIREMENTS FOR CHEMOMETRICS
• Proper application of spectroscopic data
  pre-processing, to reduce and correct
  interferences such as overlapped bands,baseline
  drifts, scattering, and pathlength variation.

12

• Figure shows NIR spectra of wheat samples that
  scatter effect is severe due to packing and size
  variation.
• Multiplicative Scatter Correction (MSC)
  pretreatment is recommended to build reliable
  relationship between wheat protein content and
  spectral data, for scatter correction.

NIR spectra 1000-2500 nm of wheat samples.
13

• Strong calibration and diagnostics means of
  sample selection and variable selection,
  statistic result calculation to build
  representative and reliable models.
• Model validation and integration means to supply
  rigorous prediction,
• measurement QC and real-time product quality and
  process monitoring.
• Spectroscopists need to use the following methods
  within a chemometrics software package to
  explore their data
• Principal Component Analysis (PCA)
• Regression (PLS, PCR, MLR, 3-way PLS) and
  Prediction
• SIMCA and PLS-DA Classification
• Design of Experiments
• ANOVA and Response Surface Methodology
• Multivariate Curve Resolution (MCR)
• Clustering (K-Means)

14

• PROCESS ANALYSIS
• Understanding and managing processes is vital in
  any manufacturing or company. In many cases,
  those processes can be complex, with a large
  number of variables potentially affecting the
  outcome.
• Most manufacturers rely on traditional
  Statistical Process Control (SPC) software which
  uses univariate statistical analysis such as
  mean, median, standard deviation etc.
• Unfortunately, univariate statistics often miss
  the underlying patterns in process data. This is
  where advanced Multivariate Statistical Process
  Control (MSPC) can give manufacturers and
  engineers an edge.

15

• MULTIVARIATE DATA ANALYSIS
• Multivariate data analysis is an advanced
  statistical approach which identifies all of
  the critical variables and underlying patterns in
  a data set.
• Importantly, it also shows the relationships
  between variables and how they impact on each
  other essential when trying to understand
  complex process behavior.
• Multivariate Statistical Process Control (MSPC)
  applies these powerful methods to process and
  manufacturing data, giving you a better
  understanding and control over your processes.

16

• ADVANTAGES OF MULTIVARIATE STATISTICAL PROCESS
  CONTROL OVER TRADITIONAL SPC
• Less control charts less margin for error-MSPC
  simplifies the job of process operators by
  showing all process variables, including
  relationships which cannot be detected with
  univariate statistics, on just one or two control
  charts. This removes the need for control charts
  for every individual variable.
• See the full picture and find underlying patterns
  in data-Multivariate analysis cuts through data
  to show which process parameters are interacting
  with each other and crucially, which are related
  to defects. MSPC also helps you identify which
  defects are related and which are diametrically
  opposed to each other.
• Make process improvements based on deeper
  understanding of process
• behavior-Many manufacturers try new strategies
  to resolve or improve processes, only to find the
  adjustments create unforeseen problems elsewhere.
  MSPC helps you understand the interaction between
  variables, allowing you to model and predict the
  effect of a new strategy before implementing it.

17

• BENEFITS OF MULTIVARIATE PROCESS MONITORING AND
  CONTROL FOR INDUSTRY
• Reduce problems during scale-up from RD to
  production-
• Accelerate pilot plant activities and optimize
  testing during the development. Problems with raw
  materials or design flaws can be spotted faster
  and easier.
• Prevent process failures-Reduce production
  problems and failures by identifying drift in a
  process before they become Critical. Enables
  process operators, managers to make informed
  decisions in real-time on the plant floor.
• Improve product quality-Powerful diagnostics let
  you identify precisely where faults occur so you
  can improve product quality, thereby reducing
  warranty claims and even the risk of recall.
  On-line quality control helps minimize the need
  for expensive and time consuming off-line
  testing.
• Reduce process costs-Less failures and higher
  quality invariably leads to lower costs through
  reduced waste and scrap, energy, rework, overtime
  etc. Even a small improvement in process
  efficiency can lead to a noticeable improvement
  on the bottom line.

