EN ISO 4259-4:2022

SLOVENSKI STANDARD
01-september-2022
Nafta in sorodni proizvodi - Natančnost merilnih metod in rezultatov - 4. del:
Uporaba grafikonov statističnega nadzora stanja "pod statističnim nadzorom“ za
izvajanje standardne preskusne metode v enem laboratoriju (ISO 4259-4:2021)
Petroleum and related products - Precision of measurement methods and results - Part
4: Use of statistical control charts to validate 'in-statistical-control' status for the execution
of a standard test method in a single laboratory (ISO 4259-4:2021)
Mineralölerzeugnisse - Präzision von Messverfahren und Ergebnissen - Teil 4:
Verwendung von Kontrollkarten zur Validierung des Status der statistischen Kontrolle bei
der Durchführung von genormten Prüfverfahren in einem einzelnen Labor (ISO 4259-
4:2021)
Produits pétroliers et connexes - Fidélité des méthodes de mesure et de leurs résultats -
Partie 4: Utilisation de cartes de contrôle statistique pour valider l’état 'sous maîtrise
statistique' pour l’exécution d'une méthode d'essai normalisée dans un seul laboratoire
(ISO 4259-4:2021)
Ta slovenski standard je istoveten z: EN ISO 4259-4:2022
ICS:
75.080 Naftni proizvodi na splošno Petroleum products in
general
75.180.30 Oprema za merjenje Volumetric equipment and
prostornine in merjenje measurements
2003-01.Slovenski inštitut za standardizacijo. Razmnoževanje celote ali delov tega standarda ni dovoljeno.

EN ISO 4259-4
EUROPEAN STANDARD
NORME EUROPÉENNE
June 2022
EUROPÄISCHE NORM
ICS 75.080
English Version
Petroleum and related products - Precision of
measurement methods and results - Part 4: Use of
statistical control charts to validate 'in-statistical-control'
status for the execution of a standard test method in a
single laboratory (ISO 4259-4:2021)
Produits pétroliers et connexes - Fidélité des méthodes Mineralölerzeugnisse - Präzision von Messverfahren
de mesure et de leurs résultats - Partie 4: Utilisation de und Ergebnissen - Teil 4: Verwendung von
cartes de contrôle statistique pour valider l'état 'sous Kontrollkarten zur Validierung des Status der
maîtrise statistique' pour l'exécution d'une méthode statistischen Kontrolle bei der Durchführung von
d'essai normalisée dans un seul laboratoire (ISO 4259- genormten Prüfverfahren in einem einzelnen Labor
4:2021) (ISO 4259-4:2021)
This European Standard was approved by CEN on 30 November 2021.

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, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Latvia, Lithuania, Luxembourg, Malta, Netherlands, Norway,
Poland, Portugal, Republic of North Macedonia, 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: Rue de la Science 23, B-1040 Brussels
© 2022 CEN All rights of exploitation in any form and by any means reserved Ref. No. EN ISO 4259-4:2022 E
worldwide for CEN national Members.

Contents Page
European foreword . 3

European foreword
This document (EN ISO 4259-4:2022) has been prepared by Technical Committee ISO/TC 28
"Petroleum and related products, fuels and lubricants from natural or synthetic sources" in
collaboration with Technical Committee CEN/TC 19 “Gaseous and liquid fuels, lubricants and related
products of petroleum, synthetic and biological origin” 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 December 2022, and conflicting national standards
shall be withdrawn at the latest by December 2022.
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.
Any feedback and questions on this document should be directed to the users’ national standards
body/national committee. A complete listing of these bodies can be found on the CEN website.
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, France, Germany, Greece, Hungary, Iceland,
Ireland, Italy, Latvia, Lithuania, Luxembourg, Malta, Netherlands, Norway, Poland, Portugal, Republic of
North Macedonia, Romania, Serbia, Slovakia, Slovenia, Spain, Sweden, Switzerland, Turkey and the
United Kingdom.
Endorsement notice
The text of ISO 4259-4:2021 has been approved by CEN as EN ISO 4259-4:2022 without any
modification.
