You are on page 1of 74

NB-CPD/SG10/12/091

GNB-CPD Guidance from the Group of Notified Bodies Issued: 9 March 2012
for the Construction Products Directive
SG10 89/106/EEC
APPROVED -
GUIDANCE

GNB-CPD position paper from SG10 - EN 771-1 to EN 771-6


Evaluation of conformity for masonry units

General scope, limitations and aim of this guidance for notified bodies
This position paper contains guidance for notified bodies (NBs) involved in the attestation of
conformity of FPC of masonry units according to EN 771-1 to EN 771-6. The purpose is to help
NBs work equivalently and come to common judgments. This guidance contains informative
material (which NBs should or may follow) and/or normative guidance (which NBs shall follow or at
least work equivalently to as circumstances demand).

The primary document for NBs is the edition of the relevant harmonized standard that is currently
cited in the Official Journal of the EU to which the manufacturer works. This guidance is thought
necessary to provide clarity and completeness for NBs so that they can work equivalently. It
supplements and makes practical for NBs the harmonized standards EN 771-1 to EN 771-6,
approved AG guidance, and Standing Committee guidance in the form of GPs, which also apply -
unless otherwise explicitly stated in this guidance. This position paper should not contradict nor
extend the scope of the work and role of a NB, nor impose additional burdens on the manufacturer,
beyond those laid down in the CPD and EN 771-1 to EN 771-6.

This guidance should be considered valid until the relevant standards are amended to include the
guidance (as thought fit by the CEN/TC); or until guidance from Commission, SCC or AG has
changed on relevant matters. Whereupon, the paper should be considered for withdrawal/revision
and be replaced by new guidance as necessary.

This position paper was considered approved by SG10 on 30 September 2011 and by Advisory
Group on 22 February 2012.

This position paper was developed primarily by CEN/TC 125 Masonry, to offer a statistical method
for the evaluation of conformity of masonry units. It has been published as a SG10 position paper to
enable it to be made available as soon as possible. It is expected that it will subsequently be
published as a CEN Technical Report.

NB-CPD/SG10/12/091 Page 1 of 74
Contents
1 Foreword ..................................................................................................................................................... 3
2 Symbols ...................................................................................................................................................... 3
3 Reference list .............................................................................................................................................. 4
4 General ....................................................................................................................................................... 4
5 Factory production control .......................................................................................................................... 5
5.1 General ................................................................................................................................................................. 5
5.2 Testing and measuring equipment ....................................................................................................................... 6
5.3 Production equipment ........................................................................................................................................... 6
5.4 Raw materials ....................................................................................................................................................... 6
5.5 Production process ............................................................................................................................................... 7
5.6 Finished product testing ....................................................................................................................................... 8
5.6.1 Inspection lot ............................................................................................................................................................................ 9
5.6.2 Spot sampling and sample sizes .............................................................................................................................................. 9
5.6.3 Production types ..................................................................................................................................................................... 11
5.6.4 Method A: Batch control ......................................................................................................................................................... 11
5.6.5 Method B: Rolling inspection ............................................................................................................................................... 12
5.6.6 Evaluation of test results ........................................................................................................................................................ 14
5.6.7 How to come from unknown to known standard deviation? ................................................................................................... 18
5.6.8 Conformity .............................................................................................................................................................................. 19
5.6.9 A simple and conservative approach ..................................................................................................................................... 24
5.6.10 Non-conforming products ....................................................................................................................................................... 24
5.6.11 Guidance ................................................................................................................................................................................ 25
5.6.12 Records .................................................................................................................................................................................. 28

6 Initial type tests ......................................................................................................................................... 28


Annex A Tables for acceptance coefficient kn depending on the used fractile p and confidence level (taken
from ISO 16269-6 (2005))............................................................................................................... 30
Annex B Examples of statistical evaluation ................................................................................................... 46

NB-CPD/SG10/12/091 Page 2 of 74
1 Foreword

By agreement with CEN/ TC 125, SG10 has prepared this position paper to have a tool available for
Notified Bodies (NBs) and manufacturers.

It is laid down in the hENs that the manufacturer shall demonstrate compliance for his product with
the requirements of the harmonised standards EN 771-1 to EN 771-6.

The purpose of this guidance document is to put statistical evaluation into practice. It can be used
for the evaluation of different properties at the different stages of the FPC with the aim to minimise
testing costs for the manufacturer and to ensure that the requirements are fulfilled. Detailed
examples are given in the Annexes.

To maintain equivalent use and interpretation of this document, notified bodies are strongly invited
to raise any questions, remarks or problems related to the use of this document with the secretariat
of the NB-CPD/SG10. The address of the secretariat can be found in GNB-CPD Monitoring report
NB-CPD/M02 Officials of the GNB-CPD. At the time of writing, this can be found on the CIRCA
website in folder http://circa.europa.eu/Members/irc/nbg/cdpgnb/library?l=/monitoring_gnb-
cpd&vm=detailed&sb=Title, but GNB-CPD information is expected to be transferred to an area of
the CIRCABC website.

2 Symbols

kn is the acceptance coefficient


k1 is the acceptance coefficient one-sided tolerance interval
k2 is the acceptance coefficient two-sided tolerance interval
kc is the corrected acceptance coefficient
kk is the acceptance coefficient for known standard deviation
ku is the acceptance coefficient for unknown standard deviation
n is the number of test samples within the spot sample
xm is the mean test result
xi is the test result for test sample i
i is the number of the individual test sample
xest is the estimated test result of the spot sample
s is the standard deviation of the test results
ss is the standard deviation of the test results of a spot sample
is the known standard deviation
l is the number of inspection lots
10,dry,unit is the thermal conductivity of the unit
p is the fractile
is the confidence level

NB-CPD/SG10/12/091 Page 3 of 74
3 Reference list

EN 771-1:2011 Specification for masonry units - Part :1 Clay masonry units

EN 771-2:2011 Specification for masonry units - Part 2: Calcium silicate masonry units

EN 771-3:2011 Specification for masonry units - Part 3: Aggregate concrete masonry


units (dense and lightweight aggregates)

EN 771-4:2011 Specification for masonry units - Part 4: Autoclaved aerated concrete


masonry units

EN 771-5:2011 Specification for masonry units Part 5: Manufactured stone masonry


units

EN 771-6:2011 Specification for masonry units Part 6: Natural stone masonry units

EN 1990:2002/A1:2005 Eurocode - Basis of structural design

EN 1996-1-1:2005 Eurocode 6: Design of masonry structures - Part 1-1: General rules for
reinforced and unreinforced masonry structures

EN 1996-1-2:2005 Eurocode 6: Design of masonry structures - Part 1-2: General rules -


Structural fire design

EN 1996-2:2006 Eurocode 6: Design of masonry structures - Part 2: Design


considerations, selection of materials and execution of masonry

EN 1996-3:2006 Eurocode 6: Design of masonry structures - Part 3: Simplified calculation


methods for unreinforced masonry structures

4 General

It is specified in the EN 771 series that the manufacturer shall demonstrate compliance for his
product with the requirements of the relevant European Standard and with the declared values for
the product properties by carrying out both:
initial type testing of the product (ITT);
factory production control (FPC).

If the manufacturer intends to declare that the units are Category I units, then the units have to fulfil
the definition of Category I units which is Units with a declared compressive strength with a
probability of failure to reach it not exceeding 5 %, which means that the manufacturer is declaring
that the customer can be 95 % confident that the delivered units fulfilled the declared compressive
strength. To be able to demonstrate this it is necessary for the manufacturer to operate a FPC that
includes a statistical evaluation.

The confidence level for a property has to be fixed depending on how important the property is in a
building. The higher the confidence level is the lower is the risk that the product does not fulfil the
declared values. When dealing with the safety of a building it is necessary to presuppose a
minimum confidence level fulfilled by the used products, otherwise the partial safety factors cannot
be fixed.

NB-CPD/SG10/12/091 Page 4 of 74
It is not possible to operate with a 100 % confidence level for a property to be tested by a
destructive test, and for properties tested by a non-destructive test it will be too expensive to
operate with a 100 % confidence level. A confidence level of 95 % is very high and considered
more acceptable.

Confidence levels other than 95 % can be used, e.g. the safety system specified in the Eurocode,
EN 1990, to which the Eurocode for masonry (EN 1996) refers for safety aspects, is based on the
assumption that declared values for the used product properties fulfil a confidence level of 75 %.

For characteristics, where a certain minimum confidence level is not fixed in a technical specification
or in a contract to be fulfilled, the manufacturer is free to fix the confidence level he will operate with,
and the higher the chosen level is, the lower is the risk that the manufacturer is running that the
delivered products do not fulfil the declared values. The risk the manufacturer is running is fixed by
a combination of the actual variation in test results over time, the frequencies of checking and
testing, the way the FPC system is developed and how close the declared value is to the tested
values.

In the product standard the conformity criteria are related to a consignment, that is a delivery to a
building site. The product standard defines a declared value as a value that the manufacturer is
confident in achieving, bearing in mind the precision of test and the variability of the production
process, and when the declared values are accompanying the product to the building site, they are
valid for the delivered consignment. Since it is impractical to test each consignment the
manufacturer has to plan the FPC system in such a way that the effect of the variations of product
characteristics during the production is taken into account when declaring the characteristics for the
consignment. In some production processes products are naturally separated into batches and a
consignment is quite often only a part of a batch. If a production is based on a continuous flow a
consignment is only a part of the continuous production.

5 Factory production control

5.1 General

The factory production control (FPC) system may be developed in such a way that the checking
procedures are:
mainly related to the process only (full process control and consequently only a small amount
of finished product testing), or;
mainly related to the finished products only (and consequently limited process control), or;
any combination of both.

It may even be so that the amount of process control and finished product testing varies depending
on the property to be assessed. If the test for the property is low cost, e.g. test of dimensions, and
if the property is less important in relation to the end use then it may be the right solution to use
finished product testing. But if the testing of the property is expensive, e.g. frost resistance tests,
then the solution may be to base the assessment on process control using proxy tests.

