EN ISO 12099:2017

SLOVENSKI STANDARD
01-november-2017
1DGRPHãþD
SIST EN ISO 12099:2010
.UPDåLWRLQPOHYVNLSURL]YRGL6PHUQLFH]DXSRUDEREOLåQMHLQIUDUGHþH
VSHNWURPHWULMH,62
Animal feeding stuffs, cereals and milled cereal products - Guidelines for the application
of near infrared spectrometry (ISO 12099:2017)
Futtermittel, Getreide und gemahlene Getreideerzeugnisse - Anleitung für die
Anwendung von Nahinfrarot-Spektrometrie (ISO 12099:2017)
Aliments des animaux, céréales et produits de mouture des céréales - Lignes directrices
pour l'application de la spectrométrie dans le proche infrarouge (ISO 12099:2017)
Ta slovenski standard je istoveten z: EN ISO 12099:2017
ICS:
65.120 Krmila Animal feeding stuffs
67.060 äLWDVWURþQLFHLQSURL]YRGLL] Cereals, pulses and derived
QMLK products
2003-01.Slovenski inštitut za standardizacijo. Razmnoževanje celote ali delov tega standarda ni dovoljeno.

EN ISO 12099
EUROPEAN STANDARD
NORME EUROPÉENNE
September 2017
EUROPÄISCHE NORM
ICS 65.120 Supersedes EN ISO 12099:2010
English Version
Animal feeding stuffs, cereals and milled cereal products -
Guidelines for the application of near infrared
spectrometry (ISO 12099:2017)
Aliments des animaux, céréales et produits de mouture Futtermittel, Getreide und gemahlene
des céréales - Lignes directrices pour l'application de la Getreideerzeugnisse - Anleitung für die Anwendung
spectrométrie dans le proche infrarouge (ISO von Nahinfrarot-Spektrometrie (ISO 12099:2017)
12099:2017)
This European Standard was approved by CEN on 14 July 2017.

CEN members are bound to comply with the CEN/CENELEC Internal Regulations which stipulate the conditions for giving this
European Standard the status of a national standard without any alteration. Up-to-date lists and bibliographical references
concerning such national standards may be obtained on application to the CEN-CENELEC Management Centre or to any CEN
member.
This European Standard exists in three official versions (English, French, German). A version in any other language made by
translation under the responsibility of a CEN member into its own language and notified to the CEN-CENELEC Management
Centre has the same status as the official versions.

CEN members are the national standards bodies of Austria, Belgium, Bulgaria, Croatia, Cyprus, Czech Republic, Denmark, Estonia,
Finland, Former Yugoslav Republic of Macedonia, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Latvia, Lithuania,
Luxembourg, Malta, Netherlands, Norway, Poland, Portugal, Romania, Serbia, Slovakia, Slovenia, Spain, Sweden, Switzerland,
Turkey and United Kingdom.
EUROPEAN COMMITTEE FOR STANDARDIZATION
COMITÉ EUROPÉEN DE NORMALISATION

EUROPÄISCHES KOMITEE FÜR NORMUNG

CEN-CENELEC Management Centre: Avenue Marnix 17, B-1000 Brussels
© 2017 CEN All rights of exploitation in any form and by any means reserved Ref. No. EN ISO 12099:2017 E
worldwide for CEN national Members.