INTERNATIONAL ISO
STANDARD 4259-4
First edition
2021-12
Petroleum and related products —
Precision of measurement methods
and results —
Part 4:
Use of statistical control charts to
validate 'in-statistical-control' status
for the execution of a standard test
method in a single laboratory
Produits pétroliers et connexes — Fidélité des méthodes de mesure et
de leurs résultats —
Partie 4: Utilisation de cartes de contrôle statistique pour valider
l’état 'sous maîtrise statistique' pour l’exécution d'une méthode
d'essai normalisée dans un seul laboratoire
Reference number
ISO 4259-4:2021(E)
ISO 4259-4:2021(E)
© ISO 2021
All rights reserved. Unless otherwise specified, or required in the context of its implementation, 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
CP 401 • Ch. de Blandonnet 8
CH-1214 Vernier, Geneva
Phone: +41 22 749 01 11
Email: copyright@iso.org
Website: www.iso.org
Published in Switzerland
ii
ISO 4259-4:2021(E)
Contents Page
Foreword .iv
Introduction .v
1 Scope . 1
2 Normative references . 1
3 Terms, definitions, symbols and abbreviated terms . 1
3.1 Specific terms and definitions . 2
3.2 Symbols and abbreviated terms . 2
4 Statistical control in the execution of a standard test method by a laboratory.3
4.1 General . 3
4.2 Control chart description . 4
4.2.1 General . 4
4.2.2 I- and MR-charts . 4
4.2.3 I-chart sensitivity enhancement strategy . 4
4.2.4 In-statistical-control conditions . 5
4.3 Control chart work process . 5
4.3.1 General . 5
4.3.2 Stage 1 of control chart work process . 5
4.3.3 Stage 2 of control chart work process . 9
4.4 QC material batch change transition . 11
4.4.1 General . 11
4.4.2 Procedure 1, concurrent testing .12
4.4.3 Procedure 2, Q-chart .12
4.4.4 Procedure 3, dynamically updated I-chart with EWMA .12
5 Guidance for insufficient variation or non-normal data .13
5.1 General requirement . 13
5.2 How to deal with insufficient variation or non-normal data .13
5.2.1 Insufficient variation .13
5.2.2 Non-normal data . 14
Annex A (informative) Details of the control chart work process .15
Annex B (normative) Check procedures .34
Bibliography .36
iii
ISO 4259-4:2021(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 of 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
www.iso.org/iso/foreword.html.
This document was prepared by Technical Committee ISO/TC 28, Petroleum and related products, fuels
and lubricants from natural or synthetic sources, in collaboration with the European Committee for
Standardization (CEN) Technical Committee CEN/TC 19, Gaseous and liquid fuels, lubricants and related
products of petroleum, synthetic and biological origin, in accordance with the Agreement on technical
cooperation between ISO and CEN (Vienna Agreement).
A list of all parts in the ISO 4259 series can be found on the ISO website.
Any feedback or questions on this document should be directed to the user’s national standards body. A
complete listing of these bodies can be found at www.iso.org/members.html.
iv
ISO 4259-4:2021(E)
Introduction
In the current global business environment, measurement data ‘trustworthiness’ is a key business
driver and an implicit expectation from customers and regulatory entities. Data trustworthiness
means the data quality meets expectations and is ‘fit-for-use’. Trustworthy data can only be produced
by measurement systems that are demonstrated to be stable and are under common cause variation
only.
This document describes the applications of specific statistical control charts selected from those that
are widely used by the manufacturing sector for the purpose of monitoring and demonstrating the
in-statistical-control status of a laboratory in the execution of a standardized test method to produce
trustworthy data.
[9]
In ISO 4259-2 , the requirement for assessment of product quality conformance to specification, is to be
interpreted that each laboratory’s test result is obtained from a test method that is in-statistical-control
in terms of precision and bias, to be substantiated by in-house statistical quality control (SQC) charts
or other equivalent statistical techniques. While in-house techniques are used by many laboratories for
test method quality assurance, standardization on how to establish in-statistical-control is necessary
[9]
to ensure consistency in application of ISO 4259-2 . Addressing the aforementioned necessity is the
[1]
motivation of this document, which is based on ASTM D6299 .
v
INTERNATIONAL STANDARD ISO 4259-4:2021(E)
Petroleum and related products — Precision of
measurement methods and results —
Part 4:
Use of statistical control charts to validate 'in-statistical-
control' status for the execution of a standard test method
in a single laboratory
1 Scope
This document specifies the process and methodology for the construction, operation, and maintenance
of statistical control charts to assess if a laboratory's execution of a standard test method is in-
statistical-control and how to establish and validate the 'in-statistical-control' status.
