ISO 16269-6:2005
(Main)Statistical interpretation of data — Part 6: Determination of statistical tolerance intervals
Statistical interpretation of data — Part 6: Determination of statistical tolerance intervals
ISO 16269-6:2005 describes procedures for establishing tolerance intervals that include at least a specified proportion of the population with a specified confidence level. Both one-sided and two-sided statistical tolerance intervals are provided, a one-sided interval having either an upper or a lower limit while a two-sided interval has both upper and lower limits. Two methods are provided, a parametric method for the case where the characteristic being studied has a normal distribution and a distribution-free method for the case where nothing is known about the distribution except that it is continuous.
Interprétation statistique des données — Partie 6: Détermination des intervalles statistiques de tolérance
L'ISO 16269-6:2005 décrit des méthodes permettant d'établir les intervalles statistiques de tolérance qui comprennent au moins une proportion spécifiée de la population avec un niveau de confiance spécifié. Des intervalles statistiques de tolérance unilatéraux et bilatéraux sont fournis, l'intervalle statistique de tolérance unilatéral étant caractérisé par une limite supérieure ou par une limite inférieure, tandis que l'intervalle statistique bilatéral possède à la fois une limite supérieure et une limite inférieure. Deux méthodes sont exposées: une méthode paramétrique, lorsque la caractéristique étudiée a une distribution normale, et une méthode non paramétrique, lorsque rien n'est connu de la distribution si ce n'est qu'elle est continue.
Statistično tolmačenje podatkov – 6. del: Ugotavljanje statističnih tolerančnih intervalov
General Information
Relations
Buy Standard
Standards Content (Sample)
INTERNATIONAL ISO
STANDARD 16269-6
First edition
2005-04-01
Statistical interpretation of data —
Part 6:
Determination of statistical tolerance
intervals
Interprétation statistique des données —
Partie 6: Détermination des intervalles statistiques de tolérance
Reference number
ISO 16269-6:2005(E)
©
ISO 2005
---------------------- Page: 1 ----------------------
ISO 16269-6:2005(E)
PDF disclaimer
This PDF file may contain embedded typefaces. In accordance with Adobe's licensing policy, this file may be printed or viewed but
shall not be edited unless the typefaces which are embedded are licensed to and installed on the computer performing the editing. In
downloading this file, parties accept therein the responsibility of not infringing Adobe's licensing policy. The ISO Central Secretariat
accepts no liability in this area.
Adobe is a trademark of Adobe Systems Incorporated.
Details of the software products used to create this PDF file can be found in the General Info relative to the file; the PDF-creation
parameters were optimized for printing. Every care has been taken to ensure that the file is suitable for use by ISO member bodies. In
the unlikely event that a problem relating to it is found, please inform the Central Secretariat at the address given below.
© ISO 2005
All rights reserved. Unless otherwise specified, no part of this publication may be reproduced or utilized in any form or by any means,
electronic or mechanical, including photocopying and microfilm, without permission in writing from either ISO at the address below or
ISO's member body in the country of the requester.
ISO copyright office
Case postale 56 • CH-1211 Geneva 20
Tel. + 41 22 749 01 11
Fax + 41 22 749 09 47
E-mail copyright@iso.org
Web www.iso.org
Published in Switzerland
ii © ISO 2005 – All rights reserved
---------------------- Page: 2 ----------------------
ISO 16269-6:2005(E)
Contents Page
Foreword. iv
Introduction . v
1 Scope. 1
2 Normative references . 1
3 Terms, definitions and symbols . 1
3.1 Terms and definitions. 1
3.2 Symbols . 2
4 Procedures . 3
4.1 Normal population with known variance and known mean. 3
4.2 Normal population with known variance and unknown mean . 3
4.3 Normal population with unknown variance and unknown mean. 3
4.4 Any continuous distribution of unknown type . 3
5 Examples. 3
5.1 Data. 3
5.2 Example 1: One-sided statistical tolerance interval under known variance. 4
5.3 Example 2: Two-sided statistical tolerance interval under known variance . 4
5.4 Example 3: One-sided statistical tolerance interval under unknown variance . 5
5.5 Example 4: Two-sided statistical tolerance interval under unknown variance. 6
5.6 Example 5: Distribution-free statistical tolerance interval for continuous distribution . 6
Annex A (informative) Forms for tolerance intervals. 8
Annex B (normative) One-sided statistical tolerance limit factors, k (n; p; 1 − α), for known σ . 14
1
Annex C (normative) Two-sided statistical tolerance limit factors, k (n; p; 1 − α), for known σ. 17
2
Annex D (normative) One-sided statistical tolerance limit factors, k (n; p; 1 − α), for unknown σ. 20
3
Annex E (normative) Two-sided statistical tolerance limit factors, k (n; p; 1 − α), for unknown σ. 23
4
Annex F (normative) One-sided distribution-free statistical tolerance intervals. 26
Annex G (normative) Two-sided distribution-free statistical tolerance intervals. 27
Annex H (informative) Construction of a distribution-free statistical tolerance interval for any type
of distribution . 28
Annex I (informative) Computation of factors for two-sided parametric statistical tolerance
intervals. 29
Bibliography . 30
© ISO 2005 – All rights reserved iii
---------------------- Page: 3 ----------------------
ISO 16269-6:2005(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.
International Standards are drafted in accordance with the rules given in the ISO/IEC Directives, Part 2.
The main task of technical committees is to prepare International Standards. Draft International Standards
adopted by the technical committees are circulated to the member bodies for voting. Publication as an
International Standard requires approval by at least 75 % of the member bodies casting a vote.
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.
ISO 16269-6 was prepared by Technical Committee ISO/TC 69, Applications of statistical methods.
This first edition of ISO 16269-6 cancels and replaces ISO 3207:1975, which has been technically revised.
ISO 16269 consists of the following parts, under the general title Statistical interpretation of data:
Part 6: Determination of statistical tolerance intervals
Part 7: Median — Estimation and confidence intervals
Part 8: Determination of prediction intervals
iv © ISO 2005 – All rights reserved
---------------------- Page: 4 ----------------------
ISO 16269-6:2005(E)
Introduction
A statistical tolerance interval is an estimated interval, based on a sample, which can be asserted with
confidence 1 − α, for example 95 %, to contain at least a specified proportion p of the items in the population.
The limits of a statistical tolerance interval are called statistical tolerance limits. The confidence level 1 − α is
the probability that a statistical tolerance interval constructed in the prescribed manner will contain at least a
proportion p of the population. Conversely, the probability that this interval will contain less than the proportion
p of the population is α. This part of ISO 16269 describes both one-sided and two-sided statistical tolerance
intervals; a one-sided interval is constructed with an upper or a lower limit while a two-sided interval is
constructed with both an upper and a lower limit.
Tolerance intervals are functions of the observations of the sample, i.e. statistics, and they will generally take
different values for different samples. It is necessary that the observations be independent for the procedures
provided in this part of ISO 16269 to be valid.
Two types of tolerance interval are provided in this part of ISO 16269, parametric and distribution-free. The
parametric approach is based on the assumption that the characteristic being studied in the population has a
normal distribution; hence the confidence that the calculated statistical tolerance interval contains at least a
proportion p of the population can only be taken to be 1 − α if the normality assumption is true. For normally
distributed characteristics, the statistical tolerance interval is determined using one of the Forms A, B, C or D
given in Annex A.
Parametric methods for distributions other than the normal are not considered in this part of ISO 16269. If
departure from normality is suspected in the population, distribution-free statistical tolerance intervals may be
constructed. The procedure for the determination of a statistical tolerance interval for any continuous
distribution is provided in Forms E and F of Annex A.
The tolerance limits discussed in this part of ISO 16269 can be used to compare the natural capability of a
process with one or two given specification limits, either an upper one U or a lower one L or both in statistical
process management. An indication of this is the fact that these tolerance limits have also been called natural
process limits. See ISO 3534-2:1993, 3.2.4, and the general remarks in ISO 3207 which will be cancelled and
replaced by this part of ISO 16269.
Above the upper specification limit U there is the upper fraction nonconforming p (ISO 3534-2:—, 3.2.5.5 and
U
3.3.1.4) and below the lower specification limit L there is the lower fraction nonconforming p (ISO 3534-2:—,
L
3.2.5.6 and 3.3.1.5). The sum p + p = p is called the total fraction nonconforming. (ISO 3534-2:—, 3.2.5.7).
U L T
Between the specification limits U and L there is the fraction conforming 1 − p .
T
In statistical process management the limits U and L are fixed in advance and the fractions p , p and p are
U L T
either calculated, if the distribution is assumed to be known, or otherwise estimated. There are many
applications of statistical tolerance intervals, although the above shows an example to a quality control
problem. Wider applications and more statistical intervals are introduced in many textbooks such as Hahn and
[10]
Meeker .
In contrast, for the tolerance intervals considered in this part of ISO 16269, the confidence level for the interval
estimator and the proportion of the distribution within the interval (corresponding to the fraction conforming
mentioned above) are fixed in advance, and the limits are estimated. These limits may be compared with U
and L. Hence the appropriateness of the given specification limits U and L can be compared with the actual
properties of the process. The one-sided tolerance intervals are used when only either the upper specification
limit U or the lower specification limit L is relevant, while the two-sided intervals are used when both the upper
and the lower specification limits are considered simultaneously.
The terminology with regard to these different limits and intervals has been confusing as the “specification
limits” were earlier also called “tolerance limits” (see the terminology standard ISO 3534-2:1993, 1.4.3, where
both these terms as well as the term “limiting values” were all used as synonyms for this concept). In the latest
© ISO 2005 – All rights reserved v
---------------------- Page: 5 ----------------------
ISO 16269-6:2005(E)
revision of ISO 3534-2:—, only the term specification limits have been kept for this concept. Furthermore, the
[5]
Guide for the expression of uncertainty in measurement uses the term “coverage factor” defined as a
“numerical factor used as a multiplier of the combined standard uncertainty in order to obtain an expanded
uncertainty”. This use of “coverage” differs from the use of the term in this part of ISO 16269.
vi © ISO 2005 – All rights reserved
---------------------- Page: 6 ----------------------
INTERNATIONAL STANDARD ISO 16269-6:2005(E)
Statistical interpretation of data —
Part 6:
Determination of statistical tolerance intervals
1 Scope
This part of ISO 16269 describes procedures for establishing tolerance intervals that include at least a
specified proportion of the population with a specified confidence level. Both one-sided and two-sided
statistical tolerance intervals are provided, a one-sided interval having either an upper or a lower limit while a
two-sided interval has both upper and lower limits. Two methods are provided, a parametric method for the
case where the characteristic being studied has a normal distribution and a distribution-free method for the
case where nothing is known about the distribution except that it is continuous.
2 Normative references
The following referenced documents are indispensable for the application 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 3534-1, Statistics — Vocabulary and symbols — Part 1: Probability and general statistical terms
1)
ISO 3534-2:— , Statistics — Vocabulary and symbols — Part 2: Applied statistics
3 Terms, definitions and symbols
3.1 Terms and definitions
For the purposes of this document, the terms and definition given in ISO 3534-1, ISO 3534-2 and the following
apply.
3.1.1
statistical tolerance interval
interval determined from a random sample in such a way that one may have a specified level of confidence
that the interval covers at least a specified proportion of the sampled population
NOTE The confidence level in this context is the long-run proportion of intervals constructed in this manner that will
include at least the specified proportion of the sampled population.
3.1.2
statistical tolerance limit
statistic representing an end point of a statistical tolerance interval
NOTE Statistical tolerance intervals can be either one-sided, in which case they have either an upper or a lower
statistical tolerance limit, or two-sided, in which case they have both.
1) To be published. (Revision of ISO 3534-2:1993)
© ISO 2005 – All rights reserved 1
---------------------- Page: 7 ----------------------
ISO 16269-6:2005(E)
3.1.3
coverage
proportion of items in a population lying within a statistical tolerance interval
NOTE This concept is not to be confused with the concept coverage factor used in the Guide for the expression of
[5]
uncertainty in measurement (GUM ) .
