ASTM D6122-99
(Practice)Standard Practice for Validation of Multivariate Process Infrared Spectrophotometers
Standard Practice for Validation of Multivariate Process Infrared Spectrophotometers
SCOPE
1.1 This practice covers requirements for the validation of measurements made by on-line, process near- or mid-infrared analyzers, or both, used in the calculation of physical, chemical, or quality parameters of liquid petroleum products. The parameters are calculated from spectroscopic data using multivariate modeling methods. The requirements include verification of adequate instrument performance, verification of the applicability of the calibration model to the spectrum of the sample under test, and verification of equivalence between the result calculated from the infrared measurements and the result produced by the primary method used for the development of the calibration model.
1.2 This practice does not cover procedures for establishing the calibration model used by the analyzer. Calibration procedures are covered in Practices E 1655 and references therein.
1.3 This practice is intended as a review for experienced persons. For novices, this practice will serve as an overview of techniques used to verify instrument performance, to verify model applicability to the spectrum of the sample under test, and to verify equivalence between the parameters calculated from the infrared measurement and the results of the primary method measurement.
1.4 This practice teaches and recommends appropriate statistical tools, outlier detection methods, for determining whether the spectrum of the sample under test is a member of the population of spectra used for the analyzer calibration. The statistical tools are used to determine if the infrared measurement results in a valid property or parameter estimate.
1.5 The outlier detection methods do not define criteria to determine whether the sample, or the instrument is the cause of an outlier measurement. Thus, the operator who is measuring samples on a routine basis will find criteria to determine that a spectral measurement lies outside the calibration, but will not have specific information on the cause of the outlier. This practice does suggest methods by which instrument performance tests can be used to indicate if the outlier methods are responding to changes in the instrument response.
1.6 This practice is not intended as a quantitative performance standard for the comparison of analyzers of different design.
1.7 Although this practice deals primarily with validation of on-line, process infrared analyzers, the procedures and statistical tests described herein are also applicable to at-line and laboratory infrared analyzers which employ multivariate models.
1.8 This standard does not purport to address all of the safety concerns, if any associated with its use. It is the responsibility of the user of this standard to consult and establish appropriate safety and health practices and determine the applicability of regulatory limitations prior to use.
General Information
Relations
Standards Content (Sample)
NOTICE: This standard has either been superseded and replaced by a new version or discontinued.
Contact ASTM International (www.astm.org) for the latest information.
Designation: D 6122 – 99 An American National Standard
Standard Practice for
Validation of Multivariate Process Infrared
Spectrophotometers
This standard is issued under the fixed designation D 6122; the number immediately following the designation indicates the year of
original adoption or, in the case of revision, the year of last revision. A number in parentheses indicates the year of last reapproval. A
superscript epsilon (e) indicates an editorial change since the last revision or reapproval.
1. Scope 1.6 This practice is not intended as a quantitative perfor-
mance standard for the comparison of analyzers of different
1.1 This practice covers requirements for the validation of
design.
measurements made by on-line, process near- or mid-infrared
1.7 Although this practice deals primarily with validation of
analyzers, or both, used in the calculation of physical, chemi-
on-line, process infrared analyzers, the procedures and statis-
cal, or quality parameters of liquid petroleum products. The
tical tests described herein are also applicable to at-line and
parameters are calculated from spectroscopic data using mul-
laboratory infrared analyzers which employ multivariate mod-
tivariate modeling methods. The requirements include verifi-
els.
cation of adequate instrument performance, verification of the
1.8 This standard does not purport to address all of the
applicability of the calibration model to the spectrum of the
safety concerns, if any associated with its use. It is the
sample under test, and verification of equivalence between the
responsibility of the user of this standard to consult and
result calculated from the infrared measurements and the result
establish appropriate safety and health practices and deter-
produced by the primary method used for the development of
mine the applicability of regulatory limitations prior to use.
the calibration model.
1.2 This practice does not cover procedures for establishing
2. Referenced Documents
the calibration model used by the analyzer. Calibration proce-
2.1 ASTM Standards:
dures are covered in Practices E 1655 and references therein.
D 1265 Practice for Sampling Liquefied Petroleum Gases
1.3 This practice is intended as a review for experienced
D 3764 Practice for Validation of Process Stream Analyz-
persons. For novices, this practice will serve as an overview of
ers
techniques used to verify instrument performance, to verify
D 4057 Practice for Manual Sampling of Petroleum and
model applicability to the spectrum of the sample under test,
Petroleum Products
and to verify equivalence between the parameters calculated
D 4177 Practice for Automatic Sampling of Petroleum and
from the infrared measurement and the results of the primary
Petroleum Products
method measurement.