It specifies control charts that are most appropriate for ISO/TC 28 test methods where the dominant
common cause variation is associated with the long term, multiple operator conditions. The control
charts specified for determination of in-statistical-control are: individual (I), moving range of 2 (MR ),
and either the exponentially weighted moving average (EWMA) or zone-based run rules [similar to
[3]
Western Electric (WE) run rules ] as sensitivity enhancement strategy to support the I-chart.
The procedures in this document have been primarily designed for numerical results obtained from
testing of control samples prepared from a homogenous source of petroleum and related products in a
manner that preserves the homogeneity of properties of interest between control samples. If the test
method permits, a certified reference material (CRM) sample is used as a control sample provided the
sample composition is representative of the material being tested and is not a pure compound; if this is
done then the laboratory best establishes its own mean for the CRM sample.
This document is applicable to properties of interest that are (known to be) stable over time, and for
data sets with sufficient resolution to support validation of the assumption that the data distribution
can be approximately represented by the normal (Gaussian) model. Mitigating strategies are suggested
for situations where the assumption cannot be validated.
2 Normative references
The following documents are referred to in the text in such a way that some of their content support
requirements of this document. For dated references, only the edition cited applies. For undated
references, the latest edition of the referenced document (including any amendments) applies.
ISO 4259-1:2017, Petroleum and related products — Precision of measurement methods and results —
Part 1: Determination of precision data in relation to methods of test
3 Terms, definitions, symbols and abbreviated terms
For the purposes of this document, the terms and definitions given in ISO 4259-1 and the following
apply.
ISO and IEC maintain terminology databases for use in standardization at the following addresses:
— ISO Online browsing platform: available at https:// www .iso .org/ obp
— IEC Electropedia: available at https:// www .electropedia .org/
ISO 4259-4:2021(E)
3.1 Specific terms and definitions
3.1.1
common cause
factors that contribute to common cause variation (3.1.2)
Note 1 to entry: See Figure 1 for illustration.
3.1.2
common cause variation
variation amongst results collected under site precision conditions (3.1.4) from repeated execution of
a test method on the same test material attributable to known, unknown, or unknowable factors that
are intentionally not or cannot be rigidly controlled as part of the normal and correct execution of all
aspects of the test method
3.1.3
in-statistical-control
situation wherein the test results produced by the user on control samples are reasonably consistent
with expectation over time with common cause variation scattered around a stable expected centre
3.1.4
site precision conditions
conditions under which single test results are obtained in time intervals, separated by at least 8 h, by
the testing population in a single laboratory executing the same test method using the same apparatus
on test specimens taken at random from the same material over the normal daily operating envelope
3.1.5
quality control sample
QC sample
specimen taken from a stable and homogeneous material with composition and properties similar
to sample normally tested by the laboratory, prepared in a manner that preserves the homogeneity
of property of interest between test specimens, stored in a manner that preserves the properties of
interest over time, and available in sufficient quantity for repeated long-term testing
3.2 Symbols and abbreviated terms
AD Anderson Darling
ARV assigned reference value
CRM certified reference material
EWMA exponentially weighted moving average
GESD generalized extreme studentized deviation
MR moving range of two
moving range average associated with s
MR known
known
PT proficiency testing
QC quality control
q-q quantile-quantile, term used to describe the plot type comparing the z-score of a data point with is
numerical value
a
s statistically pooled standard deviation obtained using final achieved standard deviations from a
known
group of retired control charts where all the final achieved control chart averages are determined
to be not statistically significantly different using the appropriate clause(s) in stage 1
a
The range spanned by the final achieved control chart averages are referred to as the working range associated with
s . See concept illustrated in Table 1.
known
ISO 4259-4:2021(E)
Table 1 — Statistically pooled standard deviation concept
Material Property (unit) s df Working range MR
known
known
summer gasoline vapour pressure (kpa) 0,55 60 49,85 to 50,68 0,62
winter gasoline vapour pressure (kpa) 0,83 85 104,67 to 105,91 0,93
4 Statistical control in the execution of a standard test method by a laboratory
4.1 General
The execution of a standard test method by a laboratory is in this document considered as the execution
of a series of inter-connected work processes. Each work process is subject to variation caused by
known, unknown, or sometimes unknowable causes that are inherent to a process over long time
horizon such that every outcome of the process is affected. These causes are referred to as common
causes. The effect on the final process outcome due to common causes are referred to as common cause
variation.