3.1.4
normal population
normally distributed population
3.2 Symbols
For the purposes of this part of ISO 16269, the following symbols apply.
i suffix of an observation
k (n; p; 1 − α) factor used to determine x or x when the value of σ is known for one-sided tolerance
L U
1
interval
k (n; p; 1 − α) factor used to determine x and x when the value of σ is known for two-sided tolerance
L U
2
interval
k (n; p; 1 − α) factor used to determine x or x when the value of σ is unknown for one-sided tolerance
L U
3
interval
k (n; p; 1 − α) factor used to determine x and x when the value of σ is unknown for two-sided tolerance
L U
4
interval
n number of observations in the sample
p minimum proportion of the population claimed to be lying in the statistical tolerance interval
u p-fractile of the standard normal distribution
p
x ith observed value (1in=,2,.,)
i
x maximum value of the observed values: x = max {x , x , …, x }
max max 1 2 n
x minimum value of the observed values: x = min {x , x , …, x }
min min 1 2 n
x lower limit of the statistical tolerance interval
L
x upper limit of the statistical tolerance interval
U
n
1
x sample mean, xx=
i
∑
n
i = 1
2
nn
2
nx − x
ii
∑∑
n
ii==11
1 2
s sample standard deviation;sx=−x=
()
i
∑
nn−−11n
()
i = 1
1 − α confidence level for the claim that the proportion of the population lying within the tolerance
interval is greater than or equal to the specified level p
µ population mean
σ population standard deviation
2 © ISO 2005 – All rights reserved
---------------------- Page: 8 ----------------------
ISO 16269-6:2005(E)
4 Procedures
4.1 Normal population with known variance and known mean
2
When the values of the mean, µ, and the variance, σ , of a normally distributed population are known, the
distribution of the characteristic under investigation is fully determined. There is exactly a proportion p of the
population:
a) to the right of x = µ − u × σ (one-sided interval);
p
L
b) to the left of x = µ + u × σ (one-sided interval);
p
U
u u
c) between x = µ − × σ and x = µ + × σ (two-sided interval).
(1+ p)/ 2 (1+ p)/ 2
L U
NOTE As such statements are known to be true, they are made with 100 % confidence.
In the above equations, u is p-fractile of the standard normal distribution. Numerical values of u may be
p p
read from the bottom line of the Tables B.1 to B.6 and Tables C.1 to C.6.
4.2 Normal population with known variance and unknown mean
Forms A and B, given in Annex A, are applicable to the case where the variance of the normal population is
known while the mean is unknown. Form A applies to the one-sided case, while Form B applies to the
two-sided case.
4.3 Normal population with unknown variance and unknown mean
Forms C and D, given in Annex A, are applicable to the case where both the mean and the variance of the
normal population are unknown. Form C applies to the one-sided case, while Form D applies to the two-sided
case.
4.4 Any continuous distribution of unknown type
If the characteristic under investigation is a continuous variable from a population of unknown form, and if a
sample of n independent random observations of the characteristic has been taken, then a statistical tolerance
interval can be determined from the ranked observations. The procedure given in Forms E and F of Annex A
provide the determination of the coverage or sample size needed for tolerance intervals determined from the
extreme values x or x of the sample of observations with given confidence level 1 − α.
min max
NOTE Statistical tolerance intervals that do not depend on the shape of the sampled population are called
distribution-free tolerance intervals.
This part of ISO 16269 does not provide procedures for distributions of known type other than the normal
distribution. However, if the distribution is continuous, the distribution-free method may be used. Selected
references to scientific literature that may assist in determining tolerance intervals for other distributions are
also provided at the end of this document.
5 Examples
5.1 Data
Forms A to D, given in Annex A, are illustrated by examples using the numerical values of ISO 2854:1976,
Clause 2, paragraph 1 of the introductory remarks, Table X, yarn 2: 12 measures of the breaking load of
cotton yarn. It should be noted that the number of observations, n = 12, given here for these examples is
[1]
considerably lower than the one recommended in ISO 2602 . The numerical data and calculations in the
different examples are expressed in centi-newtons (see Table 1).
© ISO 2005 – All rights reserved 3
---------------------- Page: 9 ----------------------
ISO 16269-6:2005(E)
Table 1 — Data for Examples 1 to 4
Values in centi-newtons
228,6 232,7 238,8 317,2 315,8 275,1 222,2 236,7 224,7 251,2 210,4 270,7
x
These measurements were obtained from a batch of 12,000 bobbins, from one production job, packed in
120 boxes each containing 100 bobbins. Twelve boxes have been drawn at random from the batch and a
bobbin has been drawn at random from each of these boxes. Test pieces of 50 cm length have been cut from
the yarn on these bobbins, at about 5 m distance from the free end. The tests themselves have been carried
out on the central parts of these test pieces. Previous information makes it reasonable to assume that the
breaking loads measured in these conditions have virtually a normal distribution. It is demonstrated in
ISO 2954:1976 that the data do not contradict the assumption of a normal distribution.
These results yield the following:
Sample size: n = 12
Sample mean: x==3 024,1/12 252,01
2
2
nx − x
()
∑∑ 166 772,27
Sample standard deviation:
s== = 1263,426 3= 35,545
nn−×11211
()
The formal presentation of the calculations will be given only for Form C in Annex A (one-sided interval,
unknown variance).
5.2 Example 1: One-sided statistical tolerance interval under known variance
Suppose that previously obtained measurements have shown that the dispersion is constant from one batch
σ = 33,150
to another from the same supplier, and is represented by a standard deviation , although the mean
is not constant. A limit x is required such that it is possible to assert with confidence level 1 − α = 0,95 (95 %)
L
that at least 0,95 (95 %) of the breaking loads of the items in the batch, when measured under the same
conditions, are above x .
L
Table B.4 gives
k (12; 0,95; 0,95) = 2,120
1
whence
x = xk−−(n;p;1 ασ)×= 252,01− 2,120× 33,150= 181,732
L 1
A smaller value of the lower limit x would be obtained if a larger proportion of the population (for example
L
p = 0,99) and/or a higher confidence level (for example 1 − α = 0,99) were required.
5.3 Example 2: Two-sided statistical tolerance interval under known variance
Under the same conditions as in Example 1, suppose that limits x and x are required such that it is
L U
possible to assert with a confidence level 1 − α = 0,95 that at least a proportion of p = 0,90 (90 %) of the
breaking load of the batch falls between x and x .
L U
Table C.4 gives
k (12; 0,90; 0,95) = 1,889
2
4 © ISO 2005 – All rights reserved
---------------------- Page: 10 ----------------------
ISO 16269-6:2005(E)
whence
x = xk−−(n;p;1 ασ)×= 252,01−1,889× 33,150= 189,390
L 2
x = xk+−(n;p;1 ασ)×= 252,01+1,889× 33,150= 314,630
U 2
Comparison with Example 1 should make it clear that assuring that at least 90 % of a population lies between
the limits x and x is not the same thing as assuring that no more than 5 % lies beyond each limit.
L U
5.4 Example 3: One-sided statistical tolerance interval under unknown variance
Here, it is supposed that the standard deviation of the population is unknown and has to be estimated from the
sample. The same requirements will be assumed as for the case where the standard deviation is known
(Example 1), thus, p = 0,95 and 1 − α = 0,95. The presentation of the results is given in detail below.
Determination of the statistical tolerance interval of proportion p:
a) one-sided interval “to the right”
Determined values:
b) proportion of the population selected for the tolerance interval: p = 0,95
c) chosen confidence level: 1 − α = 0,95
d) sample size: n = 12
Value of tolerance factor from Table D.4:
k (n; p; 1 − α) = 2,737
3
Calculations:
xx==/n 252,01
∑
2
2
nx − x
()
∑∑
s = = 35,545
nn −1
()
kn( ;p;1−×α) s= 97,286 7
3
Results: one-sided interval “to the right”
The tolerance interval which will contain at least a proportion p of the population with confidence level 1 − α
has a lower limit
x =−xkn( ;p;1− α)×s= 154,723
L 3
© ISO 2005 – All rights reserved 5
---------------------- Page: 11 ----------------------
ISO 16269-6:2005(E)
5.5 Example 4: Two-sided statistical tolerance interval under unknown variance
Under the same conditions as in Example 2, suppose it is required to calculate the limits x and x such
L U
that it is possible to assert with a confidence level 1 − α = 0,95 that in a proportion of the batch at least equal
to p = 0,90 (90 %) the breaking load falls between x and x .
L U
Table E.4 gives
kn(;p;1−=α) 2,671
4
whence
xx=−k (n;p;1−α)×s= 252,01− 2,671× 35,545= 157,069
L 4
xx=+k (n;p;1−α)×s= 252,01+ 2,671× 35,545= 346,951
U 4
It will be noted that the value of x is smaller and the value of x higher than in Example 2 (known variance),
L U
because the use of s instead of σ requires a larger value of the tolerance factor to allow for the extra
uncertainty. It is necessary to have to pay a penalty for not knowing the population standard deviation σ and
the extension of the statistical tolerance interval takes this into account. Of course, it is not quite sure that the
value σ = 33,150 used in Examples 1 and 2 is correct. Therefore, it is wiser to use the estimate, s, together
with Tables D.4 or E.4.
5.6 Example 5: Distribution-free statistical tolerance interval for continuous distribution
In a fatigue test by rotational stress carried out on a component of an aeronautical engine, a sample of
15 items has given the results (measurement of endurance), shown in ascending order of values in Table 2.
Table 2 — Data for Example 5
x 0,200 0,330 0,450 0,490 0,780 0,920 0,950 0,970 1,040 1,710 2,220 2,275 3,650 7,000 8,800
A graphical examination of checking normality, such as probability plot, shows that the hypothesis of normality
for the population of components should almost certainly be rejected (see ISO 5479). The methods of Form E,
given in Annex A, for determination of a statistical tolerance interval are therefore applicable.
The extreme values from the sample of n = 15 measurements are:
x = 0,200, x = 8,800
min max
Suppose that the required confidence level 1 − α is 0,95.
a) What is the maximum proportion of the population of components that will fall below x = 0,200?
min
Table F.1, for 1 − α = 0,95, gives for the minimum proportion above x a value of p slightly higher than
min
0,75 (75 %). Hence, for the maximum proportion below x a value of 1 − p slightly lower than
min
0,25 (25 %).
b) What sample size is necessary for it to be possible to assert, at a confidence level 0,95, that a proportion
at least p = 0,90 (90 %) of the population of components will be found below the largest of the values from
that sample? Table F.1, for 1 − α = 0,95 and p = 0,90, gives n = 29.
6 © ISO 2005 – All rights reserved
---------------------- Page: 12 ----------------------
ISO 16269-6:2005(E)
c) At a confidence level of 0,95, what is the minimum proportion of the population of components that fall
between x = 0,200 and x = 8,800? Table G.1, for 1 − α = 0,95 and n = 15, gives p slightly below
min max
0,75 (75 %).
d) What sample size is necessary for it to be possible to assert at a confidence level 0,95 that a proportion
of at least p = 0,90 (90 %) of the population of components will be found to fall between the smallest and
the largest values from that sample? Table G.1, for 1 − α = 0,95 and p = 0,90, gives n = 46.
e) In general, if a check for normality (see ISO 5479) indicates a departure from the normal distribution,
some transformation will be recommended based on the knowledge of the collected data. For example,
fatigue data are often approximated lognormally distributed. In such cases, the data could be transformed
to normality. Tolerance intervals are then calculated and finally transformed back into the original units.
See Annex H for the construction of a statistical tolerance interval for distribution-free tolerance intervals for
any type of distribution. Annex I gives the computation of factors for two-sided parametric statistical tolerance
intervals.
© ISO 2005 – All rights reserved 7
---------------------- Page: 13 ----------------------
ISO 16269-6:2005(E)
Annex A
(informative)
Forms for tolerance intervals
Form A — One-sided statistical tolerance interval (known variance)
Determination of a one-sided statistical tolerance interval with coverage p at confidence level 1 − α
a) One-sided interval “to the left”
b) One-sided interval “to the right”
Known values:
2
c) the variance: σ =
d) the standard deviation: σ =
Determined values:
e) proportion of the population selected for the tolerance interval: p =
f) chosen confidence level: 1 − α =
g) sample size: n =
Tabulated factor:
k (n; p; 1 − α) =
1
This value can be read from the tables given in Annex B for a range of values of n, p and 1 − α.
Calculations:
xx==/n
∑
k (n; p; 1 − α) × σ =
1
Results:
a) One-sided interval “to the left”
The one-sided statistical tolerance interval with coverage p at confidence level 1 − α has upper limit
xx=+k (;np;1− α)×σ=
U 1
b) One-sided interval “to the right”
The one-sided statistical tolerance interval with coverage p at confidence level 1 − α has lower limit
x =−xkn(;p;1− α)×σ=
L 1
8 © ISO 2005 – All rights reserved
---------------------- Page: 14 ----------------------
ISO 16269-6:2005(E)
Form B — Two-sided statistical tolerance interval (known variance)
Determination of a two-sided statistical toleranc
...