Contents Page
European foreword . 3

European foreword
This document (EN ISO 12099:2017) has been prepared by Technical Committee ISO/TC 34 “Food
products" in collaboration with Technical Committee CEN/TC 327 “Animal feeding stuffs - Methods of
sampling and analysis” the secretariat of which is held by NEN.
This European Standard shall be given the status of a national standard, either by publication of an
identical text or by endorsement, at the latest by March 2018 and conflicting national standards shall be
withdrawn at the latest by March 2018.
Attention is drawn to the possibility that some of the elements of this document may be the subject of
patent rights. CEN shall not be held responsible for identifying any or all such patent rights.
This document supersedes EN ISO 12099:2010.
According to the CEN-CENELEC Internal Regulations, the national standards organizations of the
following countries are bound to implement this European Standard: Austria, Belgium, Bulgaria,
Croatia, Cyprus, Czech Republic, Denmark, Estonia, Finland, Former Yugoslav Republic of Macedonia,
France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Latvia, Lithuania, Luxembourg, Malta,
Netherlands, Norway, Poland, Portugal, Romania, Serbia, Slovakia, Slovenia, Spain, Sweden, Switzerland,
Turkey and the United Kingdom.
Endorsement notice
The text of ISO 12099:2017 has been approved by CEN as EN ISO 12099:2017 without any modification.
INTERNATIONAL ISO
STANDARD 12099
Second edition
2017-08
Animal feeding stuffs, cereals and
milled cereal products — Guidelines
for the application of near infrared
spectrometry
Aliments des animaux, céréales et produits de mouture des céréales —
Lignes directrices pour l’application de la spectrométrie dans le
proche infrarouge
Reference number
ISO 12099:2017(E)
©
ISO 2017
ISO 12099:2017(E)
© ISO 2017, Published in Switzerland
All rights reserved. Unless otherwise specified, no part of this publication may be reproduced or utilized otherwise in any form
or by any means, electronic or mechanical, including photocopying, or posting on the internet or an intranet, without prior
written permission. Permission can be requested from either ISO at the address below or ISO’s member body in the country of
the requester.
ISO copyright office
Ch. de Blandonnet 8 • CP 401
CH-1214 Vernier, Geneva, Switzerland
Tel. +41 22 749 01 11
Fax +41 22 749 09 47
copyright@iso.org
www.iso.org
ii © ISO 2017 – All rights reserved

ISO 12099:2017(E)
Contents Page
Foreword .iv
Introduction .v
1 Scope . 1
2 Normative references . 1
3 Terms and definitions . 1
4 Principle . 2
5 Apparatus . 2
6 Calibration and initial validation . 2
6.1 General . 2
6.2 Reference methods . 3
6.3 Outliers . 3
6.4 Validation of calibration models . 3
6.4.1 General. 3
6.4.2 Bias correction . 4
6.4.3 Slope adjustment . 4
6.4.4 Expansion of calibration set . 4
6.5 Changes in measuring and instrument conditions . 4
7 Statistics for performance measurement . 5
7.1 General . 5
7.2 Plot the results . 5
7.3 Bias . 6
7.4 Root mean square error of prediction (s ) . 8
RMSEP
7.5 Standard error of prediction (s ) . 8
SEP
7.6 Slope .10
8 Sampling .12
9 Procedure.12
9.1 Preparation of test sample .12
9.2 Measurement .12
9.3 Evaluation of result .12
10 Checking instrument stability .13
10.1 Control sample .13
10.2 Instrument diagnostics .13
10.3 Instruments in a network .13
11 Running performance check of calibration .13
11.1 General .13
11.2 Control charts using the difference between reference and NIR results .14
12 Precision and accuracy .15
12.1 Repeatability .15
12.2 Reproducibility .15
12.3 Accuracy .15
12.4 Uncertainty .15
13 Test report .15
Annex A (informative) Guidelines for specific NIR standards .16
Annex B (informative) Examples of outliers and control charts .17
Annex C (informative) Supplementary terms and definitions .23
Bibliography .28
ISO 12099:2017(E)
Foreword
ISO (the International Organization for Standardization) is a worldwide federation of national standards
bodies (ISO member bodies). The work of preparing International Standards is normally carried out
through ISO technical committees. Each member body interested in a subject for which a technical
committee has been established has the right to be represented on that committee. International
organizations, governmental and non-governmental, in liaison with ISO, also take part in the work.
ISO collaborates closely with the International Electrotechnical Commission (IEC) on all matters of
electrotechnical standardization.
The procedures used to develop this document and those intended for its further maintenance are
described in the ISO/IEC Directives, Part 1. In particular the different approval criteria needed for the
different types of ISO documents should be noted. This document was drafted in accordance with the
editorial rules of the ISO/IEC Directives, Part 2 (see www .iso .org/ directives).
Attention is drawn to the possibility that some of the elements of this document may be the subject of
patent rights. ISO shall not be held responsible for identifying any or all such patent rights. Details of
any patent rights identified during the development of the document will be in the Introduction and/or
on the ISO list of patent declarations received (see www .iso .org/ patents).
Any trade name used in this document is information given for the convenience of users and does not
constitute an endorsement.
For an explanation on the voluntary nature of standards, the meaning of ISO specific terms and
expressions related to conformity assessment, as well as information about ISO’s adherence to the
World Trade Organization (WTO) principles in the Technical Barriers to Trade (TBT) see the following
URL: w w w . i s o .org/ iso/ foreword .html.
This document was prepared by Technical Committee ISO/TC 34, Food products, Subcommittee SC 10,
Animal feeding stuffs.
This second edition cancels and replaces the first edition (ISO 12099:2010), which has been technically
revised.
iv © ISO 2017 – All rights reserved