Common causes for variation can be grouped into 5 categories (environment, operator, equipment,
procedure and reagent material) using a technique known as a fishbone diagram. Due to common cause
variation, repeated execution of the same test method on the same material over a long-time horizon
yields results that are not numerically identical. This effect is illustrated in Figure 1.
Figure 1 — Fishbone diagram representation of common cause variation in the execution of a
test method
The complete process associated with the execution of the specific test method is said to be 'in-
statistical-control' if the process outcomes (test results) from repeated analysis of QC samples prepared
from the same material are reasonably consistent with expectation over time, with random variation
scattered around a stable centre due to common causes only.
To determine 'in-statistical-control', a multi-step and integrated work process involving use of
statistical control charts and a QC sample is required.
NOTE For simplicity, the word 'statistical' will be omitted and the term referred to as 'control charts'
throughout the rest of this document.
ISO 4259-4:2021(E)
4.2 Control chart description
4.2.1 General
Control charts appropriate for most petroleum industry test methods that yield numeric results are the
individual (I) chart, and the moving range of 2 (MR ) chart.
[2]
NOTE 1 The I-chart is also known as the X-chart, or the Shewhart chart .
NOTE 2 For simplicity, the MR chart is referred to as the moving range (MR) chart throughout the rest of this
document.
4.2.2 I- and MR-charts
The I-chart is a graphic display of individual QC sample test results (X) collected under site precision
conditions, plotted in chronological order, overlaid with a centre line, lower and upper decision limits
that require action if exceeded. These limits are herein referred to as I-chart lower and upper control
limits (LCL_X, UCL_X). The primary purpose of the I-chart is to monitor process centre stability over
time.
NOTE Since the primary interest is to monitor the stability and common cause variation of the test process
under site precision conditions over a long time horizon, replicate analysis obtained under repeatability
conditions (see ISO 4259-1) does not contribute towards this objective, as the variation due to common causes of
interest is not contained in replicate results collected under repeatability conditions.
The MR-chart is the successive difference (with no arithmetic sign) of two individual results in the
I-chart, plotted in chronological order, also overlaid with a centre line and an upper decision limit for
action, herein referred to as MR-chart upper control limit (UCL_MR). The primary purpose of the MR-
chart is to monitor common cause variation stability between successive QC sample results over time.
The decision limits for action for both charts as well as the conditions requiring action for the strategies
in 4.2.3 are based on a very low theoretical probability (<0,3 %) of “action required” decision for a
process that is in-statistical-control, using the Normal distribution (I-chart) and W distribution (MR-
chart) as the reference statistical models. Hence, these limits represent the expectation limits for the
process outcome if it is in-statistical-control.
4.2.3 I-chart sensitivity enhancement strategy
As a direct consequence of setting the action limits for the I-chart based on a low probability of
exceedance for a process that is in-statistical-control, these limits are not sensitive to detection of small
changes in the process centre. It is therefore necessary to support I-chart with additional sensitivity
enhancement strategies to overcome this shortcoming.
This document requires use of one of the following strategies in conjunction with the I-chart:
a) Strategy 1: Zone-based run rules. Action is required if any of the following run rule conditions is
present. For definitions of zones see 4.3.2:
— two out of three consecutive individual results in Zone A on one side of the centre line;
— four out of five consecutive individual results beyond Zone C on one side of the centre line;
— nine consecutive individual results on the same side of the centre line (above or below);
or,
b) Strategy 2: Exponentially weighted moving average (EWMA). Action is required if either one of the
following conditions occur:
— any exceedance of the EWMA action limits;
— nine consecutive individual results on the same side of the centre line (above or below).