In some companies responsibility for the production is placed only on one person, and if this person
is not available, the responsibility for taking decisions is unclear. This can result in unnecessary and
costly stops of the production or the manufacture of non-conforming products. It should be in the
interest of the manufacturer to avoid this by establishing the responsibility, authority and interrelation

NB-CPD/SG10/12/091 Page 5 of 74
of all personnel who manage, perform and verify the work affecting the quality of masonry unit
products and the evaluation of conformity.

The procedures to be followed when controlling the production are of course of great importance as
the quality of the products is directly linked to that. It should be in the interest of the manufacturer to
obtain the best quality of the products and therefore to have an interest in clear procedures. The
best way of achieving this is to have them in a written form. Procedures for what to do, when
control and check parameters during the production are not obtained or fulfilled, are of the same or
may be of greater importance. Therefore the need for having them in a written form is crucial.

The manufacturer may define product groups. A product group consists of products from one
manufacturer having common values for one or more characteristics. That means that the products
belonging to a product group may differ according to the characteristics in question. If a product
group is defined, then the FPC system shall ensure that all types of units within a group are
controlled and over time also in the finished product testing, if that is part of the FPC.

Depending on the way the FPC system is developed (process control only, finished product testing
only or a combination of both) a selection of these may be considered.

5.2 Testing and measuring equipment

The accuracy of the testing or measuring equipment used in the control procedures are to be in
accordance with the test standard. If it is not defined there, then a rule of thumb can be 1/5 1/10
times of the accuracy of the value to be declared. Testing or measuring data are not helpful in itself,
unless you know that the data are accurate. It should be in the interest of the manufacturer to know
that testing and measuring data are reliable. To obtain that, all relevant weighing, measuring and
testing equipment that have an influence on the declared values, need to be verified and regularly
inspected.

A verification of testing and measuring equipment needs only to be done in the measuring area
used. If the length of a unit is 300 mm, then the measuring area for the length is approximately 290
310 mm and can be verified using a fixed measuring length, e.g. iron prism, iron block or iron bar
with a length of 300 mm. Weighing equipment can be verified by the manufacturer using fixed
weights covering the weighing area used.

5.3 Production equipment

Most production equipment contains moving parts, which need adjustment from time to time.
During production wear and tear can also happen. For that reason, it is recommended that all parts
of production equipment that have an influence on the declared values are to be controlled and
regularly inspected.

5.4 Raw materials

The product properties depend on the constituents used and variations in their quality. To eliminate
this influence as much as possible the manufacturer has to define his own acceptance criteria of
raw materials and the procedures with which to operate to ensure that these are met. This is
independent of the way the constituents are received in the factory bought from a supplier or
delivered from the manufacturers own sources. If the constituents or some of them are bought from
a supplier, the manufacturer is advised to be sure that the control system for the constituents carried
out by the supplier is sufficient. Normally it is acceptable if the control system of the supplier is

NB-CPD/SG10/12/091 Page 6 of 74
supervised by a third party, and then the manufacturer has only to check the delivery notes and
make a visual inspection to ensure that the delivery is in line with the order. If the raw materials are
delivered from the manufacturers own sources, for example the manufacturers own clay pit, then a
procedure to check, if the grain size distribution of the clay is kept constant, could be to measure
regularly the amount of clay in a test sample passing a 90 m test sieve. An example of control
data is given in Figure 1 along with the acceptance criteria fixed by the manufacturer, the upper limit
(UL) and lower limit (LL).

Figure 1. Example of variation in the amount of clay particles passing a 90 m test sieve

5.5 Production process

The production process and the controlling of production are of great importance for the properties
of the products and variation in the properties. It should be in the interest of the manufacturer to
obtain the best quality of the products and therefore to want to have the best handling of the
production. The best way of achieving this is to identify relevant measuring and check parameters
in the process, and then to fix for each parameter requirements to be fulfilled or limits (upper and
lower limits, UL and LL) between which the parameter is allowed to vary. These limits and the
frequency of measuring or checking the parameter have to be based on the manufacturers
experience and on the importance and the variation of the parameter. The manufacturer should
also specify what should be done, when control and check parameters during the production are not
fulfilling the requirement or passing the limit value.

In the following example, Figure 2, the length of the green clay masonry units is measured to control
the wear and tear of the mould in which the units are produced. In the following part of the

NB-CPD/SG10/12/091 Page 7 of 74
production process the units will shrink 0,1 mm, and the intension is to declare a length of 228,5 mm
and a tolerance of 0,5 mm. Both aspects need to be taken into account when fixing the control
limits. The reason for the dramatic drop is a renewal of the mould. The renewal of the mould
should have taken place at spot sample 11 as it was leading to a situation where all units in the
inspection lot produced between spot sample 11 and 12 did not conform to the fixed upper limit.

Figure 2. Example of variation in the length of green clay masonry units over time

It is possible to operate with two sets of control limits, a narrow and a wider range. If the parameter
is passing the control limit of the narrow range, it can be looked upon as a warning, and a small
correction of the process may be made, but when the parameter passes the control limit for the
wider range, a more radical correction of the process will be needed.

5.6 Finished product testing

When testing the finished product, it is possible to use alternative test methods if a correlation can
be established between the alternative test method and the reference test method.

It is also important to notice that a test result of a spot sample (see clause 5.6.2) is representing an
inspection lot (see clause 5.6.1). If an evaluated test result is not conforming, the whole production
since the last test has to be looked upon as non-conforming. For that reason it can be
recommended, that for properties where the reference test is time consuming and may be costly,
alternative tests or proxy tests that are less time consuming and costly are used. By doing so the
time span between the tests can be shortened and the amount of products covered by a non-
conforming test result will be less and thereby reduce the manufacturers risk.

NB-CPD/SG10/12/091 Page 8 of 74
The amount of products produced between 2 tests is an inspection lot. The frequency of testing can
vary from one property to another and thereby the inspection lot can vary from one property to
another.

5.6.1 Inspection lot

The production is divided into inspection lots.

An inspection lot must consist of units produced under uniform conditions:


same raw materials;
same dimensions;
same production process.

If a certain characteristic is the same for multiple units, where the dimension has no influence, these
units can belong to the same product family.

That means that an inspection lot for the characteristic in question can only consist of products
belonging to the same product group.

The manufacturer decides on the size of the inspection lot from:


raw material mixing lots, or;
number/volume of units, or;
number of production days.

Independent of the way the size of the inspection lot has been decided, it must be possible to draw
a representative spot sample.

5.6.2 Spot sampling and sample sizes

When the inspection lot has been decided, the sampling procedure for a spot sample has to be fixed
in such a way that the spot sample is representative for the inspection lot.

Figure 3. An example of representative sampling

NB-CPD/SG10/12/091 Page 9 of 74
In the European Product Standard sampling procedures for stacks and banded packs are given. It
is also possible to sample from the conveyer belt or in the case of fired units after the kiln.

The number of units in the spot sample is decided by the manufacturer. If somewhere a minimum
number of units has been fixed then this must be accepted.

By deciding on the size of the inspection lot the manufacturer is fixing the frequencies of tests to be
done. The size of the inspection lot should be decided based on:
how close the declared value is to the test value;
the deviation of the test values;
how much process control is going on.

These decisions allow the manufacturer to manage his own risks.

In the following Figure 4 the variation over time for the mean compressive strength is given.

Figure 4. Example of variation in mean compressive strength over time

On the basis of the test results from testing the spot sample it has to be decided whether the
inspection lot is accepted or not, see clause 5.6.8. In this respect the test results can be dealt with
separately or treated together with the previous results. It depends on the type of production (batch
production or series production).

NB-CPD/SG10/12/091 Page 10 of 74
5.6.3 Production types

A production, which is naturally separated into batches, is named a batch production. In the case
of the batch production the properties of the units may change batch by batch. A batch is normally
looked upon as a separate inspection lot. If the process control minimises the changes from one
batch to another, an inspection lot can cover more than one batch

A production, which is based on a continuous flow, is named a series production. In the case of
series production the properties of the units are the same within a series. A series production
contains normally more than one inspection lot.

5.6.4 Method A: Batch control

When a batch production is in operation, then the FPC system needs to be based on a batch
control, which means, that each batch is controlled separately.

In clause 5.6.6 when dealing with the evaluation of test results the acceptance coefficient kn is given
in Tables 1 and 2. These tables show that there is a great difference in using kn for 3 or for 6 test
results and for that reason it is recommended to operate with spot sample sizes of at least 6 units.

NB-CPD/SG10/12/091 Page 11 of 74
Figure 5. Example of Method A: Each inspection lot is evaluated individually

5.6.5 Method B: Rolling inspection

In a series production there are a series of inspection lots, which should not exceed a total number
of 5. In the following 4 are used.

Figure 6. Example with 4 inspection lots in a series

For the 1st inspection lot a spot sample size of 3 is taken and tested. For the 2nd inspection lot 3
new samples are taken and tested and evaluated together with the ones from the 1st inspection lot
and therefore the spot sample size will be 6. For the 3rd inspection lot 3 new samples are taken and
tested and evaluated together with the ones from the 1st and the 2nd inspection lot and therefore the
spot sample size will be 9. For the 4th inspection lot 3 new samples are taken and tested and
evaluated together with the ones from the 1st, 2nd and 3rd inspection lot and therefore the spot
sample size will be 12. For the 5th inspection lot 3 new samples are taken and tested and evaluated

NB-CPD/SG10/12/091 Page 12 of 74
together with the ones from the 2nd, 3rd and 4th inspection lot and therefore the spot sample size will
be 12. The described rolling system will continue for the following inspection lots. The rolling
system is illustrated in the following Figure. In clause 5.6.6 when dealing with the evaluation of test
results the acceptance coefficient kn is given in Tables 1 and 2. These tables show that there is a
great difference for 6 and for 12 test results, and the number of tests to be done is half compared to
the batch control when the size of the inspection lot is the same. Another possibility is to half the
size of the inspection lot and therefore to reduce the number of units covered by non-conformity, if
that occurs.