ISO 12099:2017(E)
Introduction
This document has been drafted using, as a basis, ISO 21543 | IDF 201, which was prepared by Technical
Committee ISO/TC 34, Food products, Subcommittee SC 5, Milk and milk products, and the International
Dairy Federation (IDF).
INTERNATIONAL STANDARD ISO 12099:2017(E)
Animal feeding stuffs, cereals and milled cereal
products — Guidelines for the application of near infrared
spectrometry
1 Scope
This document gives guidelines for the determination by near infrared spectroscopy of constituents
such as moisture, fat, protein, starch and crude fibre and parameters such as digestibility in animal
feeding stuffs, cereals and milled cereal products.
The determinations are based on spectrometric measurement in the near infrared spectral region.
2 Normative references
There are no normative references in this document.
3 Terms and definitions
For the purposes of this document, the following terms and definitions apply.
ISO and IEC maintain terminological databases for use in standardization at the following addresses:
— IEC Electropedia: available at http:// www .electropedia .org/
— ISO Online browsing platform: available at http:// www .iso .org/ obp
3.1
near infrared instrument
NIR instrument
apparatus which, when used under the conditions defined in this document, predicts constituent
contents (3.3) and technological parameters (3.4) in animal feeding stuffs (3.2), cereals and milled cereal
products through relationships to absorptions in the near infrared range
3.2
animal feeding stuffs
substance or product, including additives, whether processed, partially processed or unprocessed,
intended to be used for oral feeding to animals
EXAMPLE Raw materials, fodder, meat and bone meal, mixed feed and other end products, pet food, etc.
3.3
constituent content
mass fraction of substances determined using the appropriate, standardized or validated chemical
method
Note 1 to entry: The mass fraction is often expressed as a percentage.
Note 2 to entry: For examples of appropriate methods, see References [1] to [12].
EXAMPLE Moisture, fat, protein, crude fibre, neutral detergent fibre and acid detergent fibre.
ISO 12099:2017(E)
3.4
technological parameter
property or functionality of animal feeding stuffs (3.2), cereals and milled cereal products that can be
determined using the appropriate, standardized or validated method(s)
Note 1 to entry: It is possible to develop and validate NIR methods for other parameters and sample types than
listed above, as long as the procedure from this document is observed. The measuring units of the parameters
determined follow the units used in the reference methods.
EXAMPLE Digestibility.
4 Principle
Spectral data in the near infrared region are collected and transformed to constituent or parameter
concentrations by calibration models developed on representative samples of the products concerned.
5 Apparatus
5.1 Near infrared instruments.
Instruments based on diffuse reflectance or transmittance measurement covering the near infrared
−1 −1
wavelength region of 770 nm to 2 500 nm (12 900 cm to 4 000 cm ) or segments of this or at selected
wavelengths or wavenumbers. The optical principle may be dispersive (e.g. grating monochromators),
interferometric or non-thermal (e.g. light emitting diodes, laser diodes and lasers). The instrument
should be provided with a diagnostic test system for testing photometric noise and reproducibility,
wavelength/wavenumber accuracy and wavelength/wavenumber precision (for scanning
spectrophotometers).
The instrument should measure a sufficiently large sample volume or surface to eliminate any
significant influence of inhomogeneity derived from chemical composition or physical properties of
the test sample. The sample path length (sample thickness) in transmittance measurements should
be optimized according to the manufacturer’s recommendation with respect to signal intensity for
obtaining linearity and maximum signal/noise ratio.
5.2 Appropriate milling or grinding device, for preparing the sample (if needed).
NOTE Changes in grinding or milling conditions can influence NIR measurements due, for example, to
heating which can drive off volatile components such as water.
6 Calibration and initial validation
6.1 General
The instrument shall be calibrated before use. Calibration involves the comparison with a reference
and adjustment processes to the instrument. Because a number of different calibration systems can be
applied with NIR instruments, no specific procedure can be given for calibration.
For an explanation of methods for calibration development, see Reference [16] and the respective
manufacturer’s manual. For the validation, it is important to have a sufficient number of representative
samples, covering variations such as the following:
a) combinations and composition ranges of major and minor sample components;
b) seasonal, geographic and genetic effects on forages, feed raw material and cereals;
c) processing techniques and conditions;
d) storage conditions;
2 © ISO 2017 – All rights reserved