ISO 4259-4:2021(E)
The EWMA is a ‘time-weighted moving average’ calculated using all data points up to the most current
one, the weighting of each datum reduced with age exponentially. The rate of this weight decay is
controlled by λ. It is re-calculated with the arrival of each new datum and is judged against its own
action limits (herein referred to as the EWMA-action limits). An EWMA with λ = 0,4 has similar
[4]
detection power as Strategy 1 .
Use of Strategy 2 is recommended due to the ease of implementation and lower expected false alarm
rate than Strategy 1.
4.2.4 In-statistical-control conditions
The complete process associated with the execution of a test method is deemed to be in-statistical-
control if all of the following conditions are met:
a) all individual control sample results are within the I-chart action limits,
b) less than five out of 12 successive MR results exceed MR-chart upper control limit, and
c) no action required for the sensitivity enhancement chosen (Strategy 1 or 2).
4.3 Control chart work process
4.3.1 General
To determine if the complete process associated with the execution of a test method is in-statistical-
control, a two-stage multi-step work process involving use of control charts and QC material is specified.
Stage 1 comprises visual and statistical assessment of initial test results for a new batch of QC material
plotted in a chronological order (known as a run chart). This is followed by construction of the I-chart
and MR-chart using these results by overlaying the mean and the action limits onto the respective
charts. The action limits represent the boundaries within which the current and future test results
and MR for this QC material are expected to lie, on the assumption that the process is in-statistical-
control and the QC material remains unchanged. The control charts (I and MR) constructed in Stage 1
are deployed for Stage 2 if all in-statistical-control conditions (see 4.2.4) are met.
Stage 2 comprises of two modes, operation and maintenance. Under 'operation', future test results
(for the QC material tested in Stage 1) as they arrive in chronological order are compared against
the established action limits and chosen enhancement strategy in Stage 1. Under 'maintenance', the
statistics used in the computation of the control chart action limits from Stage 1 are re-assessed
periodically using newly accrued in-statistical-control results and updated as appropriate.
For this practice, the root-mean-square technique is used to compute the sample standard deviation
statistic, s.
4.3.2 Stage 1 of control chart work process
The primary objective of Stage 1 is to establish initial means and action limits for the I, MR control
charts and implement the chosen enhancement strategy (see 4.2.3) for a specific batch of QC material
using chronologically obtained data and the normal distribution as the reference model. Figure 2 is a
flow chart of the 15-step process defined in this subclause. The main steps are:
1) Prepare multiple QC samples from a stable and homogeneous material with composition and
properties similar to samples normally tested by the laboratory.
2) Collect a minimum of 20 QC sample test results under site precision conditions.
3) Plot the individual results chronologically (this is called a run chart) and study the plot for any
visually discernible transcription errors. Correct or discard obvious transcription errors.
ISO 4259-4:2021(E)
4) Construct a quantile-quantile (q-q) plot (as shown in Annex A) and compute the Anderson-Darling
(AD) statistic. Examine this plot in conjunction with the data for variation sufficiency. Proper
application of the control charts in this document require at least 6 unique values to provide
sufficient observable common cause variation. Insufficient variation in the data set will manifest
into the following two outcomes:
a) q-q plot shows several distinct horizontal data clusters where each horizontal cluster represent
numerically identical data (see Clause 5),
b) AD value exceeds the 0,01 sig. level of 1,0 by a large margin.
If the total number of unique values is less than six, proceed directly to Clause 5 for guidance.
5) Perform a formal statistical assessment for outliers using generalized extreme studentized
[5]
deviation (GESD) technique for outliers similar to ISO 4259-1 (or refer to ASTM D7915 ). The
recommended maximum number of outliers for 20 to 25 observations is 3 at 0,01 significance.
Reject identified outliers, obtain replacement results and repeat from step 4).
While not always possible, it is recommended that rejection of outliers be justified by corresponding
root cause(s).
6) Confirm the goodness-of-fit of the normal distribution using the Anderson-Darling (AD) statistic
computed from at least 20 non-outlier results.
— If AD is less than 1,0, continue to step 7);
— if AD is between 1,0 and 1,5, proceed to Clause 5 for guidance;
— if AD is greater than 1,5, do not proceed with this document as this is strong statistical evidence
that either the system is not in-statistical-control, or, the process results distribution is severely
non-normal.
7) Compute the average (x̅ ) and standard deviation (s ) statistics using the non-rejected
stage1 stage1
results from step 5).