Figure 7. Example of Method B, Rolling inspection: series of 4 inspection lots

Another possibility is the so-called progressive sampling procedure. For each of the 1st to 5th
inspection lots a spot size of one sample is taken and tested. These lots are evaluated together.
For the 6th and following inspection lots 1 additional sample is taken and tested and evaluated
together with the ones from the previous inspection lots. The spot size is gradually increased from 5
to 15 samples. From then on, 1 additional sample is taken from each next inspection lot but the
spot sample is limited to the last 15 samples. The spot sample size continues to be 15.

NB-CPD/SG10/12/091 Page 13 of 74
Figure 8. Example of Method B, Rolling inspection Progressive sampling: series of 15
inspection lots

5.6.6 Evaluation of test results

Where and when possible and applicable, the results of checks and testing shall be interpreted by
means of statistical techniques, by attributes or by variables to verify the product characteristics and
to determine if the production conforms to the compliance criteria and the products conform to the
declared values. One method of satisfying this conformity criterion is to use the approach given in
ISO 12491. This approach is shown in detail in this section.

When using the test results of a spot sample with a limited number of samples to estimate the
characteristics of the production there are some uncertainties. The deviation within the test results
is one uncertainty and, how representative the spot sample is for the production, is another
uncertainty. The first uncertainty is dealt with in the evaluation by taking into account the standard
deviation s of the test results of the spot sample. The second uncertainty is dealt with by using an
acceptance coefficient kn. The acceptance coefficient kn can be regarded as a factor minimising the
statistic uncertainties from spot sampling. kn is dependent on several factors:
The number of samples in the inspection lot n
The confidence level
The fractile p *)

NB-CPD/SG10/12/091 Page 14 of 74
The standard deviation is unknown. The symbol used is ku
The standard deviation is known. The symbol used is kk
One-sided limit evaluation. The symbol used is k1
Two-sided limit evaluation. The symbol used is k2

*) Be aware that a 5 % characteristic value corresponds with a fractile p = 95 and a 95 % characteristic value
corresponds also with a fractile p = 95. 50 % characteristic value corresponds with a fractile p = 50.

When evaluating the test results from a spot sample, then use the following procedure:

Calculate the mean value of the test results using the following equation:
n
xm =
1
n
x i (1)
i=1

where
xm is the mean test result
xi is the test result for test sample i
n is the number of test samples within the spot sample
i is the number of the individual test sample

Calculate the standard deviation ss for the test results of the spot sample using the following
equation:

(x xm )
2
i
i =1
s= (2)
n 1

where
s is the standard deviation for the test results
n is the number of test samples within the spot sample
i is the number of the individual test sample
xi is the test result for test sample i
xm is the mean test result

If the standard deviation is unknown and if the test results have to be compared with a lower limit
value then calculate the estimated test result xest using the following equation:

xest = xm k1,u ss (3)

If the standard deviation is unknown and if the test results have to be compared with an upper limit
value then calculate the estimated test result xest using the following equation:

xest = xm + k1,u ss (4)

NB-CPD/SG10/12/091 Page 15 of 74
If the standard deviation is unknown and if the test results have to be compared with a two-sided
limit value then calculate the estimated test result xest using the following equation:
xest = xm k2,u ss 5

If the standard deviation is known and if the test results have to be compared with a lower limit
value then calculate the estimated test result xest using the following equation:
xest = xm k1,k 6

If the standard deviation is known and if the test results have to be compared with an upper limit
value then calculate the estimated test result xest using the following equation:
xest = xm + k1,k 7

If the standard deviation is known and if the test results have to be compared with a two-sided
upper limit value then calculate the estimated test result xest using the following equation:
xest = xm k2,k (8)

where
xest is the estimated test result of the spot sample
xm is the mean test result
k1,u is the acceptance coefficient for unknown standard deviation and one-sided limit
evaluation to be taken from Table 1 or 2 or relevant tables in Annex A
k2,u is the acceptance coefficient for unknown standard deviation and two-sided limit
evaluation to be taken from relevant tables in Annex A
ss is the standard deviation for the test results of the spot sample
k1,k is the acceptance coefficient for known standard deviation and one-sided limit
evaluation to be taken from Table 1 or 2 or relevant tables in Annex A
k2,k is the acceptance coefficient for known standard deviation and two-sided limit
evaluation to be taken from relevant tables in Annex A
is the known standard deviation

Standard n=3 4 5 6 7 8 9 10 11 12 14 15
deviation

Unknown 1,69 1,18 0,95 0,82 0,74 0,67 0,62 0,58 0,55 0,52 0,47 0,46

Known 0,95 0,82 0,74 0,67 0,62 0,58 0,55 0,52 0,50 0,48 0,44 0,43

Table 1. kn for 50 % characteristic value (50 % fractile) and 95 % confidence level

NB-CPD/SG10/12/091 Page 16 of 74
Standard n=3 4 5 6 7 8 9 10 11 12 14 15
deviation

Unknown 7,66 5,14 4,20 3,71 3,40 3,19 3,03 2,91 2,82 2,74 2,62 2,57

Known 2,60 2,47 2,38 2,32 2,27 2,23 2,19 2,17 2,14 2,12 2,09 2,07

Table 2. kn for 5 % characteristic value (95 % fractile) and 95 % confidence level

More tables are given in Annex A.

The method of using the acceptance coefficient for known standard deviation kk is only valid when
the standard deviation ss of the spot sample corresponds to the following equation:
0,63 ss 1,37 (9)

If as part of the evaluation it turns out not to be the case, the manufacturer has to restart or he
decides to continue working with the unknown acceptance coefficient ku. This means that the
inspection lots have to be treated separately.

The effect of the size of the spot sample and the standard deviation of the test results of the sample
on the acceptance coefficient kn and the estimated compressive strength are shown in Table 3.

In the first example of Table 3 the spot sample representing an inspection lot consists of 6 units and
the results of the compressive strength are given on each unit. The mean value and the standard
deviation are calculated. From the table for kn for 50 % fractile and 95 % confidence level the
acceptance coefficient kn for unknown standard deviation and n = 6 are taken and the estimated
compressive strength for the inspection lot is calculated.

In the second example the spot sample size and the mean value are kept the same, but there is a
greater variation in the test results leading to a higher standard deviation, which again is leading to a
lower estimated compressive strength. A higher standard deviation is demonstrating less control
compared to the first example. When keeping the confidence level the estimated compressive
strength for the inspection lot needs to be lower.

In the third example the two previous spot samples are looked upon as one spot sample consisting
of 12 units. The mean value and the standard deviation are calculated. From Table 1 the
acceptance coefficient kn for unknown standard deviation and n = 12 are taken and the estimated
compressive strength for the lot is calculated. By enlarging the number of units to be tested of the
spot sample the estimated value is more certain leading to a higher estimated compressive strength
of the inspection lot compared to the second example, where the mean value and the standard
deviation are about the same.

NB-CPD/SG10/12/091 Page 17 of 74
Spot Mean Std. Coefficient Estimated
sample value deviation 95 %, comp.
size in MPa in MPa unknown strength in
kn MPa

6 20 1,3 0,82 19

6 20 3,2 0,82 17

12 20 3,0 0,52 18

Table 3. Example showing the effect of spot sample size and deviation

As you see, when reducing the variation in the test results by operating a better process control the
estimated value for the tested property will be higher. The same will be achieved by increasing the
number of units of the spot sample.

5.6.7 How to come from unknown to known standard deviation?

Looking at the tables for kn, Tables 1 and 2, it is clear, that there is a considerable effect in going
from an unknown to known standard deviation. In control method A (clause 5.6.4) the standard
deviation of the population is considered to be unknown at least for the first 40 test samples and the
acceptance coefficient ku has to be taken from tables for unknown standard deviation. For the next
80 test samples the standard deviation can be considered to be known, but the used acceptance
coefficient is corrected (kc). The acceptance coefficient for the known standard deviation kk is taken
from tables for known standard deviation. The corrected acceptance coefficient kc is calculated by a
linear interpolation between the acceptance coefficient ku and kk. The known standard deviation is
calculated based on the first at least 40 test results.

In control method B (clause 5.6.5) the standard deviation of the population is considered to be
unknown at least for the first 20 test samples and the acceptance coefficient ku has to be taken from
tables for unknown standard deviation. For the next 40 test samples the standard deviation can be
considered to be known, but the used acceptance coefficient is corrected (kc) as above. The
acceptance coefficient for the known standard deviation kk is taken from tables for known standard
deviation. The known standard deviation is calculated based on the first at least 20 test results.

NB-CPD/SG10/12/091 Page 18 of 74
If progressive sampling is used the standard deviation of the population is considered to be
unknown at least for the first 30 test samples and the acceptance coefficient ku has to be taken from
tables for unknown standard deviation. For the next 30 test samples the standard deviation can be
considered to be known, but the used acceptance coefficient is corrected (kc) as above. The
acceptance coefficient for the known standard deviation kk is taken from tables for known standard
deviation. The known standard deviation is calculated based on the first at least 30 test results.

5.6.8 Conformity

After calculating xest by testing the inspection lots the result has to be compared with either the
declared value or a lower or upper limit depending on the property. For compressive strength it is
the declared value or the lower limit and for dimension it is the upper and lower declared value or
the upper and lower limit. In Figure 8 the estimated mean compressive strength is based on 95 %
confidence level for the different spot samples using the calculations of the test data shown in
Figure 4. In Figure 9 the estimated 5 % characteristic compressive strength based on 95 %
confidence level is shown using the same test data.

UL

LL
DV

Figure 9. Example of variation in the estimated mean compressive strength over time

NB-CPD/SG10/12/091 Page 19 of 74
Figure 10. Example of variation in the estimated 5 % characteristic compressive strength
over time

In Figure 9 and 10 the estimated compressive strength is varying between the upper and lower limit
and therefore conforming to the fixed limit values. The declared value needs to be equal to or lower
than the lower limit value.