ISO 12099:2017(E)
e) sample and instrument temperature;
f) instrument variations (i.e. differences between instruments).
NOTE For a solid validation, at least 20 samples are needed.
6.2 Reference methods
Internationally accepted reference methods for determination of moisture, fat, protein and other
constituents and parameters should be used. See References [1] to [12] for examples.
The reference method used for calibration should be in statistical control. It is essential to know the
precision of the reference method.
Where possible, references that provide measurement traceability to the SI (International system of
units), such as certified reference materials, should be used.
6.3 Outliers
In many situations, statistical outliers are observed during calibration and validation. Outliers may be
related to NIR data (spectral outliers, hereafter referred to as “x-outliers”) or errors in reference data
or samples with a different relationship between reference data and NIR data (hereafter referred to as
“y-outliers”); see Figures B.1 to B.5 for examples.
For the purpose of validation, samples are not to be regarded as outliers if they fulfil the following
conditions:
a) if they are within the working range of the constituents/parameters in the calibration(s);
b) if they are within the spectral variation of the calibration samples, as, for example, estimated by
Mahalanobis distance;
c) if the spectral residual is below a limit defined by the calibration process;
d) if the prediction residual is below a limit defined by the calibration process.
If a sample appears as an outlier, then it should be checked initially to see if it is an x-outlier. If it exceeds
the x-outlier limits defined for the calibration, it should be removed. If it is not an x-outlier, then both
the reference value and the NIR predicted value should be checked, e.g. by repeated measurements. If
these confirm the original values, then the sample should not be deleted and the validation statistics
should include this sample. If the repeat values show that either the original reference values or the NIR
predicted ones were in error, then the new values should be used.
6.4 Validation of calibration models
6.4.1 General
Before use, calibration equations shall be validated locally on an independent test set that is
representative of the sample population to be analysed. For the determination of bias, slope and for the
determination of standard error of prediction (SEP, see 7.5), at least 20 samples are needed. Validation
shall be carried out for each sample type, constituent/parameter, temperature and other factors known
to affect or expected to have an effect the measurement. The calibration is valid only for the variations,
i.e. sample types, range and temperature, used in the validation.
NOTE 1 Calibration models can only be used in the range they have been validated.
Results obtained on the independent test set are plotted, reference against NIR, and residuals against
reference results, to give a visual impression of the performance of the calibration. The SEP is calculated
(see 7.5) and the residual plot of data corrected for mean systematic error (bias) is examined for
outliers, i.e. samples with a residual exceeding ±3 s .
SEP
ISO 12099:2017(E)
If the validation process shows that the model cannot produce acceptable statistics, then it should not
be used.
NOTE 2 What will be acceptable will depend, for example, on the performance of the reference method, the
covered range, the purpose of the analysis, etc., and is up to the parties involved to decide.
Where available and suitable, reference materials or certified reference materials can be used as part of
validation of calibration models.
The next step is to fit NIR, y , and reference data, y , by linear regression (y = a + b × y ) to
NIRS ref ref NIRS
produce statistics that describe the validation results.
6.4.2 Bias correction
The data are also examined for a bias between the methods. If the difference between means of the
NIR predicted and reference values is significantly different from zero, then this indicates that the
calibration is biased. A bias may be removed by adjusting the constant term (see 7.3) in the calibration
equation.
6.4.3 Slope adjustment
If the slope, b, is significantly different from 1, the calibration is skewed.
Adjusting the slope/intercept of the calibration is generally not recommended unless the calibration
is applied to new types of samples or instruments. If a reinvestigation of the calibration does not
detect outliers, especially outliers with high leverage, it is preferable to expand the calibration set to
include more samples. However, if the slope is adjusted, the calibration should then be tested on a new
independent test set.
6.4.4 Expansion of calibration set
If the accuracy of the calibration does not meet expectations, the calibration set should be expanded
to include more samples or a new calibration should be made. In all cases when a new calibration
is developed on an expanded calibration set, the validation process should be repeated on a new
validation set. If necessary, expansion of the calibration set should be repeated until acceptable results
are obtained on a validation set.
6.5 Changes in measuring and instrument conditions
Unless additional calibration is performed, a local validation of a NIR method stating the accuracy of
the method can generally not be considered valid if the test conditions are changed.
For example, calibrations developed for a certain population of samples may not be valid for samples
outside this population, although the analyte concentration range is unchanged. A calibration developed
on grass silages from one area may not give the same accuracy on silages from another area if the
genetic, growing and processing parameters are different.
Changes in the sample presentation technique or the measuring conditions, e.g. temperature, not
included in the calibration set may also influence the analytical results.
Calibrations developed on a certain instrument cannot always be transferred directly to an identical
instrument operating under the same principle. It may be necessary to perform bias or slope /
intercept adjustments to calibration equations. In many cases, it will be necessary to standardize
[16]
the two instruments against each other before calibration equations can be transferred .
Standardization procedures can be used to transfer calibrations between instruments of different
types provided that samples are measured in the same way (reflectance, transmittance) and that the
spectral region is common.
If the conditions are changed, a supplementary validation should be performed.
4 © ISO 2017 – All rights reserved