8) Assess if additional data from historic results for this test method can be used to improve the
estimate of s , this should be done using an F-test at the 0,025 significance level with the
known
numerically larger number in the numerator. The reproducibility statement function determines
how this should be done:
a) For methods with a constant published reproducibility: use the F-test assess if s is
stage1
statistically indistinguishable versus s .
known
b) For methods with a non-constant published reproducibility: calculate the reproducibility at
the x̅ level and the reproducibility at the mean of the previous chart using x̅ .
stage1 known
NOTE Annex A shows a worked example including the F-test to pool the sigmas.
If the ratio of Reproducibility_ x̅ / Reproducibility_ x̅ is between 0,85 and 1,15 then
stage1 known
use the F-test to assess if s is statistically indistinguishable versus s .
stage1 known
If the F-Test to pool s and s passes, pool both standard deviations (follow the procedure
stage1 known
in Annex B). Assign the pooled standard deviation s as the standard deviation s to be used
pool chart
to construct the limits in step 9).
If there is no prior information on s for this type of material, or, if the F-test fails then use s
known stage1
as s for step 9).
chart
9) Create the I-Chart for this batch of QC samples by assigning x̅ from step 7) as x̅ for this
stage1 chart
step; then, overlaying the centre line represented by x̅ , and the two control limits represented
chart
by x̅ ± 3·s onto the run chart in step 3). If the EWMA sensitivity enhancement strategy is
chart chart
ISO 4259-4:2021(E)
used, construct and plot the EWMA line with its associated control limits placed at x̅ ± 1,5·s
chart chart
onto the run chart as well.
10) Label the zones in the I-chart as follows:
— Zone C: x̅ ± 1·s (exclusive)
chart chart
— Zone B: x̅ + 1·s (inclusive) to 2·s (exclusive); x̅ − 1·s (inclusive) to − 2·s
chart chart chart chart chart chart
(exclusive).
— Zone A: x̅ + 2·s (inclusive) to 3·s (exclusive); x̅ − 2·s (inclusive) to − 3·s
chart chart chart chart chart chart
(exclusive).
11) Compute the MR results using all non-rejected results from Step 5).
12) Compute the average using all MR results and designate this as MR .
stage1
13) If s is used from step 8), use the weighted average MR computed from MR
pool wtd known
corresponding to the s and MR as the MR to be used to construct the MR-chart in
known stage1 chart
step 14). Otherwise, use MR as MR for this batch of QC material.
stage1 chart
14) Create the MR-chart by overlaying the lines represented by MR and upper control limit at 3,27
chart
· MR onto a run chart plot of all the MR values computed in step 11).
chart
15) If all of the in-statistical-control conditions (see 4.2.4) are met, proceed to Stage 2. Otherwise,
investigate and mitigate root causes for failure, then repeat from step 1).
NOTE See Annex A and B for a detailed illustration of Stage 1 steps 3) to 15) and related statistical
techniques.
ISO 4259-4:2021(E)
Figure 2 — Flow chart of Stage 1 of control chart work process
ISO 4259-4:2021(E)
4.3.3 Stage 2 of control chart work process
4.3.3.1 Operation
In the operation mode, the primary objective is to detect abnormal events by:
— the immediate evaluation of a new QC sample result and its corresponding MR result against the
action limits of the I-chart and MR-chart respectively;
— the immediate evaluation of whether any action is required from the chosen sensitivity enhancement
Strategy (1 or 2).
The interpretation and immediate response associated with violation of the chart action limits, or,
'action required' outcome from enhancement Strategy 1 or 2, are listed below:
— Violation of I-chart limit: single result at or outside x̅ ± 3 s
chart chart
— Interpretation: a unique, sudden, event has occurred that caused the test result to deviate from
the centre line by an amount that is not plausible due to common causes.
— Immediate response: re-analyse a new QC sample to confirm unique event.
— If violation still persist, declare the test process is out of statistical control. The test process
should be taken out of service until the problem is investigated and resolved.
— If re-analysis result is back in zones C or B as specified above [see 4.3.2 step 10)], the out of
statistical control situation is not confirmed. Do not declare the process out of statistical
control. However, ensure this event is duly recorded.