In Figure 11 the variation in the length of the units over time is given. The units are from the same
production as the ones checked as green units, see Figure 2.

NB-CPD/SG10/12/091 Page 20 of 74
Figure 11. Example of variation in the length of the finished units over time

As mentioned before the intention is to declare a length of 228,5 mm and a tolerance of 0,5 mm,
which means, that the upper and lower declared value is fixed by the tolerance. When the renewal
of the mould did not take place at the production spot sample 11, see Figure 2, then the length of
the units of spot sample 13 does not comply with the declared value and the belonging tolerance.

A manufacturer of units with a shape shown in Figure 12 would like to declare the thermal
conductivity, 10,dry,unit, of the unit.

Figure 12. Example of a shape of a masonry unit

NB-CPD/SG10/12/091 Page 21 of 74
By carrying out tests for masonry made of specific units it is possible for these units to establish a
relationship between the thermal conductivity, 10,dry,unit, and the gross dry density of the units as
shown in Figure 13.

Figure 13. Example of a relationship between the gross dry density and the thermal
conductivity of a unit

By testing and controlling the gross dry density it is possible to declare the thermal conductivity,
10,dry,unit, of the unit. The gross dry density is used as a proxy property for the thermal conductivity.

In Figure 14 the variation in the gross dry density over time is shown. The variation in the gross dry
density is coming from 2 contributions, variation in the shape and variation in the net dry density of
the material. When a dramatic drop occurs periodically the probable reason for the variation in the
gross dry density is a renewal of the mould and therefore the variation in the shape and not a
variation in the net dry density.

NB-CPD/SG10/12/091 Page 22 of 74
Figure 14. Example of variation in the gross dry density of the units over time

If the variation in the gross dry density is as shown in Figure 15 the reason seems to be the
variation in the shape as well as the variation in the net dry density.

If the declared thermal conductivity value has to be a 50 % fractile with a confidence level of 50 %
the test results of the spot samples have to be evaluated, e.g. by the calculation procedures
described in clause 5.6.6 using Table A1 or A5 in Annex A.

NB-CPD/SG10/12/091 Page 23 of 74
Figure 15. Example of variation in the gross dry density of the units over time

5.6.9 A simple and conservative approach

A simple and conservative approach can be to evaluate single test results of at least 1 year for a
given property and calculate the mean value and the standard deviation and fix then a band in
which new test results have to fit in. The upper band limit and lower band limit then can be 2 times
of the standard deviation away from the mean value. Then the declared value is recommended to
be 0,4 times the standard deviation away from the respective band limits. If non-conformity occurs
the evaluation of at least the last year of single test results including the non-conforming values shall
be repeated and the band limit values adjusted accordingly. The same shall happen for the
declared value. The non-conforming inspection lot can be treated as described in the next clause
using control method A.

5.6.10 Non-conforming products

When an evaluation of the test results of the last spot sample is leading to non-conformity, e.g. as
shown in Figure 11, it is important to avoid that the whole inspection lot is mixed up with the other
inspection lots. The non-conforming inspection lot has to be treated separately. It may be
reclassified by the manufacturer and given different declared values. If it is not segregated the
whole stock has to be treated as non-conforming. For that reason a procedure for dealing with non-
conforming products should be developed.

NB-CPD/SG10/12/091 Page 24 of 74
It should be in the interest of the manufacturer to avoid that the same non-conformity occurs again.
When non-conformity occurs, then it is important to try to identify the reason why, otherwise it is
difficult to find out, what to do to avoid that it occurs again. Testing can be part of the identification.

To ensure that the personnel managing the production knows what to do when check and
measuring values are passing the limit values, it is important to have the necessary instructions
documented.

Non-conformities will normally result in higher frequencies than the ones used. The background for
that is to reduce the size of the next batch that might also not comply.

5.6.11 Guidance

How to use the different possibilities?

A manufacturer is producing units in two different ways:


Product 1 is a special unit produced very rarely and only in small quantities. The
characteristics of the product may vary from production to production.
Product 2 is one of the core units of the production site. It is produced in series of variable
length sometimes only 2 days of production but it is produced within short-time intervals.

For product 1 it will be obvious to use control method A (batch control). For product 2 both
control methods A and B can be used. For product 2 it is even possible to use control method A
for some properties and for some properties control method B. If using method B a re-
declaration in connection with a non-conformity is possible based on test results obtained by
testing a new spot sample taken at random from the inspection lot following control method A,
but it is necessary to keep the test results leading to the non-conformity in the method B control
system when evaluating the next spot sample.

The following details may be used when planning the setup of the FPC system:

Control method A:
Verification of separate inspection lots.
Inspection lots are defined to be the full production series.
The minimum sample size of the spot sample is 6 units (n 6).
Level of confidence for compressive strength for Category I units is required to be 95 %.
For net dry density and dimension 75 % may be chosen. For gross dry density or net dry
density used as a proxy property to thermal conductivity a confidence level of 50 % or 90 %
may be chosen.
If the spot sample size is 6 units, the acceptance constant kn for mean compressive
strength at a 95 % confidence level is k1,u = 0,82 for unknown standard deviation and k1,k =
0,67 for known standard deviation.
If the spot sample size is 6 units, the acceptance constant kn for 5 % characteristic
compressive strength at a 95 % confidence level is k1,u = 3,71 for unknown standard
deviation and k1,k = 2,32 for known standard deviation.
If the spot sample size is 6 units, the acceptance constant kn for mean compressive
strength at a 75 % confidence level (Category II units) is k1,u = 0,30 for unknown standard
deviation and k1,k = 0,28 for known standard deviation.

NB-CPD/SG10/12/091 Page 25 of 74
Control method B:
Verification of series of inspection lots.
Inspection lot can be defined to be the units produced within 1 production week / 5 days.
The minimum sample size of the spot sample is 3 units (n 3).
Size of series are 4 inspection lots (l = 4).
In case of n = 3, the sample size used for evaluation of each inspection lot is 12.
Level of confidence for compressive strength for Category I units is required to be 95 %.
For net dry density and dimension 75 % may be chosen. For gross dry density used as a
proxy property to thermal conductivity a confidence level of 50 % or 90 % may be chosen.
If the spot sample size is 3 units, the acceptance constant kn for mean compressive
strength at a 95 % confidence level is k1,u = 0,52 for unknown standard deviation and k1,k =
0,47 for known standard deviation. If a sample size of a spot sample is raised to 6 units
instead of 3 then the acceptance constant kn for mean compressive strength is k1,u = 0,35
for unknown standard deviation and k1,k = 0,34 for known standard deviation.
If the spot sample size is 3 units, the acceptance constant kn for 5 % characteristic
compressive strength at a 95 % confidence level is k1,u = 2,74 for unknown standard
deviation and k1,k = 2,12 for known standard deviation. If a sample size of a spot sample is
raised to 6 units instead of 3 then the acceptance constant kn for mean compressive
strength is k1,u = 2,31 for unknown standard deviation and k1,k = 1,98 for known standard
deviation.
If the spot sample size is 3 units, the acceptance constant kn for mean compressive
strength at a 75 % confidence level (Category II units) is k1,u = 0,20 for unknown standard
deviation and k1,k = 0,19 for known standard deviation. If a sample size of a spot sample is
raised to 6 units instead of 3 then the acceptance constant kn for mean compressive
strength is k1,u = 0,14 for unknown standard deviation and k1,k = 0,14 for known standard
deviation.

What to do with an inspection lot where the evaluated test results for one or more properties
are leading to non-conformity?

Control method A:
Discard the inspection lot, or;
Sample a new and larger spot sample (e.g. 12 instead of 6), test the sample for the
properties leading to a non-conformity and evaluate the test results using a reduced
acceptance constant (e.g. 0,52 instead of 0,82) according to the higher number of units in
the test sample, or;
Change the declaration of the units based on ITT.

Control method B:
Discard the inspection lot, or;
Sample a new larger spot sample (e.g. 6 instead of 3 units) using control method A and
evaluate the test results using a reduced acceptance constant, according to the number of
the units in the test sample and change eventually the declaration accordingly. *)

*) Always keep the results of the inspection lot within the system when evaluating the next inspection lot
or start from the very beginning.

NB-CPD/SG10/12/091 Page 26 of 74
When a non-conformity is identified in the finished product testing it is not possible to take any
corrective actions for the tested inspection lot. It can only be discarded or re-declared. The longer
the production process of the units lasts, the larger is the number of units produced before it is
possible to correct the process, leading again to a larger number of units to be discarded or re-
declared. The example mentioned about measurement of the length of the green units, see Figure
2, demonstrates that it is possible to detect a problem (wear and tear of the mould) early in the
process, which leads to a non-conformity of the finished product in the process. Checking
dimensions, weights and temperatures are quite simple, but done at the right places in the process
they will give a lot of information valid for the control of the process and the properties of the finished
products. It may even be possible to counteract a detected problem later on in the process.

Consideration should be given to identifying the most economical way to arrange the control by the
right mix of process control and finished product testing, and to consider also the possibility of using
internal proxy tests in the process control.

The manufacturer may define product groups. A product group consists of products from one
manufacturer having common values for one or more characteristics. That means that the products
belonging to a product group may differ according to the characteristics in question. If a product
group is defined, then the FPC system shall ensure that all types of units within a group are
controlled and over time also by the finished product testing, if this is part of the FPC.

For process control the evaluation procedure described in clause 5.6.6 may be used, when
appropriate.

Traceability in the process

The clause deals with the traceability in the process from raw materials to finished products. It is
not dealing with the traceability on the market.

As mentioned earlier it should be in the interest of the manufacturer to avoid that the same non-
conformity occurring again. It is therefore important to try to identify the reason why, when it
occurred, otherwise it is difficult to find out, what to do to avoid it occurring again.

The better knowledge the manufacturer has about the variation in the raw materials, variation in the
different parts of the process and their influence on the variation in the properties of the finished
product the better he will be able to identify the reason for non-conformity. To be able to obtain that
knowledge it can be recommended that the manufacturer follows the same units all through the
process from time to time if not on every occasion and to evaluate all the checks and measurements
together and to compare the results with other similar evaluations done. Based on such an exercise
it may be possible to establish traceability in the process.