ISO 12099:2017(E)
The calibrations should be checked whenever any major part of the instrument (optical system,
detector) has been changed or repaired.
7 Statistics for performance measurement
7.1 General
The performances of a prediction model shall be determined by a set of validation samples. This set
consists of samples which are independent of the calibration set. In a plant, it will be new batches; in
agriculture, it will be a new crop or a new experiment location.
This set of samples shall be carefully analysed following the reference methods. The care to analyse
validation samples shall be emphasized and the precision of these results is more important for the
validation set than for the samples used at the calibration phase.
The number of validation samples shall be at least 20 to compute the statistics with some confidence.
The NIR protocol used for the determination of the performances of the prediction model shall be the
same as that used in routine (one measurement or two measurements).
7.2 Plot the results
It is important to visualize the results in plots, i.e. reference vs. predicted values or residuals vs.
predicted values.
The residuals are defined by Formula (1):
ˆ
ey=− y (1)
ii i
where
th
y is the i reference value (y );
i ref
th
is the i predicted value (y ) obtained when applying the multivariate NIR model.
NIRS
ˆ
y
i
The way the differences are calculated will give a negative bias when the predictions are too high and a
positive one when the predictions are too low compared to the reference values.
A plot of the data immediately gives an overview of the correlation, the bias, the slope and the presence
of obvious outliers (see Figure 1).
ISO 12099:2017(E)
Key
1 45° line (ideal line with bias = 0 and slope = 1) X Y
NIRS
2 (45° − bias) line Y Y
ref
3 linear regression line
4 outliers
5 bias
NOTE The outliers (key 4) have a strong influence on the calculation of the slope and should be removed if
the results are to be used for adjustments.
Figure 1 — Scatter plot for a validation set, y = f(a + b × y )
ref NIRS
7.3 Bias
Most of the time, a bias or systematic error is observed with NIR models. Bias can occur due to several
causes: new samples of a type not previously seen by the model, drift of the instrument, drift in wet
chemistry, changes in the process, in the sample preparation, etc.
With n, the number of independent samples, the bias (or offset) is the mean difference and can be
defined by Formula (2):
n
e= e (2)
∑ i
n
i=1
where e is the residual as defined by Formula (1) resulting in Formula (3):
i
n n
 