SLOVENSKI STANDARD
SIST ISO 16269-6:2006
01-april-2006
6WDWLVWLþQRWROPDþHQMHSRGDWNRY±GHO 8JRWDYOMDQMHVWDWLVWLþQLKWROHUDQþQLK
LQWHUYDORY
Statistical interpretation of data -- Part 6: Determination of statistical tolerance intervals
Interprétation statistique des données -- Partie 6: Détermination des intervalles
statistiques de tolérance
Ta slovenski standard je istoveten z: ISO 16269-6:2005
ICS:
03.120.30 8SRUDEDVWDWLVWLþQLKPHWRG Application of statistical
methods
SIST ISO 16269-6:2006 en
2003-01.Slovenski inštitut za standardizacijo. Razmnoževanje celote ali delov tega standarda ni dovoljeno.
---------------------- Page: 1 ----------------------
SIST ISO 16269-6:2006
---------------------- Page: 2 ----------------------
SIST ISO 16269-6:2006
INTERNATIONAL ISO
STANDARD 16269-6
First edition
2005-04-01
Statistical interpretation of data —
Part 6:
Determination of statistical tolerance
intervals
Interprétation statistique des données —
Partie 6: Détermination des intervalles statistiques de tolérance
Reference number
ISO 16269-6:2005(E)
©
ISO 2005
---------------------- Page: 3 ----------------------
SIST ISO 16269-6:2006
ISO 16269-6:2005(E)
PDF disclaimer
This PDF file may contain embedded typefaces. In accordance with Adobe's licensing policy, this file may be printed or viewed but
shall not be edited unless the typefaces which are embedded are licensed to and installed on the computer performing the editing. In
downloading this file, parties accept therein the responsibility of not infringing Adobe's licensing policy. The ISO Central Secretariat
accepts no liability in this area.
Adobe is a trademark of Adobe Systems Incorporated.
Details of the software products used to create this PDF file can be found in the General Info relative to the file; the PDF-creation
parameters were optimized for printing. Every care has been taken to ensure that the file is suitable for use by ISO member bodies. In
the unlikely event that a problem relating to it is found, please inform the Central Secretariat at the address given below.
© ISO 2005
All rights reserved. Unless otherwise specified, no part of this publication may be reproduced or utilized in any form or by any means,
electronic or mechanical, including photocopying and microfilm, without permission in writing from either ISO at the address below or
ISO's member body in the country of the requester.
ISO copyright office
Case postale 56 • CH-1211 Geneva 20
Tel. + 41 22 749 01 11
Fax + 41 22 749 09 47
E-mail copyright@iso.org
Web www.iso.org
Published in Switzerland
ii © ISO 2005 – All rights reserved
---------------------- Page: 4 ----------------------
SIST ISO 16269-6:2006
ISO 16269-6:2005(E)
Contents Page
Foreword. iv
Introduction . v
1 Scope. 1
2 Normative references . 1
3 Terms, definitions and symbols . 1
3.1 Terms and definitions. 1
3.2 Symbols . 2
4 Procedures . 3
4.1 Normal population with known variance and known mean. 3
4.2 Normal population with known variance and unknown mean . 3
4.3 Normal population with unknown variance and unknown mean. 3
4.4 Any continuous distribution of unknown type . 3
5 Examples. 3
5.1 Data. 3
5.2 Example 1: One-sided statistical tolerance interval under known variance. 4
5.3 Example 2: Two-sided statistical tolerance interval under known variance . 4
5.4 Example 3: One-sided statistical tolerance interval under unknown variance . 5
5.5 Example 4: Two-sided statistical tolerance interval under unknown variance. 6
5.6 Example 5: Distribution-free statistical tolerance interval for continuous distribution . 6
Annex A (informative) Forms for tolerance intervals. 8
Annex B (normative) One-sided statistical tolerance limit factors, k (n; p; 1 − α), for known σ . 14
1
Annex C (normative) Two-sided statistical tolerance limit factors, k (n; p; 1 − α), for known σ. 17
2
Annex D (normative) One-sided statistical tolerance limit factors, k (n; p; 1 − α), for unknown σ. 20
3
Annex E (normative) Two-sided statistical tolerance limit factors, k (n; p; 1 − α), for unknown σ. 23
4
Annex F (normative) One-sided distribution-free statistical tolerance intervals. 26
Annex G (normative) Two-sided distribution-free statistical tolerance intervals. 27
Annex H (informative) Construction of a distribution-free statistical tolerance interval for any type
of distribution . 28
Annex I (informative) Computation of factors for two-sided parametric statistical tolerance
intervals. 29
Bibliography . 30
© ISO 2005 – All rights reserved iii
---------------------- Page: 5 ----------------------
SIST ISO 16269-6:2006
ISO 16269-6:2005(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.
International Standards are drafted in accordance with the rules given in the ISO/IEC Directives, Part 2.
The main task of technical committees is to prepare International Standards. Draft International Standards
adopted by the technical committees are circulated to the member bodies for voting. Publication as an
International Standard requires approval by at least 75 % of the member bodies casting a vote.
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.
ISO 16269-6 was prepared by Technical Committee ISO/TC 69, Applications of statistical methods.
This first edition of ISO 16269-6 cancels and replaces ISO 3207:1975, which has been technically revised.
ISO 16269 consists of the following parts, under the general title Statistical interpretation of data:
Part 6: Determination of statistical tolerance intervals
Part 7: Median — Estimation and confidence intervals
Part 8: Determination of prediction intervals
iv © ISO 2005 – All rights reserved
---------------------- Page: 6 ----------------------
SIST ISO 16269-6:2006
ISO 16269-6:2005(E)
Introduction
A statistical tolerance interval is an estimated interval, based on a sample, which can be asserted with
confidence 1 − α, for example 95 %, to contain at least a specified proportion p of the items in the population.
The limits of a statistical tolerance interval are called statistical tolerance limits. The confidence level 1 − α is
the probability that a statistical tolerance interval constructed in the prescribed manner will contain at least a
proportion p of the population. Conversely, the probability that this interval will contain less than the proportion
p of the population is α. This part of ISO 16269 describes both one-sided and two-sided statistical tolerance
intervals; a one-sided interval is constructed with an upper or a lower limit while a two-sided interval is
constructed with both an upper and a lower limit.
Tolerance intervals are functions of the observations of the sample, i.e. statistics, and they will generally take
different values for different samples. It is necessary that the observations be independent for the procedures
provided in this part of ISO 16269 to be valid.
Two types of tolerance interval are provided in this part of ISO 16269, parametric and distribution-free. The
parametric approach is based on the assumption that the characteristic being studied in the population has a
normal distribution; hence the confidence that the calculated statistical tolerance interval contains at least a
proportion p of the population can only be taken to be 1 − α if the normality assumption is true. For normally
distributed characteristics, the statistical tolerance interval is determined using one of the Forms A, B, C or D
given in Annex A.
Parametric methods for distributions other than the normal are not considered in this part of ISO 16269. If
departure from normality is suspected in the population, distribution-free statistical tolerance intervals may be
constructed. The procedure for the determination of a statistical tolerance interval for any continuous
distribution is provided in Forms E and F of Annex A.
The tolerance limits discussed in this part of ISO 16269 can be used to compare the natural capability of a
process with one or two given specification limits, either an upper one U or a lower one L or both in statistical
process management. An indication of this is the fact that these tolerance limits have also been called natural
process limits. See ISO 3534-2:1993, 3.2.4, and the general remarks in ISO 3207 which will be cancelled and
replaced by this part of ISO 16269.
Above the upper specification limit U there is the upper fraction nonconforming p (ISO 3534-2:—, 3.2.5.5 and
U
3.3.1.4) and below the lower specification limit L there is the lower fraction nonconforming p (ISO 3534-2:—,
L
3.2.5.6 and 3.3.1.5). The sum p + p = p is called the total fraction nonconforming. (ISO 3534-2:—, 3.2.5.7).
U L T
Between the specification limits U and L there is the fraction conforming 1 − p .
T
In statistical process management the limits U and L are fixed in advance and the fractions p , p and p are
U L T
either calculated, if the distribution is assumed to be known, or otherwise estimated. There are many
applications of statistical tolerance intervals, although the above shows an example to a quality control
problem. Wider applications and more statistical intervals are introduced in many textbooks such as Hahn and
[10]
Meeker .
In contrast, for the tolerance intervals considered in this part of ISO 16269, the confidence level for the interval
estimator and the proportion of the distribution within the interval (corresponding to the fraction conforming
mentioned above) are fixed in advance, and the limits are estimated. These limits may be compared with U
and L. Hence the appropriateness of the given specification limits U and L can be compared with the actual
properties of the process. The one-sided tolerance intervals are used when only either the upper specification
limit U or the lower specification limit L is relevant, while the two-sided intervals are used when both the upper
and the lower specification limits are considered simultaneously.
The terminology with regard to these different limits and intervals has been confusing as the “specification
limits” were earlier also called “tolerance limits” (see the terminology standard ISO 3534-2:1993, 1.4.3, where
both these terms as well as the term “limiting values” were all used as synonyms for this concept). In the latest
© ISO 2005 – All rights reserved v
---------------------- Page: 7 ----------------------
SIST ISO 16269-6:2006
ISO 16269-6:2005(E)
revision of ISO 3534-2:—, only the term specification limits have been kept for this concept. Furthermore, the
[5]
Guide for the expression of uncertainty in measurement uses the term “coverage factor” defined as a
“numerical factor used as a multiplier of the combined standard uncertainty in order to obtain an expanded
uncertainty”. This use of “coverage” differs from the use of the term in this part of ISO 16269.
vi © ISO 2005 – All rights reserved
---------------------- Page: 8 ----------------------
SIST ISO 16269-6:2006
INTERNATIONAL STANDARD ISO 16269-6:2005(E)
Statistical interpretation of data —
Part 6:
Determination of statistical tolerance intervals
1 Scope
This part of ISO 16269 describes procedures for establishing tolerance intervals that include at least a
specified proportion of the population with a specified confidence level. Both one-sided and two-sided
statistical tolerance intervals are provided, a one-sided interval having either an upper or a lower limit while a
two-sided interval has both upper and lower limits. Two methods are provided, a parametric method for the
case where the characteristic being studied has a normal distribution and a distribution-free method for the
case where nothing is known about the distribution except that it is continuous.
2 Normative references
The following referenced documents are indispensable for the application 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 3534-1, Statistics — Vocabulary and symbols — Part 1: Probability and general statistical terms
1)
ISO 3534-2:— , Statistics — Vocabulary and symbols — Part 2: Applied statistics
3 Terms, definitions and symbols
3.1 Terms and definitions
For the purposes of this document, the terms and definition given in ISO 3534-1, ISO 3534-2 and the following
apply.
3.1.1
statistical tolerance interval
interval determined from a random sample in such a way that one may have a specified level of confidence
that the interval covers at least a specified proportion of the sampled population
NOTE The confidence level in this context is the long-run proportion of intervals constructed in this manner that will
include at least the specified proportion of the sampled population.
3.1.2
statistical tolerance limit
statistic representing an end point of a statistical tolerance interval
NOTE Statistical tolerance intervals can be either one-sided, in which case they have either an upper or a lower
statistical tolerance limit, or two-sided, in which case they have both.
1) To be published. (Revision of ISO 3534-2:1993)
© ISO 2005 – All rights reserved 1
---------------------- Page: 9 ----------------------
SIST ISO 16269-6:2006
ISO 16269-6:2005(E)
3.1.3
coverage
proportion of items in a population lying within a statistical tolerance interval
NOTE This concept is not to be confused with the concept coverage factor used in the Guide for the expression of
[5]
uncertainty in measurement (GUM ) .