Marking and stock control of products

The more variations there are in the production in relation to the type of products and properties the
higher is the need for instructions dealing with the marking procedure and how to handle and to
control the stock. It is important that 2 types of units with the same shape but not the same
properties are marked in such a way that they will not be mixed up. Inspection lots of products
should be identifiable and traceable.

NB-CPD/SG10/12/091 Page 27 of 74
5.6.12 Records

Many years of experience have shown that it can be dangerous to have only one person who knows
all the information required for the production of masonry units and how to control it. The more this
information is in a written form the more it is available for others, and at the same time it is easier to
establish an overview in writing. It can be recommended to describe step by step what needs to be
done in the whole process from the raw material to the finished product leaving the gate of the
factory in order to be able to produce a high quality product. This can include specifying the position
of each check and observation points and control procedures. It is really valuable for the machine
operator to have information of corrective actions available when control parameters are passing the
control limits.

Experiences show also that on a busy day it is easy to forget important observations made during
the production control if these observations are not recorded. To make it easier to record
observations it can be recommended to use tables.

Samples are taken during the process and from finished products and these samples need to be
representative for the inspection lot. For that reason the sampling procedure is important and so
should be specified. When the frequency of testing is fixing the size of the inspection lot and
thereby the manufacturers risk the frequency should be carefully considered, decided and
recorded. If test results and FPC system give evidence of problems then the frequencies may be
reconsidered and reduced compared to the ones used.

6 Initial type tests

It is important for a manufacturer to produce what is possible to sell and not to try to sell what is
possible to produce. A manufacturer would like to fulfil the market needs and therefore intends to
develop and to produce units with specific properties. To ensure that these properties are available
it is necessary after completion of the development of a new product type and before
commencement of the manufacture and offering for sale, that appropriate initial type tests had been
carried out to confirm that the properties predicted from the development meet the requirements of
the product standard and the values to be declared for the unit.

If the manufacturer is trying to sell what is possible to produce and nothing else then the full finished
product test done as part of the control method A can act as an initial type test if the reference test
methods are used and the sampling procedure for ITT. In that respect the declared values, which
may vary from batch to batch have to be determined batch by batch and have to be based on an
evaluation of the same test results (see clause 5.6.6). It will not be possible to sell the units before
the test results are available.

If in control method B non-conformity occurs and the inspection lot is re-declared following control
method A using the reference test methods and the sampling procedure for ITT, then the test can
be regarded as an initial type test.

Whenever a major change in the source, blend, or nature of raw materials occurs, or when there is
a change in processing conditions, leading to what the manufacturer considers will constitute a new
product type being produced, the appropriate initial type test shall be repeated. If the manufacturer
has doubts it can be recommended to check whether some of the characteristics have changed or
not by using the FPC test procedures.

The manufacturer may define product groups. The products belonging to a product group may
differ according to the characteristics in question.

NB-CPD/SG10/12/091 Page 28 of 74
In the ITT process a manufacturer may take into consideration already existing test results.

A manufacturer may use the ITT results obtained by someone else (e.g. another manufacturer or
an association) to justify his own declaration of conformity regarding a product that is manufactured
according to the same design and with raw materials, constituents and manufacturing methods of
the same kind, provided that permission is given, and the test is valid for both products.

NB-CPD/SG10/12/091 Page 29 of 74
Annex A Tables for acceptance coefficient kn depending on the used
fractile p and confidence level (taken from ISO 16269-6
(2005))

n fractile : p n fractile : p
0,50 0,75 0,90 0,95 0,50 0,75 0,90 0,95
2 0,000 0,675 1,282 1,645 51 0,000 0,675 1,282 1,645
3 0,000 0,675 1,282 1,645 52 0,000 0,675 1,282 1,645
4 0,000 0,675 1,282 1,645 53 0,000 0,675 1,282 1,645
5 0,000 0,675 1,282 1,645 54 0,000 0,675 1,282 1,645
6 0,000 0,675 1,282 1,645 55 0,000 0,675 1,282 1,645
7 0,000 0,675 1,282 1,645 56 0,000 0,675 1,282 1,645
8 0,000 0,675 1,282 1,645 57 0,000 0,675 1,282 1,645
9 0,000 0,675 1,282 1,645 58 0,000 0,675 1,282 1,645
10 0,000 0,675 1,282 1,645 59 0,000 0,675 1,282 1,645
11 0,000 0,675 1,282 1,645 60 0,000 0,675 1,282 1,645
12 0,000 0,675 1,282 1,645 61 0,000 0,675 1,282 1,645
13 0,000 0,675 1,282 1,645 62 0,000 0,675 1,282 1,645
14 0,000 0,675 1,282 1,645 63 0,000 0,675 1,282 1,645
15 0,000 0,675 1,282 1,645 64 0,000 0,675 1,282 1,645
16 0,000 0,675 1,282 1,645 65 0,000 0,675 1,282 1,645
17 0,000 0,675 1,282 1,645 66 0,000 0,675 1,282 1,645
18 0,000 0,675 1,282 1,645 67 0,000 0,675 1,282 1,645
19 0,000 0,675 1,282 1,645 68 0,000 0,675 1,282 1,645
20 0,000 0,675 1,282 1,645 69 0,000 0,675 1,282 1,645
21 0,000 0,675 1,282 1,645 70 0,000 0,675 1,282 1,645
22 0,000 0,675 1,282 1,645 71 0,000 0,675 1,282 1,645
23 0,000 0,675 1,282 1,645 72 0,000 0,675 1,282 1,645
24 0,000 0,675 1,282 1,645 73 0,000 0,675 1,282 1,645
25 0,000 0,675 1,282 1,645 74 0,000 0,675 1,282 1,645
26 0,000 0,675 1,282 1,645 75 0,000 0,675 1,282 1,645
27 0,000 0,675 1,282 1,645 76 0,000 0,675 1,282 1,645
28 0,000 0,675 1,282 1,645 77 0,000 0,675 1,282 1,645
29 0,000 0,675 1,282 1,645 78 0,000 0,675 1,282 1,645
30 0,000 0,675 1,282 1,645 79 0,000 0,675 1,282 1,645
31 0,000 0,675 1,282 1,645 80 0,000 0,675 1,282 1,645
32 0,000 0,675 1,282 1,645 81 0,000 0,675 1,282 1,645
33 0,000 0,675 1,282 1,645 82 0,000 0,675 1,282 1,645
34 0,000 0,675 1,282 1,645 83 0,000 0,675 1,282 1,645
35 0,000 0,675 1,282 1,645 84 0,000 0,675 1,282 1,645
36 0,000 0,675 1,282 1,645 85 0,000 0,675 1,282 1,645
37 0,000 0,675 1,282 1,645 86 0,000 0,675 1,282 1,645
38 0,000 0,675 1,282 1,645 87 0,000 0,675 1,282 1,645
39 0,000 0,675 1,282 1,645 88 0,000 0,675 1,282 1,645
40 0,000 0,675 1,282 1,645 89 0,000 0,675 1,282 1,645
41 0,000 0,675 1,282 1,645 90 0,000 0,675 1,282 1,645
42 0,000 0,675 1,282 1,645 91 0,000 0,675 1,282 1,645
43 0,000 0,675 1,282 1,645 92 0,000 0,675 1,282 1,645
44 0,000 0,675 1,282 1,645 93 0,000 0,675 1,282 1,645
45 0,000 0,675 1,282 1,645 94 0,000 0,675 1,282 1,645
46 0,000 0,675 1,282 1,645 95 0,000 0,675 1,282 1,645
47 0,000 0,675 1,282 1,645 96 0,000 0,675 1,282 1,645
48 0,000 0,675 1,282 1,645 97 0,000 0,675 1,282 1,645
49 0,000 0,675 1,282 1,645 98 0,000 0,675 1,282 1,645
50 0,000 0,675 1,282 1,645 99 0,000 0,675 1,282 1,645
100 0,000 0,675 1,282 1,645

Table A1. k1 for one-sided statistical tolerance, standard deviation: known and confidence
level = 50 %