 
e =−yy =−yy (3)
∑∑i i
 
n
i==11i
 
6 © ISO 2017 – All rights reserved

ISO 12099:2017(E)
where
th
y is the i reference value;
i
ˆ th
y is the i predicted value;
i
y is the mean of the reference values;

y is the mean of the predicted values.
The significance of the bias is checked by a t test. The calculation of the bias confidence limits (BCLs),
T , determines the limits for accepting or rejecting formula performance on the small set of samples
b
chosen from the new population; see Formula (4):
 
Tt=± sn/ (4)
b (/12−α ) SEP
 
where
α is the probability of making a type I error;
t is the appropriate student’s t value for a two-tailed test with degrees of freedom associated
with s and the selected probability of a type I error;
SEP
n is the number of independent samples;
s is the standard error of prediction (defined in 7.5).
SEP
As an example, with n = 20, and s = 1, the BCLs are as in Formula (5):
SEP
T =± 20,/91× 20 =±04, 8 (5)
()
b
This means that the bias tested with 20 samples shall be higher than 48 % of the standard error of
prediction to be considered as different from zero.
Table 1 — Values of the t distribution with a probability α = 0,05 (5 %)
n t values n t values
10 2,23 75 1,99
15 2,13 100 1,98
20 2,09 200 1,97
30 2,04 500 1,96
40 2,02 1000 1,96
50 2,01 — —
a
NOTE  The Excel function “TINV” can be used.
a
Excel is the trade name of a product supplied by Microsoft. This
information is given for the convenience of users of this document and does
not constitute an endorsement by ISO of the product named. Equivalent
products may be used if they can be shown to lead to the same results.
ISO 12099:2017(E)
7.4 Root mean square error of prediction (s )
RMSEP
The s is defined by Formula (6):
RMSEP
n
e
∑ i
i=1
s = (6)
RMSEP
n
where
th
e is the residual of the i sample;
i
n is the number of independent samples.
This value can be compared with s and s (see Annex C).
SEC SECV
S includes the random error (s ) and the systematic error (bias). It includes also the error of the
RMSEP SEP
reference methods (as do s and s ); see Formula (7):
SEC SECV
()n−1
s = se+ (7)
RMSEPSEP
n
where
n is the number of independent samples;
s is the standard error of prediction (defined in 7.5);
SEP
is the bias or systematic error.
e
There is no direct test for S . This is the reason to separate the systemic error (bias or e ) and the
RMSEP
random error s .
SEP
7.5 Standard error of prediction (s )
SEP
The standard error of prediction (s ), i.e. standard deviation of the residuals, which expresses the
SEP
accuracy of routine NIR results corrected for the mean difference (bias) between routine NIR and
reference method, can be calculated by using Formula (8):
n
ee−
()
∑ i
i=1
s = (8)
SEP
n−1
where
n is the number of independent samples;
th
e is the residual of the i sample;
i
is the bias or systematic error.
e
The s should be related to the s (respectively, s ; see Annex C) to check the validity of the
SEP SEC SECV
calibration model for the selected validation set.
8 © ISO 2017 – All rights reserved