3.1.4
normal population
normally distributed population
3.2 Symbols
For the purposes of this part of ISO 16269, the following symbols apply.
i suffix of an observation
k (n; p; 1 − α) factor used to determine x or x when the value of σ is known for one-sided tolerance
L U
1
interval
k (n; p; 1 − α) factor used to determine x and x when the value of σ is known for two-sided tolerance
L U
2
interval
k (n; p; 1 − α) factor used to determine x or x when the value of σ is unknown for one-sided tolerance
L U
3
interval
k (n; p; 1 − α) factor used to determine x and x when the value of σ is unknown for two-sided tolerance
L U
4
interval
n number of observations in the sample
p minimum proportion of the population claimed to be lying in the statistical tolerance interval
u p-fractile of the standard normal distribution
p
x ith observed value (1in=,2,.,)
i
x maximum value of the observed values: x = max {x , x , …, x }
max max 1 2 n
x minimum value of the observed values: x = min {x , x , …, x }
min min 1 2 n
x lower limit of the statistical tolerance interval
L
x upper limit of the statistical tolerance interval
U
n
1
x sample mean, xx=
i
∑
n
i = 1
2
nn
2
nx − x
ii
∑∑
n
ii==11
1 2
s sample standard deviation;sx=−x=
()
i
∑
nn−−11n
()
i = 1
1 − α confidence level for the claim that the proportion of the population lying within the tolerance
interval is greater than or equal to the specified level p
µ population mean
σ population standard deviation
2 © ISO 2005 – All rights reserved
---------------------- Page: 10 ----------------------
SIST ISO 16269-6:2006
ISO 16269-6:2005(E)
4 Procedures
4.1 Normal population with known variance and known mean
2
When the values of the mean, µ, and the variance, σ , of a normally distributed population are known, the
distribution of the characteristic under investigation is fully determined. There is exactly a proportion p of the
population:
a) to the right of x = µ − u × σ (one-sided interval);
p
L
b) to the left of x = µ + u × σ (one-sided interval);
p
U
u u
c) between x = µ − × σ and x = µ + × σ (two-sided interval).
(1+ p)/ 2 (1+ p)/ 2
L U
NOTE As such statements are known to be true, they are made with 100 % confidence.
In the above equations, u is p-fractile of the standard normal distribution. Numerical values of u may be
p p
read from the bottom line of the Tables B.1 to B.6 and Tables C.1 to C.6.
4.2 Normal population with known variance and unknown mean
Forms A and B, given in Annex A, are applicable to the case where the variance of the normal population is
known while the mean is unknown. Form A applies to the one-sided case, while Form B applies to the
two-sided case.
4.3 Normal population with unknown variance and unknown mean
Forms C and D, given in Annex A, are applicable to the case where both the mean and the variance of the
normal population are unknown. Form C applies to the one-sided case, while Form D applies to the two-sided
case.
4.4 Any continuous distribution of unknown type
If the characteristic under investigation is a continuous variable from a population of unknown form, and if a
sample of n independent random observations of the characteristic has been taken, then a statistical tolerance
interval can be determined from the ranked observations. The procedure given in Forms E and F of Annex A
provide the determination of the coverage or sample size needed for tolerance intervals determined from the
extreme values x or x of the sample of observations with given confidence level 1 − α.
min max
NOTE Statistical tolerance intervals that do not depend on the shape of the sampled population are called
distribution-free tolerance intervals.
This part of ISO 16269 does not provide procedures for distributions of known type other than the normal
distribution. However, if the distribution is continuous, the distribution-free method may be used. Selected
references to scientific literature that may assist in determining tolerance intervals for other distributions are
also provided at the end of this document.
5 Examples
5.1 Data
Forms A to D, given in Annex A, are illustrated by examples using the numerical values of ISO 2854:1976,
Clause 2, paragraph 1 of the introductory remarks, Table X, yarn 2: 12 measures of the breaking load of
cotton yarn. It should be noted that the number of observations, n = 12, given here for these examples is
[1]
considerably lower than the one recommended in ISO 2602 . The numerical data and calculations in the
different examples are expressed in centi-newtons (see Table 1).
© ISO 2005 – All rights reserved 3
---------------------- Page: 11 ----------------------
SIST ISO 16269-6:2006
ISO 16269-6:2005(E)
Table 1 — Data for Examples 1 to 4
Values in centi-newtons
228,6 232,7 238,8 317,2 315,8 275,1 222,2 236,7 224,7 251,2 210,4 270,7
x
These measurements were obtained from a batch of 12,000 bobbins, from one production job, packed in
120 boxes each containing 100 bobbins. Twelve boxes have been drawn at random from the batch and a
bobbin has been drawn at random from each of these boxes. Test pieces of 50 cm length have been cut from
the yarn on these bobbins, at about 5 m distance from the free end. The tests themselves have been carried
out on the central parts of these test pieces. Previous information makes it reasonable to assume that the
breaking loads measured in these conditions have virtually a normal distribution. It is demonstrated in
ISO 2954:1976 that the data do not contradict the assumption of a normal distribution.
These results yield the following:
Sample size: n = 12
Sample mean: x==3 024,1/12 252,01
2
2
nx − x
()
∑∑ 166 772,27
Sample standard deviation:
s== = 1263,426 3= 35,545
nn−×11211
()
The formal presentation of the calculations will be given only for Form C in Annex A (one-sided interval,
unknown variance).
5.2 Example 1: One-sided statistical tolerance interval under known variance
Suppose that previously obtained measurements have shown that the dispersion is constant from one batch
σ = 33,150
to another from the same supplier, and is represented by a standard deviation , although the mean
is not constant. A limit x is required such that it is possible to assert with confidence level 1 − α = 0,95 (95 %)
L
that at least 0,95 (95 %) of the breaking loads of the items in the batch, when measured under the same
conditions, are above x .
L
Table B.4 gives
k (12; 0,95; 0,95) = 2,120
1
whence
x = xk−−(n;p;1 ασ)×= 252,01− 2,120× 33,150= 181,732
L 1
A smaller value of the lower limit x would be obtained if a larger proportion of the population (for example
L
p = 0,99) and/or a higher confidence level (for example 1 − α = 0,99) were required.
5.3 Example 2: Two-sided statistical tolerance interval under known variance
Under the same conditions as in Example 1, suppose that limits x and x are required such that it is
L U
possible to assert with a confidence level 1 − α = 0,95 that at least a proportion of p = 0,90 (90 %) of the
breaking load of the batch falls between x and x .
L U
Table C.4 gives
k (12; 0,90; 0,95) = 1,889
2
4 © ISO 2005 – All rights reserved
---------------------- Page: 12 ----------------------
SIST ISO 16269-6:2006
ISO 16269-6:2005(E)
whence
x = xk−−(n;p;1 ασ)×= 252,01−1,889× 33,150= 189,390
L 2
x = xk+−(n;p;1 ασ)×= 252,01+1,889× 33,150= 314,630
U 2
Comparison with Example 1 should make it clear that assuring that at least 90 % of a population lies between
the limits x and x is not the same thing as assuring that no more than 5 % lies beyond each limit.
L U
5.4 Example 3: One-sided statistical tolerance interval under unknown variance
Here, it is supposed that the standard deviation of the population is unknown and has to be estimated from the
sample. The same requirements will be assumed as for the case where the standard deviation is known
(Example 1), thus, p = 0,95 and 1 − α = 0,95. The presentation of the results is given in detail below.
Determination of the statistical tolerance interval of proportion p:
a) one-sided interval “to the right”
Determined values:
b) proportion of the population selected for the tolerance interval: p = 0,95
c) chosen confidence level: 1 − α = 0,95
d) sample size: n = 12
Value of tolerance factor from Table D.4:
k (n; p; 1 − α) = 2,737
3
Calculations:
xx==/n 252,01
∑
2
2
nx − x
()
∑∑
s = = 35,545
nn −1
()
kn( ;p;1−×α) s= 97,286 7
3
Results: one-sided interval “to the right”
The tolerance interval which will contain at least a proportion p of the population with confidence level 1 − α
has a lower limit
x =−xkn( ;p;1− α)×s= 154,723
L 3
© ISO 2005 – All rights reserved 5
---------------------- Page: 13 ----------------------
SIST ISO 16269-6:2006
ISO 16269-6:2005(E)
5.5 Example 4: Two-sided statistical tolerance interval under unknown variance
Under the same conditions as in Example 2, suppose it is required to calculate the limits x and x such
L U
that it is possible to assert with a confidence level 1 − α = 0,95 that in a proportion of the batch at least equal
to p = 0,90 (90 %) the breaking load falls between x and x .
L U
Table E.4 gives
kn(;p;1−=α) 2,671
4
whence
xx=−k (n;p;1−α)×s= 252,01− 2,671× 35,545= 157,069
L 4
xx=+k (n;p;1−α)×s= 252,01+ 2,671× 35,545= 346,951
U 4
It will be noted that the value of x is smaller and the value of x higher than in Example 2 (known variance),
L U
because the use of s instead of σ requires a larger value of the tolerance factor to allow for the extra
uncertainty. It is necessary to have to pay a penalty for not knowing the population standard deviation σ and
the extension of the statistical tolerance interval takes this into account. Of course, it is not quite sure that the
value σ = 33,150 used in Examples 1 and 2 is correct. Therefore, it is wiser to use the estimate, s, together
with Tables D.4 or E.4.
5.6 Example 5: Distribution-free statistical tolerance interval for continuous distribution
In a fatigue test by rotational stress carried out on a component of an aeronautical engine, a sample of
15 items has given the results (measurement of endurance), shown in ascending order of values in Table 2.
Table 2 — Data for Example 5
x 0,200 0,330 0,450 0,490 0,780 0,920 0,950 0,970 1,040 1,710 2,220 2,275 3,650 7,000 8,800
A graphical examination of checking normality, such as probability plot, shows that the hypothesis of normality
for the population of components should almost certainly be rejected (see ISO 5479). The methods of Form E,
given in Annex A, for determination of a statistical tolerance interval are therefore applicable.
The extreme values from the sample of n = 15 measurements are:
x = 0,200, x = 8,800
min max
Suppose that the required confidence level 1 − α is 0,95.
a) What is the maximum proportion of the population of components that will fall below x = 0,200?
min
Table F.1, for 1 − α = 0,95, gives for the minimum proportion above x a value of p slightly higher than
min
0,75 (75 %). Hence, for the maximum proportion below x a value of 1 − p slightly lower than
min
0,25 (25 %).
b) What sample size is necessary for it to be possible to assert, at a confidence level 0,95, that a proportion
at least p = 0,90 (90 %) of the population of components will be found below the largest of the values from
that sample? Table F.1, for 1 − α = 0,95 and p = 0,90, gives n = 29.
6 © ISO 2005 – All rights reserved
---------------------- Page: 14 ----------------------
SIST ISO 16269-6:2006
ISO 16269-6:2005(E)
c) At a confidence level of 0,95, what is the minimum proportion of the population of components that fall
between x = 0,200 and x = 8,800? Table G.1, for 1 − α = 0,95 and n = 15, gives p slightly below
min max
0,75 (75 %).
d) What sample size is necessary for it to be possible to assert at a confidence level 0,95 that a proportion
of at least p = 0,90 (90 %) of the population of components will be found to fall between the smallest and
the largest values from that sample? Table G.1, for 1 − α = 0,95 and p = 0,90, gives n = 46.
e) In general, if a check for normality (see ISO 5479) indicates a departure from the normal distribution,
some transformation will be recommended based on the knowledge of the collected data. For example,
fatigue data are often approximated lognormally distributed. In such cases, the data could be transformed
to normality. Tolerance intervals are then calculated and finally transformed back into the original units.
See Annex H for the construction of a statistical tolerance interval for distribution-free tolerance intervals for
any type of distribution. Annex I gives the computation of factors for two-sided parametric statistical tolerance
intervals.
© ISO 2005 – All rights reserved 7
---------------------- Page: 15 ----------------------
SIST ISO 16269-6:2006
ISO 16269-6:2005(E)
Annex A
(informative)
Forms for tolerance intervals
Form A — One-sided statistical tolerance interval (known variance)
Determination of a one-sided statistical tolerance interval with cov
...
INTERNATIONAL ISO
STANDARD 16269-6
First edition
2005-04-01
Statistical interpretation of data —
Part 6:
Determination of statistical tolerance
intervals
Interprétation statistique des données —
Partie 6: Détermination des intervalles statistiques de tolérance
Reference number
ISO 16269-6:2005(E)
©
ISO 2005
---------------------- Page: 1 ----------------------
ISO 16269-6:2005(E)
PDF disclaimer
This PDF file may contain embedded typefaces. In accordance with Adobe's licensing policy, this file may be printed or viewed but
shall not be edited unless the typefaces which are embedded are licensed to and installed on the computer performing the editing. In
downloading this file, parties accept therein the responsibility of not infringing Adobe's licensing policy. The ISO Central Secretariat
accepts no liability in this area.
Adobe is a trademark of Adobe Systems Incorporated.
Details of the software products used to create this PDF file can be found in the General Info relative to the file; the PDF-creation
parameters were optimized for printing. Every care has been taken to ensure that the file is suitable for use by ISO member bodies. In
the unlikely event that a problem relating to it is found, please inform the Central Secretariat at the address given below.
© ISO 2005
All rights reserved. Unless otherwise specified, no part of this publication may be reproduced or utilized in any form or by any means,
electronic or mechanical, including photocopying and microfilm, without permission in writing from either ISO at the address below or
ISO's member body in the country of the requester.