NB-CPD/SG10/12/091 Page 30 of 74
n fractile : p n fractile : p
0,50 0,75 0,90 0,95 0,50 0,75 0,90 0,95
2 0,477 1,152 1,759 2,122 51 0,095 0,769 1,376 1,740
3 0,390 1,064 0,671 2,035 52 0,094 0,768 1,375 1,739
4 0,388 1,012 0,619 1,983 53 0,094 0,768 1,375 1,738
5 0,302 0,977 1,584 1,947 54 0,093 0,767 1,374 1,737
6 0,276 0,950 1,557 1,921 55 0,092 0,766 1,373 1,737
7 0,255 0,930 1,537 1,900 56 0,091 0,765 1,372 1,736
8 0,239 0,913 1,521 1,884 57 0,090 0,764 1,371 1,735
9 0,225 0,900 1,507 1,870 58 0,090 0,764 1,371 1,734
10 0,214 0,888 1,495 1,859 59 0,089 0,763 1,370 1,733
11 0,204 0,878 1,485 1,849 60 0,088 0,762 1,369 1,732
12 0,195 0,870 1,477 1,840 61 0,087 0,761 1,368 1,731
13 0,188 0,862 1,469 1,832 62 0,087 0,761 1,368 1,731
14 0,181 0,855 1,462 1,826 63 0,086 0,760 1,367 1,730
15 0,175 0,849 1,456 1,820 64 0,085 0,760 1,367 1,730
16 0,169 0,844 1,451 1,814 65 0,085 0,759 1,366 1,729
17 0,164 0,839 1,446 1,809 66 0,084 0,758 1,365 1,728
18 0,159 0,834 1,441 1,804 67 0,083 0,758 1,365 1,728
19 0,155 0,830 1,437 1,800 68 0,082 0,757 1,364 1,727
20 0,151 0,826 1,433 1,796 69 0,082 0,757 1,364 1,727
21 0,148 0,823 1,430 1,793 70 0,081 0,756 1,363 1,726
22 0,144 0,819 1,426 1,789 71 0,081 0,755 1,362 1,726
23 0,141 0,816 1,423 1,786 72 0,080 0,755 1,362 1,725
24 0,138 0,813 1,420 1,783 73 0,080 0,754 1,361 1,725
25 0,136 0,810 1,417 1,781 74 0,079 0,754 1,361 1,724
26 0,133 0,807 1,414 1,778 75 0,079 0,753 1,360 1,724
27 0,131 0,805 1,412 1,776 76 0,078 0,752 1,359 1,723
28 0,128 0,802 1,410 1,773 77 0,078 0,752 1,359 1,723
29 0,126 0,800 1,408 1,771 78 0,077 0,751 1,358 1,722
30 0,124 0,798 1,405 1,768 79 0,077 0,751 1,358 1,722
31 0,122 0,796 1,403 1,766 80 0,076 0,75 1,357 1,721
32 0,120 0,794 1,401 1,764 81 0,076 0,750 1,357 1,721
33 0,119 0,793 1,400 1,763 82 0,075 0,749 1,356 1,720
34 0,117 0,791 1,398 1,761 83 0,075 0,749 1,356 1,720
35 0,115 0,789 1,396 1,759 84 0,074 0,748 1,355 1,719
36 0,113 0,788 1,395 1,758 85 0,074 0,748 1,355 1,719
37 0,112 0,786 1,393 1,756 86 0,074 0,748 1,355 1,718
38 0,110 0,785 1,392 1,755 87 0,073 0,747 1,354 1,718
39 0,109 0,783 1,390 1,753 88 0,073 0,747 1,354 1,717
40 0,107 0,782 1,389 1,752 89 0,072 0,746 1,353 1,717
41 0,106 0,781 1,388 1,751 90 0,072 0,746 1,353 1,716
42 0,105 0,780 1,387 1,750 91 0,072 0,746 1,353 1,716
43 0,103 0,778 1,385 1,748 92 0,071 0,745 1,352 1,715
44 0,102 0,777 1,384 1,747 93 0,071 0,745 1,352 1,715
45 0,101 0,776 1,383 1,746 94 0,070 0,744 1,352 1,715
46 0,100 0,775 1,382 1,745 95 0,070 0,744 1,352 1,715
47 0,099 0,774 1,381 1,744 96 0,070 0,744 1,351 1,714
48 0,098 0,772 1,379 1,743 97 0,069 0,743 1,351 1,714
49 0,097 0,771 1,378 1,742 98 0,069 0,743 1,351 1,714
50 0,096 0,770 1,377 1,741 99 0,068 0,742 1,350 1,713
100 0,068 0,742 1,35 1,713

Table A2. k1 for one-sided statistical tolerance, standard deviation: known and confidence
level = 75 %

NB-CPD/SG10/12/091 Page 31 of 74
n fractile : p n fractile : p
0,50 0,75 0,90 0,95 0,50 0,75 0,90 0,95
2 0,907 1,581 2,188 2,552 51 0,180 0,854 1,461 1,825
3 0,740 1,415 2,022 2,385 52 0,179 0,853 1,460 1,824
4 0,641 1,316 1,923 2,286 53 0,177 0,851 1,458 1,822
5 0,574 1,248 1,855 2,218 54 0,176 0,850 1,457 1,821
6 0,524 1,198 1,805 2,169 55 0,174 0,848 1,455 1,819
7 0,485 1,159 1,766 2,130 56 0,172 0,846 1,453 1,817
8 0,454 1,128 1,735 2,098 57 0,171 0,845 1,452 1,816
9 0,428 1,102 1,709 2,073 58 0,169 0,843 1,450 1,814
10 0,406 1,080 1,687 2,051 59 0,168 0,842 1,449 1,813
11 0,387 1,061 1,668 2,032 60 0,166 0,840 1,447 1,811
12 0,370 1,045 1,652 2,015 61 0,165 0,839 1,446 1,810
13 0,356 1,030 1,637 2,001 62 0,164 0,838 1,445 1,809
14 0,343 1,017 1,625 1,998 63 0,162 0,836 1,443 1,807
15 0,331 1,006 1,613 1,976 64 0,161 0,835 1,442 1,806
16 0,321 0,995 1,602 1,966 65 0,160 0,834 1,441 1,805
17 0,311 0,986 1,593 1,956 66 0,159 0,833 1,440 1,804
18 0,303 0,977 1,584 1,947 67 0,158 0,832 1,439 1,803
19 0,295 0,969 1,576 1,939 68 0,156 0,830 1,437 1,801
20 0,287 0,962 1,569 1,932 69 0,155 0,829 1,436 1,800
21 0,281 0,955 1,562 1,926 70 0,154 0,828 1,435 1,799
22 0,274 0,948 1,555 1,919 71 0,153 0,827 1,434 1,798
23 0,268 0,943 1,550 1,913 72 0,152 0,826 1,433 1,797
24 0,262 0,937 1,544 1,907 73 0,151 0,825 1,432 1,796
25 0,257 0,932 1,539 1,902 74 0,150 0,824 1,431 1,795
26 0,252 0,926 1,533 1,897 75 0,149 0,823 1,430 1,794
27 0,248 0,922 1,529 1,893 76 0,148 0,822 1,429 1,793
28 0,243 0,917 1,524 1,888 77 0,147 0,821 1,428 1,792
29 0,239 0,913 1,520 1,884 78 0,146 0,820 1,427 1,791
30 0,234 0,909 1,516 1,879 79 0,145 0,819 1,426 1,790
31 0,231 0,906 1,513 1,876 80 0,144 0,818 1,425 1,789
32 0,227 0,902 1,509 1,872 81 0,143 0,817 1,424 1,788
33 0,224 0,899 1,506 1,869 82 0,142 0,816 1,423 1,787
34 0,220 0,895 1,502 1,865 83 0,142 0,816 1,423 1,786
35 0,217 0,892 1,499 1,862 84 0,141 0,815 1,422 1,785
36 0,214 0,889 1,496 1,859 85 0,140 0,814 1,421 1,785
37 0,211 0,886 1,493 1,856 86 0,139 0,813 1,420 1,784
38 0,209 0,884 1,491 1,854 87 0,138 0,812 1,419 1,783
39 0,206 0,881 1,488 1,851 88 0,138 0,812 1,419 1,782
40 0,203 0,878 1,485 1,848 89 0,137 0,811 1,418 1,781
41 0,201 0,876 1,483 1,846 90 0,136 0,810 1,417 1,780
42 0,199 0,873 1,480 1,843 91 0,135 0,809 1,416 1,779
43 0,196 0,871 1,478 1,841 92 0,135 0,809 1,416 1,779
44 0,194 0,868 1,475 1,838 93 0,134 0,808 1,415 1,778
45 0,192 0,866 1,473 1,836 94 0,133 0,807 1,414 1,778
46 0,190 0,864 1,471 1,834 95 0,133 0,807 1,414 1,777
47 0,188 0,862 1,469 1,832 96 0,132 0,806 1,413 1,776
48 0,186 0,860 1,467 1,831 97 0,131 0,805 1,412 1,776
49 0,184 0,858 1,465 1,829 98 0,130 0,804 1,411 1,775
50 0,182 0,856 1,463 1,827 99 0,130 0,804 1,411 1,775
100 0,129 0,803 1,410 1,774

Table A3. k1 for one-sided statistical tolerance, standard deviation: known and confidence
level = 90 %

NB-CPD/SG10/12/091 Page 32 of 74
n fractile : p n fractile : p
0,50 0,75 0,90 0,95 0,50 0,75 0,90 0,95
2 1,164 0,838 2,445 2,828 51 0,231 0,906 1,513 1,876
3 0,950 0,625 2,232 2,595 52 0,229 0,904 1,511 1,874
4 0,823 1,497 2,104 2,468 53 0,227 0,902 1,509 1,872
5 0,736 1,411 2,018 2,381 54 0,225 0,900 1,507 1,870
6 0,672 1,346 1,954 2,317 55 0,223 0,898 1,505 1,868
7 0,622 1,297 1,904 2,267 56 0,221 0,895 1,502 1,866
8 0,582 1,257 1,864 2,227 57 0,219 0,893 1,500 1,864
9 0,549 1,223 1,830 2,194 58 0,217 0,891 1,498 1,862
10 0,521 1,195 1,802 2,166 59 0,215 0,889 1,496 1,860
11 0,496 1,171 1,778 2,141 60 0,213 0,887 1,494 1,858
12 0,475 1,150 1,757 2,120 61 0,211 0,886 1,493 1,856
13 0,457 1,131 1,738 2,102 62 0,210 0,884 1,491 1,855
14 0,440 1,115 1,722 2,085 63 0,208 0,883 1,490 1,853
15 0,425 1,100 1,707 2,070 64 0,207 0,881 1,488 1,852
16 0,412 1,086 1,693 2,057 65 0,205 0,880 1,487 1,850
17 0,399 1,074 1,691 2,044 66 0,203 0,878 1,485 1,848
18 0,388 1,063 1,670 2,033 67 0,202 0,877 1,484 1,847
19 0,378 1,052 1,659 2,023 68 0,200 0,875 1,482 1,845
20 0,368 1,043 1,650 2,013 69 0,199 0,874 1,481 1,844
21 0,360 1,035 1,642 2,005 70 0,197 0,872 1,479 1,842
22 0,351 1,026 1,633 1,996 71 0,196 0,871 1,478 1,841
23 0,344 1,019 1,626 1,989 72 0,194 0,869 1,476 1,839
24 0,336 1,011 1,618 1,981 73 0,193 0,868 1,475 1,838
25 0,330 1,005 1,612 1,975 74 0,192 0,867 1,474 1,837
26 0,323 0,998 1,605 1,968 75 0,191 0,866 1,473 1,836
27 0,317 0,992 1,599 1,962 76 0,189 0,864 1,471 1,834
28 0,311 0,986 1,593 1,956 77 0,188 0,863 1,470 1,833
29 0,306 0,981 1,588 1,951 78 0,187 0,862 1,469 1,832
30 0,301 0,975 1,582 1,946 79 0,185 0,860 1,467 1,830
31 0,297 0,971 1,578 1,941 80 0,184 0,859 1,466 1,829
32 0,292 0,966 1,573 1,937 81 0,183 0,858 1,465 1,828
33 0,288 0,962 1,569 1,932 82 0,182 0,857 1,464 1,827
34 0,283 0,957 1,564 1,928 83 0,181 0,856 1,463 1,826
35 0,279 0,953 1,560 1,923 84 0,180 0,855 1,462 1,825
36 0,275 0,949 1,556 1,919 85 0,179 0,854 1,461 1,824
37 0,272 0,946 1,553 1,916 86 0,178 0,852 1,459 1,823
38 0,268 0,942 1,549 1,912 87 0,177 0,851 1,458 1,822
39 0,265 0,939 1,546 1,909 88 0,176 0,850 1,457 1,821
40 0,261 0,935 1,542 1,905 89 0,175 0,849 1,456 1,820
41 0,258 0,932 1,539 1,902 90 0,174 0,848 1,455 1,819
42 0,255 0,929 1,536 1,899 91 0,173 0,847 1,454 1,818
43 0,252 0,926 1,533 1,897 92 0,172 0,846 1,453 1,817
44 0,249 0,923 1,530 1,894 93 0,171 0,845 1,453 1,816
45 0,246 0,920 1,527 1,891 94 0,170 0,844 1,452 1,815
46 0,243 0,918 1,525 1,888 95 0,170 0,844 1,451 1,815
47 0,241 0,915 1,522 1,886 96 0,169 0,843 1,450 1,814
48 0,238 0,913 1,520 1,883 97 0,168 0,842 1,449 1,813
49 0,236 0,910 1,517 1,881 98 0,167 0,841 1,449 1,812
50 0,233 0,908 1,515 1,878 99 0,166 0,840 1,448 1,811
100 0,165 0,839 1,447 1,810