ISO 12099:2017(E)
The unexplained error confidence limits (UECLs), T , are calculated from an F-test (ratio of two
UE
variances) (see Reference [18] and Table 2). See Formula (9):
Ts= F (9)
UE SEC (,α vM,)
where
s is the standard error of calibration (see Annex C);
SEC
α is the probability of making a Type I error;
v is n − 1 numerator degrees of freedom associated with s of the test set;
SEP
n is the number of samples in the validation process;
M is n – p − 1 denominator degrees of freedom associated with s (standard error of calibration)
c SEC
[n is the number of calibration samples, p is the number of terms or PLS factors of the model
c
or weights in the case of ANN (see Annex C). In ANN, weights are all unknown parameters in
the model].
NOTE 1 s can be replaced by s , which is a better statistic than s ; very often, s is too optimistic;
SEC SECV SEC SEC
(s > s > s ).
SEP SECV SEC
EXAMPLE Where n = 20, α = 0,05, M = 100, and s = 1, gives the following value: T = 1,30.
SEP UE
With 20 samples, a s that is up to 30 % larger than the s can be accepted.
SEP SEC
1)
NOTE 2 The Excel function “FINV” can be used.
The F-test cannot be used to compare two calibrations on the same validation set. It needs (as here) two
independent sets to work. Another test shall be used to compare two or more models on the same data set.
1) Excel is the trade name of a product supplied by Microsoft. This information is given for the convenience of
users of this document and does not constitute an endorsement by ISO of the product named. Equivalent products
may be used if they can be shown to lead to the same results.
ISO 12099:2017(E)
Table 2 — F values and squared root of the F values in function of the degrees of freedom of the
numerator associated with s and of the denominator associated with s
SEP SEC
F(α,v,M)
Fvα ,,M
()
df (s ) df (s )
SEC SEC
df (s ) 50 100 200 500 1 000 df (s ) 50 100 200 500 1 000
SEP SEP
10 2,03 1,93 1,88 1,85 1,84 10 1,42 1,39 1,37 1,36 1,36
11 1,99 1,89 1,84 1,81 1,80 11 1,41 1,37 1,36 1,34 1,34
12 1,95 1,85 1,80 1,77 1,76 12 1,40 1,36 1,34 1,33 1,33
13 1,92 1,82 1,77 1,74 1,73 13 1,39 1,35 1,33 1,32 1,32
14 1,89 1,79 1,74 1,71 1,70 14 1,38 1,34 1,32 1,31 1,30
15 1,87 1,77 1,72 1,69 1,68 15 1,37 1,33 1,31 1,30 1,29
16 1,85 1,75 1,69 1,66 1,65 16 1,36 1,32 1,30 1,29 1,29
17 1,83 1,73 1,67 1,64 1,63 17 1,35 1,31 1,29 1,28 1,28
18 1,81 1,71 1,66 1,62 1,61 18 1,30 1,31 1,29 1,27 1,27
19 1,80 1,69 1,64 1,61 1,60 19 1,34 1,30 1,28 1,27 1,26
29 1,69 1,58 1,52 1,49 1,48 29 1,30 1,26 1,23 1,22 1,22
49 1,60 1,48 1,42 1,38 1,37 49 1,27 1,22 1,19 1,17 1,17
99 1,53 1,39 1,32 1,28 1,26 99 1,24 1,18 1,15 1,13 1,12
199 1,48 1,34 1,26 1,21 1,19 199 1,22 1,16 1,12 1,10 1,09
499 1,46 1,31 1,22 1,16 1,13 499 1,21 1,14 1,11 1,08 1,07
999 1,45 1,30 1,21 1,14 1,11 999 1,20 1,14 1,10 1,07 1,05
NOTE 1  See explanation to Formula (9).
NOTE 2  df is the degree of freedom; n − 1 for s ; and n − p − 1 for s .
SEP c SEC
7.6 Slope
ˆ
The slope, b, of the simple regression ya=+ by⋅ is often reported in the NIR reports and publications.
Notice that the slope shall be calculated with the reference values as the dependent variable and the
predicted NIR values as the independent variable, if the calculated slope is intended to be used for
adjustment of NIR results (like in the case of the inverse multivariate regression used to build the
prediction model).
From the least squares fitting, slope and intercept are calculated by Formula (10) and Formula (11),
respectively:

s
yy
b = (10)
s
ˆ
y
where
 is the covariance between reference and predicted values;
s
yy
is the variance of the n predicted values.
s
ˆ
y
10 © ISO 2017 – All rights reserved