ISO copyright office
Case postale 56 • CH-1211 Geneva 20
Tel. + 41 22 749 01 11
Fax + 41 22 749 09 47
E-mail copyright@iso.org
Web www.iso.org
Published in Switzerland
ii © ISO 2005 – All rights reserved
---------------------- Page: 2 ----------------------
ISO 16269-6:2005(E)
Contents Page
Foreword. iv
Introduction . v
1 Scope. 1
2 Normative references . 1
3 Terms, definitions and symbols . 1
3.1 Terms and definitions. 1
3.2 Symbols . 2
4 Procedures . 3
4.1 Normal population with known variance and known mean. 3
4.2 Normal population with known variance and unknown mean . 3
4.3 Normal population with unknown variance and unknown mean. 3
4.4 Any continuous distribution of unknown type . 3
5 Examples. 3
5.1 Data. 3
5.2 Example 1: One-sided statistical tolerance interval under known variance. 4
5.3 Example 2: Two-sided statistical tolerance interval under known variance . 4
5.4 Example 3: One-sided statistical tolerance interval under unknown variance . 5
5.5 Example 4: Two-sided statistical tolerance interval under unknown variance. 6
5.6 Example 5: Distribution-free statistical tolerance interval for continuous distribution . 6
Annex A (informative) Forms for tolerance intervals. 8
Annex B (normative) One-sided statistical tolerance limit factors, k (n; p; 1 − α), for known σ . 14
1
Annex C (normative) Two-sided statistical tolerance limit factors, k (n; p; 1 − α), for known σ. 17
2
Annex D (normative) One-sided statistical tolerance limit factors, k (n; p; 1 − α), for unknown σ. 20
3
Annex E (normative) Two-sided statistical tolerance limit factors, k (n; p; 1 − α), for unknown σ. 23
4
Annex F (normative) One-sided distribution-free statistical tolerance intervals. 26
Annex G (normative) Two-sided distribution-free statistical tolerance intervals. 27
Annex H (informative) Construction of a distribution-free statistical tolerance interval for any type
of distribution . 28
Annex I (informative) Computation of factors for two-sided parametric statistical tolerance
intervals. 29
Bibliography . 30
© ISO 2005 – All rights reserved iii
---------------------- Page: 3 ----------------------
ISO 16269-6:2005(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.
International Standards are drafted in accordance with the rules given in the ISO/IEC Directives, Part 2.
The main task of technical committees is to prepare International Standards. Draft International Standards
adopted by the technical committees are circulated to the member bodies for voting. Publication as an
International Standard requires approval by at least 75 % of the member bodies casting a vote.
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.
ISO 16269-6 was prepared by Technical Committee ISO/TC 69, Applications of statistical methods.
This first edition of ISO 16269-6 cancels and replaces ISO 3207:1975, which has been technically revised.
ISO 16269 consists of the following parts, under the general title Statistical interpretation of data:
Part 6: Determination of statistical tolerance intervals
Part 7: Median — Estimation and confidence intervals
Part 8: Determination of prediction intervals
iv © ISO 2005 – All rights reserved
---------------------- Page: 4 ----------------------
ISO 16269-6:2005(E)
Introduction
A statistical tolerance interval is an estimated interval, based on a sample, which can be asserted with
confidence 1 − α, for example 95 %, to contain at least a specified proportion p of the items in the population.
The limits of a statistical tolerance interval are called statistical tolerance limits. The confidence level 1 − α is
the probability that a statistical tolerance interval constructed in the prescribed manner will contain at least a
proportion p of the population. Conversely, the probability that this interval will contain less than the proportion
p of the population is α. This part of ISO 16269 describes both one-sided and two-sided statistical tolerance
intervals; a one-sided interval is constructed with an upper or a lower limit while a two-sided interval is
constructed with both an upper and a lower limit.
Tolerance intervals are functions of the observations of the sample, i.e. statistics, and they will generally take
different values for different samples. It is necessary that the observations be independent for the procedures
provided in this part of ISO 16269 to be valid.
Two types of tolerance interval are provided in this part of ISO 16269, parametric and distribution-free. The
parametric approach is based on the assumption that the characteristic being studied in the population has a
normal distribution; hence the confidence that the calculated statistical tolerance interval contains at least a
proportion p of the population can only be taken to be 1 − α if the normality assumption is true. For normally
distributed characteristics, the statistical tolerance interval is determined using one of the Forms A, B, C or D
given in Annex A.
Parametric methods for distributions other than the normal are not considered in this part of ISO 16269. If
departure from normality is suspected in the population, distribution-free statistical tolerance intervals may be
constructed. The procedure for the determination of a statistical tolerance interval for any continuous
distribution is provided in Forms E and F of Annex A.
The tolerance limits discussed in this part of ISO 16269 can be used to compare the natural capability of a
process with one or two given specification limits, either an upper one U or a lower one L or both in statistical
process management. An indication of this is the fact that these tolerance limits have also been called natural
process limits. See ISO 3534-2:1993, 3.2.4, and the general remarks in ISO 3207 which will be cancelled and
replaced by this part of ISO 16269.
Above the upper specification limit U there is the upper fraction nonconforming p (ISO 3534-2:—, 3.2.5.5 and
U
3.3.1.4) and below the lower specification limit L there is the lower fraction nonconforming p (ISO 3534-2:—,
L
3.2.5.6 and 3.3.1.5). The sum p + p = p is called the total fraction nonconforming. (ISO 3534-2:—, 3.2.5.7).
U L T
Between the specification limits U and L there is the fraction conforming 1 − p .
T
In statistical process management the limits U and L are fixed in advance and the fractions p , p and p are
U L T
either calculated, if the distribution is assumed to be known, or otherwise estimated. There are many
applications of statistical tolerance intervals, although the above shows an example to a quality control
problem. Wider applications and more statistical intervals are introduced in many textbooks such as Hahn and
[10]
Meeker .
In contrast, for the tolerance intervals considered in this part of ISO 16269, the confidence level for the interval
estimator and the proportion of the distribution within the interval (corresponding to the fraction conforming
mentioned above) are fixed in advance, and the limits are estimated. These limits may be compared with U
and L. Hence the appropriateness of the given specification limits U and L can be compared with the actual
properties of the process. The one-sided tolerance intervals are used when only either the upper specification
limit U or the lower specification limit L is relevant, while the two-sided intervals are used when both the upper
and the lower specification limits are considered simultaneously.
The terminology with regard to these different limits and intervals has been confusing as the “specification
limits” were earlier also called “tolerance limits” (see the terminology standard ISO 3534-2:1993, 1.4.3, where
both these terms as well as the term “limiting values” were all used as synonyms for this concept). In the latest
© ISO 2005 – All rights reserved v
---------------------- Page: 5 ----------------------
ISO 16269-6:2005(E)
revision of ISO 3534-2:—, only the term specification limits have been kept for this concept. Furthermore, the
[5]
Guide for the expression of uncertainty in measurement uses the term “coverage factor” defined as a
“numerical factor used as a multiplier of the combined standard uncertainty in order to obtain an expanded
uncertainty”. This use of “coverage” differs from the use of the term in this part of ISO 16269.
vi © ISO 2005 – All rights reserved
---------------------- Page: 6 ----------------------
INTERNATIONAL STANDARD ISO 16269-6:2005(E)
Statistical interpretation of data —
Part 6:
Determination of statistical tolerance intervals
1 Scope
This part of ISO 16269 describes procedures for establishing tolerance intervals that include at least a
specified proportion of the population with a specified confidence level. Both one-sided and two-sided
statistical tolerance intervals are provided, a one-sided interval having either an upper or a lower limit while a
two-sided interval has both upper and lower limits. Two methods are provided, a parametric method for the
case where the characteristic being studied has a normal distribution and a distribution-free method for the
case where nothing is known about the distribution except that it is continuous.
2 Normative references
The following referenced documents are indispensable for the application 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 3534-1, Statistics — Vocabulary and symbols — Part 1: Probability and general statistical terms
1)
ISO 3534-2:— , Statistics — Vocabulary and symbols — Part 2: Applied statistics
3 Terms, definitions and symbols
3.1 Terms and definitions
For the purposes of this document, the terms and definition given in ISO 3534-1, ISO 3534-2 and the following
apply.
3.1.1
statistical tolerance interval
interval determined from a random sample in such a way that one may have a specified level of confidence
that the interval covers at least a specified proportion of the sampled population
NOTE The confidence level in this context is the long-run proportion of intervals constructed in this manner that will
include at least the specified proportion of the sampled population.
3.1.2
statistical tolerance limit
statistic representing an end point of a statistical tolerance interval
NOTE Statistical tolerance intervals can be either one-sided, in which case they have either an upper or a lower
statistical tolerance limit, or two-sided, in which case they have both.
1) To be published. (Revision of ISO 3534-2:1993)
© ISO 2005 – All rights reserved 1
---------------------- Page: 7 ----------------------
ISO 16269-6:2005(E)
3.1.3
coverage
proportion of items in a population lying within a statistical tolerance interval
NOTE This concept is not to be confused with the concept coverage factor used in the Guide for the expression of
[5]
uncertainty in measurement (GUM ) .
3.1.4
normal population
normally distributed population
3.2 Symbols
For the purposes of this part of ISO 16269, the following symbols apply.
i suffix of an observation
k (n; p; 1 − α) factor used to determine x or x when the value of σ is known for one-sided tolerance
L U
1
interval
k (n; p; 1 − α) factor used to determine x and x when the value of σ is known for two-sided tolerance
L U
2
interval
k (n; p; 1 − α) factor used to determine x or x when the value of σ is unknown for one-sided tolerance
L U
3
interval
k (n; p; 1 − α) factor used to determine x and x when the value of σ is unknown for two-sided tolerance
L U
4
interval
n number of observations in the sample
p minimum proportion of the population claimed to be lying in the statistical tolerance interval
u p-fractile of the standard normal distribution
p
x ith observed value (1in=,2,.,)
i
x maximum value of the observed values: x = max {x , x , …, x }
max max 1 2 n
x minimum value of the observed values: x = min {x , x , …, x }
min min 1 2 n
x lower limit of the statistical tolerance interval
L
x upper limit of the statistical tolerance interval
U
n
1
x sample mean, xx=
i
∑
n
i = 1
2
nn
2
nx − x
ii
∑∑
n
ii==11
1 2
s sample standard deviation;sx=−x=
()
i
∑
nn−−11n
()
i = 1
1 − α confidence level for the claim that the proportion of the population lying within the tolerance
interval is greater than or equal to the specified level p
µ population mean
σ population standard deviation
2 © ISO 2005 – All rights reserved
---------------------- Page: 8 ----------------------
ISO 16269-6:2005(E)
4 Procedures
4.1 Normal population with known variance and known mean
2
When the values of the mean, µ, and the variance, σ , of a normally distributed population are known, the
distribution of the characteristic under investigation is fully determined. There is exactly a proportion p of the
population:
a) to the right of x = µ − u × σ (one-sided interval);
p
L
b) to the left of x = µ + u × σ (one-sided interval);
p
U
u u
c) between x = µ − × σ and x = µ + × σ (two-sided interval).
(1+ p)/ 2 (1+ p)/ 2
L U
NOTE As such statements are known to be true, they are made with 100 % confidence.
In the above equations, u is p-fractile of the standard normal distribution. Numerical values of u may be
p p
read from the bottom line of the Tables B.1 to B.6 and Tables C.1 to C.6.
4.2 Normal population with known variance and unknown mean
Forms A and B, given in Annex A, are applicable to the case where the variance of the normal population is
known while the mean is unknown. Form A applies to the one-sided case, while Form B applies to the
two-sided case.
4.3 Normal population with unknown variance and unknown mean
Forms C and D, given in Annex A, are applicable to the case where both the mean and the variance of the
normal population are unknown. Form C applies to the one-sided case, while Form D applies to the two-sided
case.
4.4 Any continuous distribution of unknown type
If the characteristic under investigation is a continuous variable from a population of unknown form, and if a
sample of n independent random observations of the characteristic has been taken, then a statistical tolerance
interval can be determined from the ranked observations. The procedure given in Forms E and F of Annex A
provide the determination of the coverage or sample size needed for tolerance intervals determined from the
extreme values x or x of the sample of observations with given confidence level 1 − α.
min max
NOTE Statistical tolerance intervals that do not depend on the shape of the sampled population are called
distribution-free tolerance intervals.
This part of ISO 16269 does not provide procedures for distributions of known type other than the normal
distribution. However, if the distribution is continuous, the distribution-free method may be used. Selected
references to scientific literature that may assist in determining tolerance intervals for other distributions are
also provided at the end of this document.