— For the purpose of Stage 2 Annex B - Maintenance:
— If the system is declared out of statistical control, exclude both the initial and re-analysis result.
— If out of statistical control is not confirmed, exclude the initial out of control result and use
only the re-analysis result if the initial out of control result exceeds the control limit by an
amount > 0,25 s and there is no MR upper control limit exceedance associate with either
chart
the original or the retest result. This is for the purpose of minimising the rejection of legitimate
data. Otherwise, exclude the re-analysis result and use only the initial out of control result.
In the case of redefining control limits, care shall be taken to ensure that initial points that
exceed the control limit by ≤ 0,25 s be included in the performance re-evaluation and the
chart
re-analysis points rejected. This approach ensures that true system performance is evaluated
and not just points that were within limits leading to ever decreasing sigma values with no
identification of system deterioration.
— Violation of MR-chart upper limit: single MR outside its upper action limit:
— Interpretation: an unusually large difference has occurred between the current result and the
previous result; this event is usually (but not necessarily) associated with an I-chart violation;
— Immediate response: re-run a new QC sample if this is not associated with an I-chart violation.
Investigate for possible causes of large step change:
— If it is determined that this is not due to QC sample integrity, or, if this is not caused by
an I-chart violation, due to the higher false alarm rate of this statistic, 5 or more of this
event over the past 12 is reasonable statistical evidence that the variation (precision) of
the process has deteriorated. Perform an on-demand precision (variance) comparison
using the F-test (similar to precision comparison in maintenance mode and illustrated in
Annex B) between the most recent 20 in-statistical-control results versus the current chart
variance (s ) .
chart
ISO 4259-4:2021(E)
”Action required” from I-chart sensitivity enhancement strategies (either 1 or 2)
— Interpretation: it is highly possible that the current process centre has moved away from the
expected average (x̅ ) established from stage 1.
chart
— Immediate response: confirm if this violation is instrument or QC sample related by testing
CRM’s or stable retains of PT or previously tested production samples that are known to have
been properly stored in a manner that preserves the property of interest during storage.
Compare the result obtained to the value expected. For certified reference materials (CRM) or
proficiency testing (PT) samples, use the assigned reference value (ARV) as the value expected.
For retain of previously tested production sample, use the previous test result.
— If the difference between the test result obtained and the value expected is greater than
1,5·s for CRM or PT retain, or greater than 2·s for previously tested production
chart chart
sample, conclude that this “action required” signal is likely not due to QC sample integrity.
Declare the test process is out of statistical control. The test process should be taken out of
service until the problem is investigated and resolved. CRM/PT with ARV within working
range of s is preferred but not mandatory.
chart
— If the difference between the test result obtained versus the value expected for CRM or
PT retain is ≤ 1,5·s , or ≤ 2·s for previously tested production sample, conclude that
chart chart
the QC sample integrity is suspect. Switch to a new QC supply per 4.4 on QC material batch
change.
4.3.3.2 Maintenance
4.3.3.2.1 General
In the maintenance mode, the primary objective is to reduce the uncertainties of the statistics used
for I-chart parameters (centre, control limits) through periodic statistical assessment of accrued in-
statistical-control data. Two scenarios are described in 4.3.3.2.2 and 4.3.3.2.3.
4.3.3.2.2 Scenario 1
A minimum of 20 new in-statistical-control QC results that are not used to compute the control chart
limits has been accrued while the control chart is in Operation mode (i.e.: an active control chart) for
the same batch of control sample.
Under this scenario, the current control chart is actively being populated by results from the same batch
of QC material. Condition on the assumption that both the test process and this QC material remain
unchanged, the new accrued in-statistical-control data represents a ‘repeated statistical sampling’
of the same target population of data. Therefore, the control chart centre and control limits may be
recalculated and updated on a go-forward basis using more reliable statistics computed using a larger
data set from the same target data population by combining the new accrued in control data with the
data used to update the current control limits. Combining new accrued data in the aforementioned
manner is condition upon no compelling evidence that invalidates the “remain unchanged” assumption,
which can be supported by 'not statistically significant' outcomes from the following statistical tests at
the 0,05 significance level:
First, perform the F-test of variance of new accrued in-statistical-control data versus the variance used
to compute the current control limits. If F-test is not significant
...