ISO 12099:2017(E)

ay=−by (11)
where
is the mean of the reference values;
y
b is the slope;

is the mean of the predicted values.
y
As for the bias, a t test can be calculated to check the hypothesis that b = 1 as in Formula (12):
sn. −1
()
ˆ
y
tb=−1 . (12)
obs
s
res
where
is the variance of the n predicted values;
s
ˆ
y
n is the number of independent samples;
s is the residual standard deviation.
res
The residual standard deviation, s , is defined in Formula (13):
res
n
 2
ya−+by 
()
∑ ii
 
i=1
s = (13)
res
n−2
where
n is the number of independent samples;
a is the intercept (see Formula (11);
b is the slope (see Formula (10);
th
y is the i reference value;
i

th
is the i predicted value obtained when applying the multivariate NIR model.
y
i
NOTE s is like s when the predicted values are corrected for slope and intercept. Be aware to not
res SEP
confuse bias and intercept. See also Figure 1.
The bias equals the intercept only when the slope is exactly one.
The slope, b, will be considered as different from 1 when Formula (14) applies:
tt≥ (14)
obs
()12−α/
ISO 12099:2017(E)
where
t is the observed t value, calculated according to Formula (12);
obs
t is the t value obtained from Table 1 for a probability of α = 0,05 (5 %).
(1−α/2)
A too-narrow range or uneven distribution will lead to useless correction of the slope even when
the s is correct. The slope can only be adjusted when the validation set covers a large part of the
SEP
calibration range.
EXAMPLE For n = 20 samples with a residual standard deviation [see Formula (13)] of 1, a standard deviation
of the predicted values of S = 2 and a calculated slope of b = 1,2, the observed t value is 1,7 and then the
obs
ˆ
y
slope is not significantly different from 1 as the t value (see Table 1) for n = 20 samples is 2,09. If the slope is 1,3,
the t value is 2,6 and then the slope is significantly different than 1.
obs
8 Sampling
Sampling is not part of the method specified in this document.
NOTE Recommended sampling procedures are given in ISO 6497 and ISO 24333.
It is important that the laboratory receives a sample which is truly representative and has not been
damaged or changed during transport or storage.
9 Procedure
9.1 Preparation of test sample
All laboratory samples should usually be kept under conditions that will not change the composition of
the sample from the time of sampling to the time of commencing the procedure.
The preparation of samples for routine measurements needs to be made in the same way as the
preparation of the validation samples. It is necessary to apply standard conditions.
Before the analysis, the sample should be taken in such a way as to obtain a sample representative of
the material to be analysed.
For specific procedures, see specific NIR standards.
Guidelines for specific NIR standards are given in Annex A.
9.2 Measurement
Follow the instructions of the instrument manufacturer/supplier.
The prepared sample should reach a temperature within the range included in the validation.
9.3 Evaluation of result
For the routine results to be valid, they shall be within the range of the calibration model used.
Results obtained on samples detected as spectral outliers cannot be regarded as reliable.
If multiple measurements are made on the same sample, calculate the arithmetic mean if the
repeatability conditions (see 12.1) are observed.
For the expression of results, refer to specific NIR standards.
12 © ISO 2017 – All rights reserved

ISO 12099:2017(E)
10 Checking instrument stability
10.1 Control sample
At least one control sample should be measured at least once per day to check instrument hardware
stability and to detect any malfunction. Knowledge of the true concentration of the analyte in the
control sample is not necessary. The sample material should be stable and, as far as possible, resemble
the samples to be analysed. The parameter(s) measured should be stable and, as far as possible, identical
to or at least biochemically close to the sample analyte. A sample is prepared as described in 9.1 and
stored in such a way as to maximize the storage life. These samples are normally stable for lengthy
periods but the stability should be tested in the actual
...