5 Examples
5.1 Data
Forms A to D, given in Annex A, are illustrated by examples using the numerical values of ISO 2854:1976,
Clause 2, paragraph 1 of the introductory remarks, Table X, yarn 2: 12 measures of the breaking load of
cotton yarn. It should be noted that the number of observations, n = 12, given here for these examples is
[1]
considerably lower than the one recommended in ISO 2602 . The numerical data and calculations in the
different examples are expressed in centi-newtons (see Table 1).
© ISO 2005 – All rights reserved 3
---------------------- Page: 9 ----------------------
ISO 16269-6:2005(E)
Table 1 — Data for Examples 1 to 4
Values in centi-newtons
228,6 232,7 238,8 317,2 315,8 275,1 222,2 236,7 224,7 251,2 210,4 270,7
x
These measurements were obtained from a batch of 12,000 bobbins, from one production job, packed in
120 boxes each containing 100 bobbins. Twelve boxes have been drawn at random from the batch and a
bobbin has been drawn at random from each of these boxes. Test pieces of 50 cm length have been cut from
the yarn on these bobbins, at about 5 m distance from the free end. The tests themselves have been carried
out on the central parts of these test pieces. Previous information makes it reasonable to assume that the
breaking loads measured in these conditions have virtually a normal distribution. It is demonstrated in
ISO 2954:1976 that the data do not contradict the assumption of a normal distribution.
These results yield the following:
Sample size: n = 12
Sample mean: x==3 024,1/12 252,01
2
2
nx − x
()
∑∑ 166 772,27
Sample standard deviation:
s== = 1263,426 3= 35,545
nn−×11211
()
The formal presentation of the calculations will be given only for Form C in Annex A (one-sided interval,
unknown variance).
5.2 Example 1: One-sided statistical tolerance interval under known variance
Suppose that previously obtained measurements have shown that the dispersion is constant from one batch
σ = 33,150
to another from the same supplier, and is represented by a standard deviation , although the mean
is not constant. A limit x is required such that it is possible to assert with confidence level 1 − α = 0,95 (95 %)
L
that at least 0,95 (95 %) of the breaking loads of the items in the batch, when measured under the same
conditions, are above x .
L
Table B.4 gives
k (12; 0,95; 0,95) = 2,120
1
whence
x = xk−−(n;p;1 ασ)×= 252,01− 2,120× 33,150= 181,732
L 1
A smaller value of the lower limit x would be obtained if a larger proportion of the population (for example
L
p = 0,99) and/or a higher confidence level (for example 1 − α = 0,99) were required.
5.3 Example 2: Two-sided statistical tolerance interval under known variance
Under the same conditions as in Example 1, suppose that limits x and x are required such that it is
L U
possible to assert with a confidence level 1 − α = 0,95 that at least a proportion of p = 0,90 (90 %) of the
breaking load of the batch falls between x and x .
L U
Table C.4 gives
k (12; 0,90; 0,95) = 1,889
2
4 © ISO 2005 – All rights reserved
---------------------- Page: 10 ----------------------
ISO 16269-6:2005(E)
whence
x = xk−−(n;p;1 ασ)×= 252,01−1,889× 33,150= 189,390
L 2
x = xk+−(n;p;1 ασ)×= 252,01+1,889× 33,150= 314,630
U 2
Comparison with Example 1 should make it clear that assuring that at least 90 % of a population lies between
the limits x and x is not the same thing as assuring that no more than 5 % lies beyond each limit.
L U
5.4 Example 3: One-sided statistical tolerance interval under unknown variance
Here, it is supposed that the standard deviation of the population is unknown and has to be estimated from the
sample. The same requirements will be assumed as for the case where the standard deviation is known
(Example 1), thus, p = 0,95 and 1 − α = 0,95. The presentation of the results is given in detail below.
Determination of the statistical tolerance interval of proportion p:
a) one-sided interval “to the right”
Determined values:
b) proportion of the population selected for the tolerance interval: p = 0,95
c) chosen confidence level: 1 − α = 0,95
d) sample size: n = 12
Value of tolerance factor from Table D.4:
k (n; p; 1 − α) = 2,737
3
Calculations:
xx==/n 252,01
∑
2
2
nx − x
()
∑∑
s = = 35,545
nn −1
()
kn( ;p;1−×α) s= 97,286 7
3
Results: one-sided interval “to the right”
The tolerance interval which will contain at least a proportion p of the population with confidence level 1 − α
has a lower limit
x =−xkn( ;p;1− α)×s= 154,723
L 3
© ISO 2005 – All rights reserved 5
---------------------- Page: 11 ----------------------
ISO 16269-6:2005(E)
5.5 Example 4: Two-sided statistical tolerance interval under unknown variance
Under the same conditions as in Example 2, suppose it is required to calculate the limits x and x such
L U
that it is possible to assert with a confidence level 1 − α = 0,95 that in a proportion of the batch at least equal
to p = 0,90 (90 %) the breaking load falls between x and x .
L U
Table E.4 gives
kn(;p;1−=α) 2,671
4
whence
xx=−k (n;p;1−α)×s= 252,01− 2,671× 35,545= 157,069
L 4
xx=+k (n;p;1−α)×s= 252,01+ 2,671× 35,545= 346,951
U 4
It will be noted that the value of x is smaller and the value of x higher than in Example 2 (known variance),
L U
because the use of s instead of σ requires a larger value of the tolerance factor to allow for the extra
uncertainty. It is necessary to have to pay a penalty for not knowing the population standard deviation σ and
the extension of the statistical tolerance interval takes this into account. Of course, it is not quite sure that the
value σ = 33,150 used in Examples 1 and 2 is correct. Therefore, it is wiser to use the estimate, s, together
with Tables D.4 or E.4.
5.6 Example 5: Distribution-free statistical tolerance interval for continuous distribution
In a fatigue test by rotational stress carried out on a component of an aeronautical engine, a sample of
15 items has given the results (measurement of endurance), shown in ascending order of values in Table 2.
Table 2 — Data for Example 5
x 0,200 0,330 0,450 0,490 0,780 0,920 0,950 0,970 1,040 1,710 2,220 2,275 3,650 7,000 8,800
A graphical examination of checking normality, such as probability plot, shows that the hypothesis of normality
for the population of components should almost certainly be rejected (see ISO 5479). The methods of Form E,
given in Annex A, for determination of a statistical tolerance interval are therefore applicable.
The extreme values from the sample of n = 15 measurements are:
x = 0,200, x = 8,800
min max
Suppose that the required confidence level 1 − α is 0,95.
a) What is the maximum proportion of the population of components that will fall below x = 0,200?
min
Table F.1, for 1 − α = 0,95, gives for the minimum proportion above x a value of p slightly higher than
min
0,75 (75 %). Hence, for the maximum proportion below x a value of 1 − p slightly lower than
min
0,25 (25 %).
b) What sample size is necessary for it to be possible to assert, at a confidence level 0,95, that a proportion
at least p = 0,90 (90 %) of the population of components will be found below the largest of the values from
that sample? Table F.1, for 1 − α = 0,95 and p = 0,90, gives n = 29.
6 © ISO 2005 – All rights reserved
---------------------- Page: 12 ----------------------
ISO 16269-6:2005(E)
c) At a confidence level of 0,95, what is the minimum proportion of the population of components that fall
between x = 0,200 and x = 8,800? Table G.1, for 1 − α = 0,95 and n = 15, gives p slightly below
min max
0,75 (75 %).
d) What sample size is necessary for it to be possible to assert at a confidence level 0,95 that a proportion
of at least p = 0,90 (90 %) of the population of components will be found to fall between the smallest and
the largest values from that sample? Table G.1, for 1 − α = 0,95 and p = 0,90, gives n = 46.
e) In general, if a check for normality (see ISO 5479) indicates a departure from the normal distribution,
some transformation will be recommended based on the knowledge of the collected data. For example,
fatigue data are often approximated lognormally distributed. In such cases, the data could be transformed
to normality. Tolerance intervals are then calculated and finally transformed back into the original units.
See Annex H for the construction of a statistical tolerance interval for distribution-free tolerance intervals for
any type of distribution. Annex I gives the computation of factors for two-sided parametric statistical tolerance
intervals.
© ISO 2005 – All rights reserved 7
---------------------- Page: 13 ----------------------
ISO 16269-6:2005(E)
Annex A
(informative)
Forms for tolerance intervals
Form A — One-sided statistical tolerance interval (known variance)
Determination of a one-sided statistical tolerance interval with coverage p at confidence level 1 − α
a) One-sided interval “to the left”
b) One-sided interval “to the right”
Known values:
2
c) the variance: σ =
d) the standard deviation: σ =
Determined values:
e) proportion of the population selected for the tolerance interval: p =
f) chosen confidence level: 1 − α =
g) sample size: n =
Tabulated factor:
k (n; p; 1 − α) =
1
This value can be read from the tables given in Annex B for a range of values of n, p and 1 − α.
Calculations:
xx==/n
∑
k (n; p; 1 − α) × σ =
1
Results:
a) One-sided interval “to the left”
The one-sided statistical tolerance interval with coverage p at confidence level 1 − α has upper limit
xx=+k (;np;1− α)×σ=
U 1
b) One-sided interval “to the right”
The one-sided statistical tolerance interval with coverage p at confidence level 1 − α has lower limit
x =−xkn(;p;1− α)×σ=
L 1
8 © ISO 2005 – All rights reserved
---------------------- Page: 14 ----------------------
ISO 16269-6:2005(E)
Form B — Two-sided statistical tolerance interval (known variance)
Determination of a tw
...
NORME ISO
INTERNATIONALE 16269-6
Première édition
2005-04-01
Version corrigée
2006-01-01
Interprétation statistique des données —
Partie 6:
Détermination des intervalles statistiques
de dispersion
Statistical interpretation of data —
Part 6: Determination of statistical tolerance intervals
Numéro de référence
ISO 16269-6:2005(F)
©
ISO 2005
---------------------- Page: 1 ----------------------
ISO 16269-6:2005(F)
PDF – Exonération de responsabilité
Le présent fichier PDF peut contenir des polices de caractères intégrées. Conformément aux conditions de licence d'Adobe, ce fichier
peut être imprimé ou visualisé, mais ne doit pas être modifié à moins que l'ordinateur employé à cet effet ne bénéficie d'une licence
autorisant l'utilisation de ces polices et que celles-ci y soient installées. Lors du téléchargement de ce fichier, les parties concernées
acceptent de fait la responsabilité de ne pas enfreindre les conditions de licence d'Adobe. Le Secrétariat central de l'ISO décline toute
responsabilité en la matière.
Adobe est une marque déposée d'Adobe Systems Incorporated.
Les détails relatifs aux produits logiciels utilisés pour la création du présent fichier PDF sont disponibles dans la rubrique General Info
du fichier; les paramètres de création PDF ont été optimisés pour l'impression. Toutes les mesures ont été prises pour garantir
l'exploitation de ce fichier par les comités membres de l'ISO. Dans le cas peu probable où surviendrait un problème d'utilisation,
veuillez en informer le Secrétariat central à l'adresse donnée ci-dessous.
© ISO 2005
Droits de reproduction réservés. Sauf prescription différente, aucune partie de cette publication ne peut être reproduite ni utilisée sous
quelque forme que ce soit et par aucun procédé, électronique ou mécanique, y compris la photocopie et les microfilms, sans l'accord écrit
de l'ISO à l'adresse ci-après ou du comité membre de l'ISO dans le pays du demandeur.