D 6299 Practice for Applying Statistical Quality Assurance
1.4 This practice teaches and recommends appropriate sta-
Techniques to Evaluate Analytical Measurement System
tistical tools, outlier detection methods, for determining
Performance
whether the spectrum of the sample under test is a member of
E 131 Terminology Relating to Molecular Spectroscopy
the population of spectra used for the analyzer calibration. The
E 275 Practice for Describing and Measuring Performance
statistical tools are used to determine if the infrared measure-
of Ultraviolet, Visible, and Near Infrared Spectrophotom-
ment results in a valid property or parameter estimate.
eters
1.5 The outlier detection methods do not define criteria to
E 932 Practice for Describing and Measuring Performance
determine whether the sample, or the instrument is the cause of
of Dispersive Infrared Spectrophotometers
an outlier measurement. Thus, the operator who is measuring
E 1421 Practice for Describing and Measuring Performance
samples on a routine basis will find criteria to determine that a
of Fourier Transform Infrared (FT-IR) Spectrometers:
spectral measurement lies outside the calibration, but will not
Level Zero and Level One Tests
have specific information on the cause of the outlier. This
E 1655 Practices for Infrared, Multivariate, Quantitative
practice does suggest methods by which instrument perfor-
Analysis
mance tests can be used to indicate if the outlier methods are
E 1866 Guide for Establishing Spectrophotometer Perfor-
responding to changes in the instrument response.
mance Tests
E 1944 Practice for Describing and Measuring Performance
This practice is under the jurisdiction of ASTM Committee D-2 on Petroleum
Products and Lubricants and is the direct responsibility of Subcommittee D02.25 on
Validation of Process Analyzers and Statistical Quality Assurance of Measurement Annual Book of ASTM Standards, Vol 05.01.
Processes for Petroleum and Petroleum Products. Annual Book of ASTM Standards, Vol 05.02.
Current edition approved Dec. 10, 1999. Published December 1999. Originally Annual Book of ASTM Standards, Vol 05.04.
published as D 6122–97. Last previous edition D 6122–97. Annual Book of ASTM Standards, Vol 03.06.
Copyright © ASTM, 100 Barr Harbor Drive, West Conshohocken, PA 19428-2959, United States.
D 6122
of Fourier Transform Near-Infrared (FT-NIR) Spectrom- analyzer by optical fibers.
eters: Level Zero and Level One Tests
3.4.17 instrument, n—spectrophotometer, associated elec-
tronics and computer, spectrophotometer cell and, if utilized,
3. Terminology
transfer optics.
3.1 Definitions:
3.4.18 instrument standardization, n—a procedure for stan-
3.2 For definitions of terms and symbols relating to IR
dardizing the response of multiple instruments such that a
spectroscopy, refer to Terminology E 131.
common multivariate model is applicable for measurements
3.3 For definitions of terms and symbols relating to multi-
conducted by these instruments, the standardization being
variate calibration, refer to Practices E 1655.
accomplished by way of adjustment of the spectrophotometer
3.4 Definitions of Terms Specific to This Standard:
hardware or by way of mathematical treatment of the collected
3.4.1 action limit, n—the limiting value from an instrument
spectra.
performance test, beyond which the analyzer is expected to
3.4.19 line sample, n—a process or product sample which is
produce potentially invalid results.
withdrawn from a sample port in accordance with Practices
3.4.2 analyzer , n—all piping, hardware, computer, soft-
D 1265, D 4057, or D 4177, whichever is applicable, during a
ware, instrumentation and calibration model required to auto-
period when the material flowing through the analyzer is of
matically perform analysis of a process or product stream.
uniform quality and the analyzer result is essentially constant.
3.4.3 analyzer calibration, n—see multivariate calibration.
3.4.20 moving range of two control chart, n— a control
3.4.4 analyzer intermediate precision, n— a statistical mea-
chart that monitors the change in the absolute value of the
sure of the expected long-term variability of analyzer results
difference between two successive differences of the analyzer
for samples whose spectra are neither outliers, nor nearest
result minus the result from the primary method.
neighbor inliers.
3.4.21 multivariate calibration, n—an analyzer calibration
3.4.5 analyzer model, n—see multivariate model.
that relates the spectrum at multiple wavelengths or frequen-
3.4.6 analyzer repeatability, n—a statistical measure of the
cies to the physical, chemical, or quality parameters.
expected short-term variability of results produced by the
3.4.22 multivariate model, n—a multivariate, mathematical
analyzer for samples whose spectra are neither outliers nor
rule or formula used to calculate physical, chemical, or quality
nearest neighbor inliers.
parameters from the measured infrared spectrum.
3.4.7 analyzer result, n—the numerical estimate of a
3.4.23 nearest neighbor distance inlier, n— a spectrum
physical, chemical, or quality parameter produced by applying
residing within a gap in the multivariate calibration space, the
the calibration model to the spectral data collected by the
result for which is subject to possible interpolation error.
analyzer.
3.4.24 optical background, n—the spectrum of radiation
3.4.8 analyzer validation test,, n—see validation test.
incident on a sample under test, typically obtained by measur-
3.4.9 calibration transfer, n— a method of applying a
ing the radiation transmitted through the spectrophotometer
multivariate calibration developed on one analyzer to a differ-
cell when no sample is present, or when an optically thin or
ent analyzer by mathematically modifying the calibration
nonabsorbing liquid is present.
model or by instrument standardization.