Table A4. k1 for one-sided statistical tolerance, standard deviation: known and confidence
level = 95 %

NB-CPD/SG10/12/091 Page 33 of 74
n fractile : p n fractile : p
0,50 0,75 0,90 0,95 0,50 0,75 0,90 0,95
2 0,000 0,888 1,785 2,339 51 0,000 0,679 1,290 1,655
3 0,000 0,774 1,499 1,939 52 0,000 0,679 1,290 1,655
4 0,000 0,739 1,419 1,830 53 0,000 0,679 1,289 1,655
5 0,000 0,722 1,382 1,780 54 0,000 0,679 1,289 1,655
6 0,000 0,712 1,361 1,751 55 0,000 0,679 1,289 1,655
7 0,000 0,706 1,347 1,732 56 0,000 0,678 1,289 1,654
8 0,000 0,701 1,337 1,719 57 0,000 0,678 1,289 1,654
9 0,000 0,698 1,330 1,710 58 0,000 0,678 1,288 1,654
10 0,000 0,695 1,325 1,702 59 0,000 0,678 1,288 1,654
11 0,000 0,693 1,320 1,696 60 0,000 0,678 1,288 1,654
12 0,000 0,692 1,317 1,691 61 0,000 0,678 1,288 1,654
13 0,000 0,690 1,314 1,687 62 0,000 0,678 1,288 1,654
14 0,000 0,689 1,311 1,684 63 0,000 0,678 1,288 1,653
15 0,000 0,688 1,309 1,681 64 0,000 0,678 1,288 1,653
16 0,000 0,678 1,307 1,679 65 0,000 0,678 1,288 1,653
17 0,000 0,686 1,306 1,677 66 0,000 0,678 1,287 1,653
18 0,000 0,686 1,304 1,675 67 0,000 0,678 1,287 1,653
19 0,000 0,685 1,303 1,673 68 0,000 0,678 1,287 1,652
20 0,000 0,685 1,302 1,672 69 0,000 0,678 1,287 1,652
21 0,000 0,685 1,301 1,671 70 0,000 0,678 1,287 1,652
22 0,000 0,684 1,300 1,669 71 0,000 0,678 1,287 1,652
23 0,000 0,684 1,299 1,668 72 0,000 0,678 1,287 1,652
24 0,000 0,683 1,298 1,667 73 0,000 0,678 1,287 1,652
25 0,000 0,683 1,298 1,666 74 0,000 0,678 1,287 1,652
26 0,000 0,682 1,297 1,665 75 0,000 0,678 1,287 1,652
27 0,000 0,682 1,297 1,665 76 0,000 0,677 1,287 1,652
28 0,000 0,682 1,296 1,664 77 0,000 0,677 1,287 1,652
29 0,000 0,682 1,296 1,663 78 0,000 0,677 1,287 1,652
30 0,000 0,681 1,295 1,662 79 0,000 0,677 1,287 1,652
31 0,000 0,681 1,295 1,662 80 0,000 0,677 1,287 1,652
32 0,000 0,681 1,294 1,661 81 0,000 0,677 1,287 1,652
33 0,000 0,680 1,294 1,661 82 0,000 0,677 1,287 1,652
34 0,000 0,680 1,293 1,660 83 0,000 0,677 1,287 1,652
35 0,000 0,680 1,293 1,660 84 0,000 0,677 1,287 1,652
36 0,000 0,680 1,293 1,660 85 0,000 0,677 1,287 1,652
37 0,000 0,680 1,293 1,659 86 0,000 0,677 1,286 1,651
38 0,000 0,680 1,292 1,659 87 0,000 0,677 1,286 1,651
39 0,000 0,680 1,292 1,658 88 0,000 0,677 1,286 1,651
40 0,000 0,680 1,292 1,658 89 0,000 0,677 1,286 1,651
41 0,000 0,680 1,292 1,658 90 0,000 0,677 1,286 1,651
42 0,000 0,680 1,291 1,658 91 0,000 0,677 1,286 1,651
43 0,000 0,679 1,291 1,657 92 0,000 0,677 1,286 1,651
44 0,000 0,679 1,290 1,657 93 0,000 0,677 1,286 1,651
45 0,000 0,679 1,290 1,657 94 0,000 0,677 1,286 1,651
46 0,000 0,679 1,290 1,657 95 0,000 0,677 1,286 1,651
47 0,000 0,679 1,290 1,656 96 0,000 0,677 1,286 1,650
48 0,000 0,679 1,290 1,656 97 0,000 0,677 1,286 1,650
49 0,000 0,679 1,290 1,655 98 0,000 0,677 1,286 1,650
50 0,000 0,679 1,290 1,655 99 0,000 0,677 1,286 1,650
100 0,000 0,677 1,286 1,650

Table A5. k1 for one-sided statistical tolerance, standard deviation: unknown and confidence
level = 50 %

NB-CPD/SG10/12/091 Page 34 of 74
n fractile : p n fractile : p
0,50 0,75 0,90 0,95 0,50 0,75 0,90 0,95
2 0,708 2,225 3,993 5,122 51 0,096 0,788 1,425 1,809
3 0,472 1,465 2,502 3,152 52 0,095 0,787 1,423 1,808
4 0,383 1,256 2,134 2,681 53 0,094 0,786 1,422 1,806
5 0,332 1,152 1,962 2,464 54 0,093 0,785 1,420 1,805
6 0,297 1,088 1,860 2,336 55 0,093 0,784 1,419 1,803
7 0,272 1,044 1,791 2,251 56 0,092 0,782 1,418 1,801
8 0,252 0,011 1,740 2,189 57 0,091 0,781 1,416 1,800
9 0,236 0,985 1,702 2,142 58 0,090 0,780 1,415 1,798
10 0,223 0,964 1,671 2,104 59 0,089 0,779 1,413 1,797
11 0,212 0,947 1,646 2,074 60 0,088 0,778 1,412 1,795
12 0,202 0,933 1,625 2,048 61 0,087 0,777 1,411 1,794
13 0,193 0,920 1,607 2,026 62 0,087 0,776 1,410 1,793
14 0,186 0,909 1,591 2,008 63 0,086 0,776 1,409 1,791
15 0,179 0,900 1,578 1,991 64 0,086 0,775 1,408 1,790
16 0,173 0,891 1,566 1,977 65 0,085 0,774 1,407 1,789
17 0,168 0,884 1,555 1,964 66 0,084 0,773 1,405 1,788
18 0,163 0,877 1,545 1,952 67 0,084 0,772 1,404 1,787
19 0,158 0,870 1,536 1,942 68 0,083 0,772 1,403 1,785
20 0,154 0,865 1,529 1,932 69 0,083 0,771 1,402 1,784
21 0,151 0,860 1,522 1,924 70 0,082 0,770 1,401 1,783
22 0,147 0,854 1,514 1,916 71 0,081 0,769 1,400 1,782
23 0,144 0,850 1,509 1,909 72 0,081 0,769 1,399 1,781
24 0,140 0,846 1,503 1,902 73 0,080 0,768 1,399 1,780
25 0,138 0,842 1,498 1,896 74 0,080 0,767 1,398 1,779
26 0,135 0,838 1,492 1,889 75 0,079 0,767 1,397 1,778
27 0,133 0,835 1,488 1,884 76 0,078 0,766 1,396 1,777
28 0,130 0,831 1,483 1,879 77 0,078 0,765 1,395 1,776
29 0,128 0,828 1,479 1,874 78 0,077 0,764 1,395 1,775
30 0,125 0,825 1,475 1,869 79 0,077 0,764 1,394 1,774
31 0,123 0,823 1,472 1,865 80 0,076 0,763 1,393 1,773
32 0,121 0,820 1,468 1,861 81 0,076 0,763 1,392 1,772
33 0,120 0,818 1,465 1,858 82 0,075 0,762 1,392 1,771
34 0,118 0,815 1,461 1,854 83 0,075 0,762 1,391 1,771
35 0,116 0,813 1,458 1,850 84 0,074 0,761 1,390 1,770
36 0,114 0,811 1,455 1,847 85 0,074 0,761 1,390 1,769
37 0,113 0,809 1,453 1,844 86 0,074 0,760 1,389 1,768
38 0,111 0,807 1,450 1,840 87 0,073 0,760 1,388 1,767
39 0,110 0,805 1,448 1,837 88 0,073 0,759 1,387 1,767
40 0,108 0,803 1,445 1,834 89 0,072 0,759 1,387 1,766
41 0,107 0,801 1,443 1,832 90 0,072 0,758 1,386 1,765
42 0,106 0,800 1,441 1,829 91 0,072 0,758 1,385 1,764
43 0,104 0,798 1,439 1,827 92 0,071 0,757 1,385 1,764
44 0,103 0,797 1,437 1,824 93 0,071 0,757 1,384 1,763
45 0,102 0,795 1,435 1,822 94 0,070 0,756 1,384 1,762
46 0,101 0,794 1,433 1,820 95 0,070 0,756 1,383 1,762
47 0,100 0,793 1,431 1,818 96 0,070 0,755 1,382 1,761
48 0,099 0,791 1,430 1,815 97 0,069 0,755 1,382 1,760
49 0,098 0,790 1,428 1,813 98 0,069 0,754 1,381 1,759
50 0,097 0,789 1,426 1,811 99 0,068 0,754 1,381 1,759
100 0,068 0,753 1,380 1,758