ISO copyright office
Case postale 56 • CH-1211 Geneva 20
Tel. + 41 22 749 01 11
Fax. + 41 22 749 09 47
E-mail copyright@iso.org
Web www.iso.org
Publié en Suisse
ii © ISO 2005 – Tous droits réservés
---------------------- Page: 2 ----------------------
ISO 16269-6:2005(F)
Sommaire Page
Avant-propos. iv
Introduction . v
1 Domaine d'application. 1
2 Références normatives . 1
3 Termes, définitions et symboles . 1
3.1 Termes et définitions. 1
3.2 Symboles . 2
4 Méthodes . 3
4.1 Population normale avec une variance et une moyenne connues. 3
4.2 Population normale avec une variance connue et une moyenne inconnue. 3
4.3 Population normale avec une variance et une moyenne inconnues . 3
4.4 Distribution continue quelconque de type inconnu. 3
5 Exemples . 4
5.1 Données. 4
5.2 Exemple 1: intervalle statistique de dispersion unilatéral sous variance connue . 4
5.3 Exemple 2: intervalle statistique de dispersion bilatéral sous variance connue. 5
5.4 Exemple 3: intervalle statistique de dispersion unilatéral sous variance inconnue . 5
5.5 Exemple 4: intervalle statistique de dispersion bilatéral sous variance inconnue . 6
5.6 Exemple 5: intervalle statistique de dispersion non paramétrique pour une distribution
continue . 7
Annexe A (informative) Formulaires pour les intervalles de dispersion . 8
Annexe B (normative) Facteurs de la limite statistique de dispersion unilatérale, k (n; p; 1 − α), pour
1
un écart-type de la population, σ, connu. 14
Annexe C (normative) Facteurs de la limite statistique de dispersion bilatérale, k (n; p; 1 − α), pour
2
un écart-type de la population, σ, connu. 17
Annexe D (normative) Facteurs de la limite statistique de dispersion unilatérale, k (n; p; 1 − α), pour
3
un écart-type de la population, σ, inconnu . 20
Annexe E (normative) Facteurs de la limite statistique de dispersion bilatérale, k (n; p; 1 − α), pour
4
un écart-type de la population, σ, inconnu . 23
Annexe F (normative) Intervalles statistiques de dispersion unilatéraux non paramétriques . 26
Annexe G (normative) Intervalles statistiques de dispersion bilatéraux non paramétriques. 27
Annexe H (informative) Construction d'un intervalle statistique de dispersion non paramétrique
pour un type de distribution quelconque.28
Annexe I (informative) Calculs des facteurs des intervalles statistiques de dispersion bilatéraux
paramétriques . 29
Bibliographie . 30
© ISO 2005 – Tous droits réservés iii
---------------------- Page: 3 ----------------------
ISO 16269-6:2005(F)
Avant-propos
L'ISO (Organisation internationale de normalisation) est une fédération mondiale d'organismes nationaux de
normalisation (comités membres de l'ISO). L'élaboration des Normes internationales est en général confiée
aux comités techniques de l'ISO. Chaque comité membre intéressé par une étude a le droit de faire partie du
comité technique créé à cet effet. Les organisations internationales, gouvernementales et non
gouvernementales, en liaison avec l'ISO participent également aux travaux. L'ISO collabore étroitement avec
la Commission électrotechnique internationale (CEI) en ce qui concerne la normalisation électrotechnique.
Les Normes internationales sont rédigées conformément aux règles données dans les Directives ISO/CEI,
Partie 2.
La tâche principale des comités techniques est d'élaborer les Normes internationales. Les projets de Normes
internationales adoptés par les comités techniques sont soumis aux comités membres pour vote. Leur
publication comme Normes internationales requiert l'approbation de 75 % au moins des comités membres
votants.
L'attention est appelée sur le fait que certains des éléments du présent document peuvent faire l'objet de
droits de propriété intellectuelle ou de droits analogues. L'ISO ne saurait être tenue pour responsable de ne
pas avoir identifié de tels droits de propriété et averti de leur existence.
L'ISO 16269-6 a été élaborée par le comité technique ISO/TC 69, Application des méthodes statistiques.
Cette première édition de l'ISO 16269-6 annule et remplace l'ISO 3207:1975, qui a fait l'objet d'une révision
technique.
L'ISO 16269 comprend les parties suivantes, présentées sous le titre général Interprétation statistique des
données:
Partie 6: Détermination des intervalles statistiques de dispersion
Partie 7: Médiane — Estimation et intervalles de confiance
Partie 8: Détermination des intervalles de prédiction
Dans la présente version corrigée, le terme «tolérance» a été remplacé par «dispersion» et les termes «limite
de spécification» ont été remplacés par «limite de tolérance» dans l'ensemble du document.
iv © ISO 2005 – Tous droits réservés
---------------------- Page: 4 ----------------------
ISO 16269-6:2005(F)
Introduction
Un intervalle statistique de dispersion est un intervalle estimé, d'après un échantillon, pour lequel il est
possible d'affirmer avec un niveau de confiance 1 − α, par exemple 95 %, qu'il contient au moins une
proportion donnée p d'individus de la population. Les limites d'un intervalle statistique de dispersion sont
appelées limites statistiques de dispersion. Le niveau de confiance 1 − α est la probabilité selon laquelle un
intervalle statistique de dispersion construit de la manière spécifiée contiendra au moins une proportion p
d'individus de la population. Inversement, la probabilité que cet intervalle contiendra moins que la proportion p
d'individus de la population est α. La présente partie de l'ISO 16269 décrit les intervalles statistiques de
dispersion unilatéraux et les intervalles statistiques de dispersion bilatéraux; un intervalle unilatéral est
construit avec une limite inférieure ou une limite supérieure tandis qu'un intervalle bilatéral est construit avec
une limite supérieure et une limite inférieure.
Les intervalles de dispersion sont fonction des observations de l'échantillon, c'est-à-dire des statistiques, et
leurs valeurs seront généralement différentes pour des échantillons différents. Il est nécessaire que les
observations soient indépendantes pour que les méthodes indiquées dans la présente partie de l'ISO 16269
soient valables.
La présente partie de l'ISO 16269 stipule deux types d'intervalles statistiques de dispersion: l'intervalle
paramétrique et l'intervalle non paramétrique. L'approche paramétrique se fonde sur l'hypothèse selon
laquelle la caractéristique étudiée dans la population a une distribution normale; ainsi, si l'hypothèse de
normalité est avérée, le niveau de confiance avec lequel l'intervalle statistique de dispersion contient au moins
une proportion p d'individus de la population ne peut être que de 1 − α. Pour les caractéristiques distribuées
normalement, l'intervalle statistique de dispersion est déterminé à l'aide des formulaires A, B, C et D donnés
dans l'Annexe A.
La présente partie de l'ISO 16269 ne traite pas des méthodes paramétriques s'appliquant à des distributions
autres que les distributions normales. Si des écarts par rapport à la normalité sont suspectés dans la
population, des intervalles statistiques de dispersion non paramétriques peuvent être construits. La procédure
de détermination d'un intervalle statistique de dispersion pour une distribution continue quelconque est
indiquée aux formulaires E et F de l'Annexe A.
Les limites de dispersion abordées dans la présente partie de l'ISO 16269 peuvent être utilisées pour
comparer l'aptitude naturelle d'un processus avec une ou deux limites de tolérance données, soit une limite
supérieure, U, soit une limite inférieure, L, ou encore les deux, dans la gestion d'un processus statistique.
Cela est indiqué par le fait que ces limites de dispersion ont également été appelées limites naturelles du
processus. Voir l'ISO 3534-2:1993, 3.2.4, ainsi que les remarques générales de l'ISO 3207, qui sera annulée
et remplacée par la présente partie de l'ISO 16269.
Au-dessus de la limite de tolérance supérieure, U, il y a la fraction supérieure non conforme, p
U
(ISO 3534-2:—, 3.2.5.5 et 3.3.1.4), et en dessous de la limite de tolérance inférieure, L, il y a la fraction
inférieure non conforme, p (ISO 3534-2:—, 3.2.5.6 et 3.3.1.5). La somme p + p = p est appelée fraction
L U L T
totale non conforme (ISO 3534-2:—, 3.2.5.7). Entre les limites de tolérance U et L, il y a la fraction conforme
1 – p .
T
Dans la gestion du processus statistique, les limites U et L sont fixées à l'avance et les fractions p , p et p
U L T
sont soit calculées, lorsque la distribution est supposée connue, soit estimées. Il existe beaucoup
d'applications d'intervalles statistiques de dispersion, bien que l'exemple ci-dessus montre un exemple d'un
problème de contrôle qualité. Des applications plus importantes et plus d'intervalles statistiques sont introduits
[10]
dans de nombreux ouvrages tels que Hahn et Meeker .
Par contraste, pour les intervalles de dispersion dont il est question dans la présente partie de l'ISO 16269, le
niveau de confiance pour l'estimateur d'intervalle et la proportion de distribution dans l'intervalle
(correspondant à la fraction conforme mentionnée ci-dessus) sont fixés à l'avance, et les limites sont estimées.
© ISO 2005 – Tous droits réservés v
---------------------- Page: 5 ----------------------
ISO 16269-6:2005(F)
Ces limites peuvent être comparées à U et à L. Ainsi la justesse des limites de tolérance données U et L peut
être comparée aux propriétés réelles du processus. Les intervalles de dispersion unilatéraux sont utilisés
uniquement lorsque la limite de tolérance supérieure, U, ou la limite de tolérance inférieure, L, est appropriée,
tandis que les intervalles bilatéraux sont utilisés lorsque les limites supérieure et inférieure sont prises en
compte simultanément.
La terminologie relative à ces limites et intervalles différents est confuse car les «limites de tolérance» étaient
également autrefois appelées «limites de dispersion» (voir la Norme de terminologie ISO 3534-2:1993, 1.4.3,
où ces deux termes, mais aussi le terme «valeurs limites», étaient utilisés comme synonymes pour désigner
ce concept). Dans la dernière révision de l'ISO 3534-2:—, seul le terme «limites de tolérance» a été conservé
[5]
pour désigner ce concept. En outre, le Guide pour l'expression de l'incertitude de mesure (GUM) utilise le
terme «facteur d'élargissement», défini comme un «facteur numérique utilisé comme multiplicateur de
l'incertitude-type composée pour obtenir l'incertitude élargie». Cette utilisation du terme «élargissement» est
différente de celle de la présente partie de l'ISO 16269.
vi © ISO 2005 – Tous droits réservés
---------------------- Page: 6 ----------------------
NORME INTERNATIONALE ISO 16269-6:2005(F)
Interprétation statistique des données —
Partie 6:
Détermination des intervalles statistiques de dispersion
1 Domaine d'application
La présente partie de l'ISO 16269 décrit des méthodes permettant d'établir les intervalles statistiques de
dispersion qui comprennent au moins une proportion spécifiée de la population avec un niveau de confiance
spécifié. Des intervalles statistiques de dispersion unilatéraux et bilatéraux sont fournis, l'intervalle statistique
de dispersion unilatéral étant caractérisé par une limite supérieure ou par une limite inférieure, tandis que
l'intervalle statistique bilatéral possède à la fois une limite supérieure et une limite inférieure. Deux méthodes
sont exposées: une méthode paramétrique, lorsque la caractéristique étudiée a une distribution normale, et
une méthode non paramétrique, lorsque rien n'est connu de la distribution si ce n'est qu'elle est continue.
2 Références normatives
Les documents de référence suivants sont indispensables pour l'application du présent document. Pour les
références datées, seule l'édition citée s'applique. Pour les références non datées, la dernière édition du
document de référence s'applique (y compris les éventuels amendements).
ISO 3534-1, Statistique — Vocabulaire et symboles — Partie 1: Probabilité et termes statistiques généraux
1)
ISO 3534-2:— , Statistique — Vocabulaire et symboles — Partie 2: Statistique appliquée
3 Termes, définitions et symboles
3.1 Termes et définitions
Pour les besoins du présent document, les termes et définitions donnés dans l'ISO 3534-1 et l'ISO 3534-2
ainsi que les suivants s'appliquent.
3.1.1
intervalle statistique de dispersion
intervalle déterminé à partir d'un échantillon prélevé au hasard de manière qu'à un niveau de confiance
spécifié, l'intervalle couvre au moins une proportion spécifiée de la population échantillonnée
NOTE Dans ce contexte, le niveau de confiance est la proportion à long terme d'intervalles construits de cette
manière qui comprendra au moins la proportion également donnée de la population échantillonnée.
3.1.2
limite statistique de dispersion
statistique représentant un point limite d'un intervalle statistique de dispersion
NOTE Les intervalles statistiques de dispersion peuvent être soit unilatéraux, auquel cas ils ont une limite statistique
de dispersion supérieure ou inférieure, soit bilatéraux, auquel cas ils possèdent les deux limites.
1) À publier. (Révision de l'ISO 3534-2:1993)
© ISO 2005 – Tous droits réservés 1
---------------------- Page: 7 ----------------------
ISO 16269-6:2005(F)
3.1.3
élargissement
proportion des individus d'une population se trouvant dans un intervalle statistique de dispersion
NOTE Ce concept est à ne pas confondre avec le concept de facteur d'élargissement utilisé dans le Guide pour
[5]
l'expression de l'incertitude de mesure (GUM) .