18

• Optimize processes-
• A better understanding of process behavior
  enables on-going process optimization programs,
  improved productivity and increased operational
  efficiency. MSPC gives you deeper insights and
  visibility so you can reduce variations and
  forecast more accurately.
• Reduce time to market-
• Enhance process and quality improvement programs
  on individual production lines and across sites.
  Knowledge transfer and more efficient processes
  can lead to significant operational improvements,
  so you get your products to market faster and at
  lower cost.
• Increase overall equipment efficiency (OEE)-
• Pinpoint anomalies in equipment performance to
  help increase uptime, reduce unscheduled
  maintenance, implement corrective and
  preventative action (CAPA) programs and get the
  maximum return on investment in plant and
  machinery.

19

• PROCESS ANALYTICAL TECHNOLOGY (PAT) AND QUALITY
  BY DESIGN (QBD)
• PAT involves fundamental changes to working
  practices. It is based on the application of
  sophisticated and high technology process
  analyzers and the development of online
  Multi-Variate Analysis (MVA).
• Analysis of the process data is a key to
  understand the process and keep it under
  multivariate statistical control.
• The philosophy behind the Process Analytical
  Technology (PAT) initiative represents a move
  away from traditional product centric
  measurements of quality to a process centric
  focus on quality.
• This process centric approach to quality may
  start at the early design stages of the process
  (Quality by Design, QbD) so as to have the
  quality built in.
• It continues throughout the entire life-cycle of
  a product.
• The measurement of Critical Quality Attributes
  (CQA's) all along the entire process will ensure
  consistent product quality.
• These quality attributes define the Design Space
  of the process.

20
ICH Q10 approach of PAT
21
QbD- overarching paradiag
22

• VARIOUS CHEMOMETRIC METHODS ARE APPLIED FOR THE
  ANALYSIS OF COMPLEX DATA IN A NUMBER
  OF APPLICATIONS VIZ.
• Spectroscopic calibrations
• Process modeling for optimization
• Process models for monitoring and fault detection
  
• Dynamic model identification for process control
• Multivariate statistical process control
• Process analytical instrument standardization
• Analytical instrument design and development

23

• USES OF MULTIVARIATE ANALYSIS METHODS
• CONSUMER AND MARKET RESEARCH
• -Quality control and quality assurance across a
  range of industries such as food and beverage,
  paint, pharmaceuticals, chemicals, energy,
  telecommunications, etc process optimization and
  process control research and development.
• EXPLORATORY ANALYSIS
• -To explore possible outliers and indicate
  whether there are patterns or trends in the data.
• -PCA is an important part of chemometrics and
  provides the most compact representation of all
  the variation in a data table.
• -Exploratory algorithms such as principal
  component analysis (PCA) are designed to reduce
  large complex data sets into a series of
  optimized and interpretable size.

24

• REGRESSION
• -To predict related properties that are easier to
  measure.
• -The goal of chemometric regression analysis is
  to develop a model which correlates the
  information in the set of known measurements to
  the desired property.
• -Chemometric algorithms for performing regression
  include partial least squares (PLS) and principal
  component regression (PCR).
• -Chemometric regression is extensively used in
  making decisions relating to product quality in
  the on-line monitoring and process control
  industry where fast and expensive systems are
  needed to test.

25

• CONCLUSION
• Chemometrics is the bridge between connecting the
  state of a
• chemical system to the measurements of the
  system.
• It has become an essential part in the modern
  chemical and biomedical industries.
• Chemometrics software has been widely used by
  product development scientists, process
  engineers, PAT specialists, and QA/QC scientists
  to build reliable model, ensure product quality,
  classify raw material, and to monitor process end
  point in real-time.

26

• REFERENCES
• http//www.camo.com/rt/Resources/chemometrics.html
  
• CHEMICAL ANALYSIS-A SERIES OF MONOGRAPHS ON
  ANALYTICAL CHEMISTRY AND ITS APPLICATIONS, Edited
  by-J. D. WINEFORDNER,VOLUME 164.
• INTRODUCTION TO CHEMOMETRICS-RICHARD G BRERETON,
  CENTRE FOR CHEMOMETRICS