3.4.25 optical reference filter, n—an optical filter or other
3.4.10 check sample, n—a single, pure liquid hydrocarbon
device which can be inserted into the optical path in the
compound, or a known, reproducible mixture of liquid hydro-
spectrophotometer or probe producing an absorption spectrum
carbon compounds whose spectrum is constant over time such
which is known to be constant over time, such that it can be
that it can be used in a performance test.
used in place of a check or test sample in a performance test.
3.4.11 control limits, n—limits on a control chart which are
3.4.26 outlier detection limits, n—the limiting value for
used as criteria for signaling the need for action, or for judging
application of an outlier detection method to a spectrum,
whether a set of data does or does not indicate a state of
beyond which the spectrum represents an extrapolation of the
statistical control. E 456
calibration model.
3.4.12 exponentially weighted moving average control
3.4.27 outlier detection methods, n—statistical tests which
chart, n—a control chart based on the exponentially weighted
are conducted to determine if the analysis of a spectrum using
average of individual observations from a system; the obser-
a multivariate model represents an interpolation of the model.
vations may be the differences between the analyzer result, and
3.4.28 outlier spectrum, n—a spectrum whose analysis by a
the result from the primary method.
multivariate model represents an extrapolation of the model.
3.4.13 individual observation control chart, n—a control
3.4.29 performance test, n—a test that verifies that the
chart of individual observations from a system; the observa-
performance of the instrument is consistent with historical data
tions may be the differences between the analyzer result and
and adequate to produce valid results.
the result from the primary method.
3.4.14 inlier, n—see nearest neighbor distance inlier. 3.4.30 physical correction, n— a type of pos processing
where the correction made to the numerical value produced by
3.4.15 inlier detection methods, n—statistical tests which
are conducted to determine if a spectrum resides within a the multivariate model is based on a separate physical mea-
surement of, for example, sample density, sample path length,
region of the multivariate calibration space which is sparsely
populated. or particulate scattering.
3.4.16 in-line probe, n—a spectrophotometer cell installed 3.4.31 post-processing, v—performing a mathematical op-
in a process pipe or slip stream loop and connected to the eration on an intermediate analyzer result to produce the final
D 6122
result, including correcting for temperature effects, adding a verify that the instrument is functioning properly. The intent of
mean property value of the analyzer calibration, and converting these tests is to provide a rapid indication of the state of the
into appropriate units for reporting purposes. instrument. These tests are necessary but not sufficient to
3.4.32 pre-processing, v—performing mathematical opera- demonstrate valid analyzer results.
tions on raw spectral data prior to multivariate analysis or 4.4 After the initial performance test is successfully com-
model development, such as selecting wave length regions, pleted, an initial validation test is conducted to verify that the
correcting for baseline, smoothing, mean centering, and assign- results produced by the analyzer are in statistical agreement
ing weights to certain spectral positions. with results for the primary method. Once this initial validation
3.4.33 primary method, n—the analytical procedure used to is completed, the analyzer results are considered valid for
generate the reference values against which the analyzer is both samples whose spectra are neither outliers or nearest neighbor
calibrated and validated; Practices E 1655 uses the term inliers.
reference method in place of the term primary method. 4.5 During routine operation of the analyzer, validation tests
3.4.34 process analyzer system, n—see analyzer. are conducted on a regular, periodic basis to demonstrate that
3.4.35 process analyzer validation samples, n—see valida- the analyzer results remain in statistical agreement with results
tion samples. for the primary method. Between validation tests, performance
3.4.36 spectrophotometer cell, n— an apparatus which al- tests are conducted to verify that the instrument is performing
lows a liquid hydrocarbon to flow between two optical surfaces in a consistent fashion.
which are separated by a fixed distance, the sample pathlength,
5. Significance and Use
while simultaneously allowing light to pass through the liquid.
5.1 The primary purpose of this practice is to permit the user
3.4.37 test sample, n—a process or product sample, or a
to validate numerical values produced by a multivariate,
mixture of process or product samples, which has a constant
infrared or near-infrared, on-line, process analyzer calibrated to
spectrum for a finite time period, and which can be used in a
measure a specific chemical concentration, chemical property,
performance test; test samples and their spectra are generally
or physical property. The validated analyzer results are ex-
not reproducible in the long term.
pected to be equivalent, over diverse samples whose spectra
3.4.38 transfer optics, n—a device which allows movement
are neither outliers or nearest neighbor inliers, to those
of light from the spectrophotometer to a remote spectropho-
produced by the primary method to within control limits
tometer cell and back to the spectrophotometer; transfer optics
established by control charts for the prespecified statistical
include optical fibers or other optical light pipes.
confidence level.
3.4.39 validation samples, n—samples that are used to
5.2 Procedures are described for verifying that the instru-
compare the analyzer results to the primary method results
ment, the model, and the analyzer system are stable and
through the use of control charts and statistical tests; validation
prope
...
Questions, Comments and Discussion
Ask us and Technical Secretary will try to provide an answer. You can facilitate discussion about the standard in here.