3.1.4
population normale
population distribuée normalement
3.2 Symboles
Pour les besoins de la présente partie de l'ISO 16269, les symboles suivants s'appliquent.
i suffixe d'une observation
k (n; p; 1 − α) facteur utilisé pour déterminer x ou x lorsque la valeur de σ est connue pour un intervalle
1 L U
de dispersion unilatéral
k (n; p; 1 − α) facteur utilisé pour déterminer x et x lorsque la valeur de σ est connue pour un intervalle de
2 L U
dispersion bilatéral
k (n; p; 1 − α) facteur utilisé pour déterminer x ou x lorsque la valeur de σ est inconnue pour un intervalle
3 L U
de dispersion unilatéral
k (n; p; 1 − α) facteur utilisé pour déterminer x et x lorsque la valeur de σ est inconnue pour un intervalle
4 L U
de dispersion bilatéral
n nombre d'observations dans l'échantillon
p proportion minimale de la population déclarée comme se trouvant dans l'intervalle statistique
de dispersion
u fractile d'ordre p de la distribution normale réduite
p
ème
x i valeur observée (i = 1, 2, …, n)
i
x valeur maximale des valeurs observées, x = max {x , x , …, x }
max max 1 2 n
x valeur minimale des valeurs observées, x = min {x , x , …, x }
min min 1 2 n
x limite inférieure de l'intervalle statistique de dispersion
L
x limite supérieure de l'intervalle statistique de dispersion
U
n
1
x moyenne de l'échantillon, xx=
∑ i
n
i = 1
2
nn
2
nx − x
∑∑ii
n
1 ii==11
2
s écart-type de l'échantillon; sx x
=−()=
∑ i
nn−−11n
()
i = 1
2 © ISO 2005 – Tous droits réservés
---------------------- Page: 8 ----------------------
ISO 16269-6:2005(F)
1 − α niveau de confiance de la déclaration selon laquelle la proportion de la population se
trouvant dans l'intervalle de dispersion est supérieure ou égale au niveau spécifié p
µ moyenne de la population
σ écart-type de la population
4 Méthodes
4.1 Population normale avec une variance et une moyenne connues
2
Lorsque les valeurs de la moyenne, µ, et de la variance, σ , d'une population normalement distribuée sont
connues, la distribution de la caractéristique étudiée est complètement déterminée. Il y a exactement une
proportion p de la population:
a) à la droite de x = µ − u × σ (intervalle unilatéral);
L p
b) à la gauche de x = µ + u × σ (intervalle unilatéral);
U p
c) entre x = µ − u × σ et x = µ + u × σ (intervalle bilatéral).
L (1 + p)/2 U (1 + p)/2
NOTE Dans la mesure où l'on sait que ces déclarations sont justes, elles sont faites avec un niveau de confiance de
100 %.
Dans les équations ci-dessus, u est le fractile d'ordre p de la distribution normale réduite. Les valeurs
p
numériques de u sont indiquées aux dernières lignes des Tableaux B.1 à B.6 et C.1 à C.6.
p
4.2 Population normale avec une variance connue et une moyenne inconnue
Les formulaires A et B, donnés dans l'Annexe A, sont applicables lorsque la variance de la population
normale est connue alors que la moyenne est inconnue. Le formulaire A s'applique aux cas unilatéraux tandis
que le formulaire B s'applique aux cas bilatéraux.
4.3 Population normale avec une variance et une moyenne inconnues
Les formulaires C et D, donnés dans l'Annexe A, sont applicables lorsque la moyenne et la variance de la
population normale sont inconnues. Le formulaire C s'applique aux cas unilatéraux tandis que le formulaire D
s'applique aux cas bilatéraux.
4.4 Distribution continue quelconque de type inconnu
Si la caractéristique à l'étude est une variable continue provenant d'une population de forme inconnue et si un
échantillon de n observations aléatoires et indépendantes de la caractéristique a été prélevé, alors un
intervalle statistique de dispersion peut être déterminé à partir des observations ordonnées. La méthode
indiquée aux formulaires E et F de l'Annexe A permet de déterminer l'élargissement ou l'effectif de
l'échantillon nécessaires aux intervalles de dispersion déterminés à partir des valeurs extrêmes x et x
min max
de l'échantillon d'observations avec un niveau de confiance de 1 − α.
NOTE Les intervalles statistiques de dispersion qui ne sont pas fonction de la forme de la population échantillonnée
sont appelés intervalles de dispersion non paramétriques.
La présente partie de l'ISO 16269 ne préconise pas de méthode pour les distributions d'un type connu autre
que la distribution normale. Toutefois, si la distribution est continue, la méthode non paramétrique peut être
utilisée. Une sélection de références à de la littérature scientifique pouvant aider à déterminer les intervalles
de dispersion pour d'autres distributions est aussi fournie à la fin de ce document.
© ISO 2005 – Tous droits réservés 3
---------------------- Page: 9 ----------------------
ISO 16269-6:2005(F)
5 Exemples
5.1 Données
Les formulaires A à D, donnés dans l'Annexe A, sont illustrés par des exemples à l'aide des valeurs
numériques de l'ISO 2854:1976, Article 2, paragraphe 1 des remarques introductives, Tableau X, fil 2:
12 mesures de la charge de rupture du fil en coton. Il convient de noter que le nombre d'observations, n = 12,
[1]
indiqué pour ces exemples, est considérablement inférieur à celui recommandé dans l'ISO 2602 . L'unité de
mesure pour exprimer les données numériques et les calculs dans les différents exemples est le centinewton
(voir Tableau 1).
Tableau 1 — Données pour les Exemples 1 à 4
Valeurs en centinewtons
x 228,6 232,7 238,8 317,2 315,8 275,1 222,2 236,7 224,7 251,2 210,4 270,7
Ces mesures proviennent d'un lot de 12 000 bobines, d'une même série de fabrication, emballées dans
120 boîtes contenant chacune 100 bobines. Douze boîtes de ce lot ont été prélevées au hasard et une bobine
a été prise au hasard dans chacune de ces boîtes. Des éprouvettes de 50 cm de long ont été découpées
dans le fil de ces bobines, à environ 5 m de l'extrémité libre. Les essais proprement dits ont été réalisés sur
les parties centrales de ces éprouvettes. Des informations antérieures permettent de penser raisonnablement
que les charges de rupture mesurées dans ces conditions ont une distribution pratiquement normale. Il est
démontré, dans l'ISO 2954:1976, que les données ne contredisent pas l'hypothèse d'une distribution normale.
Les résultats suivants sont produits:
Effectif de l'échantillon: n = 12
Moyenne
de l'échantillon: x/==3 024,1 12 252,01
22
nx −()x
166 772,27
∑∑
Écart-type de l'échantillon: s== 1= 263,4263=35,545
nn (1−×) 12 11
La présentation formelle des calculs sera donnée uniquement pour le formulaire C de l'Annexe A (intervalle
unilatéral, variance inconnue).
5.2 Exemple 1: intervalle statistique de dispersion unilatéral sous variance connue
Supposer que les mesures obtenues précédemment montrent que la dispersion est constante d'un lot à
l'autre, provenant du même fournisseur, et qu'elle est représentée par un écart-type σ = 33,150, bien que la
moyenne ne soit pas constante. Une limite x est requise de manière à ce qu'il soit possible d'affirmer avec un
L
niveau de confiance 1 − α = 0,95 (95 %) qu'au moins 0,95 (95 %) des charges de rupture des individus du lot,
mesurées dans les mêmes conditions, sont supérieures à x .
L
La Tableau B.4 donne
k (12; 0,95; 0,95) = 2,120
1
d'où
xx=−k (n; ; p 1−ασ)× =252,01−2,120×33,150=181,732
L 1
4 © ISO 2005 – Tous droits réservés
---------------------- Page: 10 ----------------------
ISO 16269-6:2005(F)
Une plus petite valeur de la limite inférieure, x , serait obtenue si une proportion plus importante de la
L
population (par exemple, p = 0,99) et/ou un niveau de confiance supérieur (par exemple 1 − α = 0,99) étaient
requis.
5.3 Exemple 2: intervalle statistique de dispersion bilatéral sous variance connue
Dans des conditions identiques à celles de l'Exemple 1, supposer que les limites x et x sont requises de
L U
manière qu'il soit possible d'affirmer avec un niveau de confiance 1 − α = 0,95 qu'au moins une proportion de
p = 0,90 (90 %) des charges de rupture du lot se situe entre x et x .
L U
Le Tableau C.4 donne
k (12; 0,90; 0,95) = 1,889
2
d'où
xx=−k (n;p;1−ασ)× = 252,01−1,889× 33,150= 189,390
L 2
xx=+k (n;p;1−ασ)× = 252,01+1,889× 33,150= 314,630
U 2
Il convient que la comparaison de cet exemple avec l'Exemple 1 fasse bien comprendre la différence entre le
fait de garantir qu'au moins 90 % d'une population se situe entre les limites x et x et le fait de garantir qu'au
L U
plus 5 % se trouve au-delà de ces limites.
5.4 Exemple 3: intervalle statistique de dispersion unilatéral sous variance inconnue
Il est ici supposé que l'écart-type de la population est inconnu et doit être estimé à partir de l'échantillon. Les
mêmes exigences seront supposées que pour le cas où l'écart-type est connu (Exemple 1); ainsi, p = 0,95 et
1 − α = 0,95. La présentation détaillée des résultats est donnée ci-après.
© ISO 2005 – Tous droits réservés 5
---------------------- Page: 11 ----------------------
ISO 16269-6:2005(F)
Détermination de l'intervalle statistique de dispersion de la proportion p
a) Intervalle unilatéral «à droite»
Valeurs déterminées:
b) proportion de la population choisie pour l'intervalle de dispersion: p = 0,95
c) niveau de confiance choisi: 1 − α = 0,95
d) effectif de l'échantillon: n = 12
Valeur du facteur de dispersion, provenant du Tableau D.4:
k (n; p; 1 − α) = 2,737
3
Calculs:
xx==/n 252,01
∑
22
nx −()x
∑∑
s== 35,545
nn(1−)
k (n; p; 1 − α) × s = 97,286 7
3
Résultats: intervalle unilatéral «à droite»
L'intervalle de dispersion qui contiendra au moins une proportion p de la population avec un niveau de
confiance 1 − α a une limite inférieure:
xx=−k n; ; p 1−α×s=154,723
( )
L 3
5.5 Exemple 4: intervalle statistique de dispersion bilatéral sous variance inconnue
Dans les mêmes conditions que dans l'Exemple 2, supposer qu'il est requis de calculer les limites x et x de
L U
manière qu'il soit possible d'affirmer avec un niveau de confiance 1 − α = 0,95 que dans une proportion du lot
au moins égale à p = 0,90 (90 %), la charge de rupture est comprise entre x et x .
L U
La Tableau E.4 donne
kn( ; ;p 1−=α) 2,671
4
d'où
xx=−k (n; ; p 1− α)×s=252,01−2,671×35,545=157,069
L 4
xx=+k n; ; p 1−α×s=252,01+2,671×35,545=346,951
( )
U 4
Noter que la valeur de x est inférieure et que la valeur de x est supérieure à celles de l'Exemple 2 (variance
L U
connue) parce que l'utilisation de s à la place de σ nécessite une valeur de la constante de dispersion plus
importante pour tenir compte de l'incertitude supplémentaire. Il est nécessaire d'assumer le fait de ne pas
connaître l'écart-type de la population, σ; et l'extension de l'intervalle statistique de dispersion prend cela en
6 © ISO 2005 – Tous droits réservés
---------------------- Page: 12 ----------------------
ISO 16269-6:2005(F)
compte. Bien sûr, il n'est pas certain que la valeur σ = 33,150 utilisée dans les Exemples 1 et 2 soit correcte.
Donc, il est plus sage d'utiliser l'estimation, s, avec les Tableaux D.4 ou E.4.
5.6 Exemple 5: intervalle statistique de dispersion non paramétrique pour une distribution
continue
Lors d'un essai de fatigue par flexions rotatives, réalisé sur un composant d'un engin aéronautique, un
échantillon de 15 individus a donné les résultats du Tableau 2 (mesurage de l'endurance), classés par ordre
croissant.
Tableau 2 — Données de l'Exemple 5
x 0,200 0,330 0,450 0,490 0,780 0,920 0,950 0,970 1,040 1,710 2,220 2,275 3,650 7,000 8,800
Un examen graphique permettant de vérifier la normalité, comme une courbe de probabilité, montre qu'il
convient quasi certainement de rejeter l'hypothèse de normalité de la population de composants (voir
l'ISO 5479). Les méthodes du formulaire E, donné dans l'Annexe A, pour la détermination de l'intervalle
stat
...
Questions, Comments and Discussion
Ask us and Technical Secretary will try to provide an answer. You can facilitate discussion about the standard in here.