Table A6. k1 for one-sided statistical tolerance, standard deviation: unknown and confidence
level = 75 %

NB-CPD/SG10/12/091 Page 35 of 74
n fractile : p n fractile : p
0,50 0,75 0,90 0,95 0,50 0,75 0,90 0,95
2 2,177 5,843 10,253 13,090 51 0,182 0,892 1,557 1,963
3 1,089 2,603 4,259 5,312 52 0,181 0,890 1,555 1,960
4 0,819 1,973 3,188 3,957 53 0,179 0,888 1,552 1,956
5 0,686 1,698 2,743 3,400 54 0,178 0,886 1,549 1,953
6 0,603 1,540 2,494 3,092 55 0,176 0,884 1,547 1,950
7 0,545 1,436 2,333 2,894 56 0,174 0,881 1,544 1,947
8 0,501 1,360 2,219 2,755 57 0,173 0,879 1,541 1,944
9 0,466 1,303 2,133 2,650 58 0,171 0,877 1,538 1,940
10 0,438 1,257 2,066 2,569 59 0,170 0,875 1,536 1,937
11 0,414 1,220 2,012 2,503 60 0,168 0,873 1,533 1,934
12 0,394 1,189 1,967 2,449 61 0,167 0,871 1,531 1,932
13 0,377 1,162 1,929 2,403 62 0,165 0,870 1,529 1,929
14 0,361 1,139 1,896 2,364 63 0,164 0,868 1,527 1,927
15 0,348 1,119 1,867 2,329 64 0,163 0,867 1,525 1,924
16 0,336 1,101 1,842 2,299 65 0,162 0,865 1,523 1,922
17 0,325 1,085 1,820 2,273 66 0,160 0,863 1,520 1,920
18 0,315 1,071 1,800 2,249 67 0,159 0,862 1,518 1,917
19 0,306 1,058 1,782 2,228 68 0,158 0,860 1,516 1,915
20 0,297 1,046 1,766 2,208 69 0,156 0,859 1,514 1,912
21 0,290 1,036 1,752 2,191 70 0,155 0,857 1,512 1,910
22 0,283 1,026 1,737 2,174 71 0,154 0,856 1,510 1,908
23 0,277 1,017 1,725 2,160 72 0,153 0,855 1,509 1,906
24 0,270 1,008 1,713 2,146 73 0,152 0,853 1,507 1,904
25 0,265 1,001 1,703 2,134 74 0,151 0,852 1,505 1,902
26 0,259 0,993 1,692 2,121 75 0,150 0,851 1,504 1,900
27 0,254 0,986 1,683 2,110 76 0,149 0,850 1,502 1,898
28 0,249 0,979 1,674 2,099 77 0,148 0,849 1,500 1,896
29 0,245 0,973 1,666 2,090 78 0,147 0,847 1,498 1,894
30 0,240 0,967 1,658 2,080 79 0,146 0,846 1,497 1,892
31 0,236 1,162 1,651 2,072 80 0,145 0,845 1,495 1,890
32 0,232 1,357 1,644 2,064 81 0,144 0,844 1,494 1,889
33 0,229 1,553 1,638 2,057 82 0,143 0,843 1,492 1,887
34 0,225 1,748 1,631 2,049 83 0,143 0,842 1,491 1,886
35 0,221 1,943 1,624 2,041 84 0,142 0,841 1,490 1,884
36 0,218 1,739 1,619 2,035 85 0,141 0,840 1,489 1,883
37 0,215 1,535 1,614 2,029 86 0,140 0,838 1,487 1,881
38 0,213 1,331 1,608 2,023 87 0,139 0,837 1,486 1,880
39 0,210 1,127 1,603 2,017 88 0,139 0,836 1,485 1,878
40 0,207 0,923 1,598 2,011 89 0,138 0,835 1,483 1,877
41 0,204 0,920 1,594 2,006 90 0,137 0,834 1,482 1,875
42 0,202 0,917 1,590 2,001 91 0,136 0,833 1,481 1,874
43 0,199 0,913 1,585 1,996 92 0,136 0,832 1,480 1,872
44 0,197 0,910 1,581 1,991 93 0,135 0,831 1,479 1,871
45 0,194 0,907 1,577 1,986 94 0,134 0,830 1,478 1,870
46 0,192 0,904 1,574 1,982 95 0,134 0,830 1,477 1,869
47 0,190 0,902 1,570 1,978 96 0,133 0,829 1,475 1,867
48 0,188 0,899 1,567 1,974 97 0,132 0,828 1,474 1,866
49 0,186 0,897 1,563 1,970 98 0,131 0,827 1,473 1,865
50 0,184 0,894 1,560 1,966 99 0,131 0,826 1,472 1,863
100 0,130 0,825 1,471 1,862

Table A7. k1 for one-sided statistical tolerance, standard deviation: unknown and confidence
level = 90 %

NB-CPD/SG10/12/091 Page 36 of 74
n fractile : p n fractile : p
0,50 0,75 0,90 0,95 0,50 0,75 0,90 0,95
2 4,465 11,763 20,582 26,260 51 0,236 0,958 1,642 2,061
3 1,686 3,807 6,156 7,656 52 0,234 0,955 1,639 2,057
4 1,177 2,618 4,162 5,144 53 0,231 0,953 1,635 2,052
5 0,954 2,150 3,407 4,203 54 0,229 0,950 1,631 2,048
6 0,823 1,896 3,007 3,708 55 0,227 0,947 1,628 2,044
7 0,735 1,733 2,756 3,400 56 0,225 0,944 1,624 2,040
8 0,670 1,618 2,582 3,188 57 0,223 0,941 1,620 2,036
9 0,620 1,533 2,454 3,032 58 0,220 0,939 1,616 2,031
10 0,580 1,466 2,355 2,911 59 0,218 0,936 1,613 2,027
11 0,547 1,412 2,276 2,815 60 0,216 0,933 1,609 2,023
12 0,519 1,367 2,211 2,737 61 0,214 0,931 1,606 2,020
13 0,495 1,329 2,156 2,671 62 0,213 0,929 1,604 2,016
14 0,474 1,296 2,109 2,615 63 0,211 0,927 1,601 2,013
15 0,455 1,268 2,069 2,567 64 0,210 0,925 1,598 2,010
16 0,439 1,243 2,033 2,524 65 0,208 0,923 1,596 2,007
17 0,424 1,221 2,002 2,487 66 0,206 0,920 1,593 2,003
18 0,411 1,201 1,974 2,453 67 0,205 0,918 1,590 2,000
19 0,398 1,183 1,949 2,424 68 0,203 0,916 1,587 1,997
20 0,387 1,167 1,926 2,397 69 0,202 0,914 1,585 1,993
21 0,377 1,153 1,907 2,373 70 0,200 0,912 1,582 1,990
22 0,367 1,138 1,887 2,349 71 0,199 0,910 1,580 1,988
23 0,359 1,126 1,870 2,330 72 0,197 0,909 1,578 1,985
24 0,350 1,114 1,853 2,310 73 0,196 0,907 1,575 1,983
25 0,343 1,104 1,839 2,293 74 0,195 0,905 1,573 1,980
26 0,335 1,093 1,825 2,276 75 0,194 0,904 1,571 1,978
27 0,329 1,084 1,813 2,261 76 0,192 0,902 1,569 1,975
28 0,322 1,075 1,800 2,246 77 0,191 0,900 1,567 1,973
29 0,317 1,067 1,789 2,233 78 0,190 0,898 1,564 1,970
30 0,311 1,059 1,778 2,220 79 0,188 0,897 1,562 1,968
31 0,306 1,052 1,769 2,209 80 0,187 0,895 1,560 1,965
32 0,301 1,046 1,760 2,199 81 0,186 0,894 1,558 1,963
33 0,296 1,039 1,751 2,188 82 0,185 0,892 1,556 1,961
34 0,291 1,033 1,742 2,178 83 0,184 0,891 1,555 1,959
35 0,286 1,026 1,733 2,167 84 0,183 0,890 1,553 1,957
36 0,282 1,021 1,726 2,159 85 0,182 0,889 1,551 1,955
37 0,278 1,016 1,719 2,151 86 0,180 0,887 1,549 1,952
38 0,275 1,010 1,712 2,142 87 0,179 0,886 1,547 1,950
39 0,271 1,005 1,705 2,134 88 0,178 0,885 1,546 1,948
40 0,267 1,000 1,698 2,126 89 0,177 0,883 1,544 1,946
41 0,264 0,996 1,692 2,119 90 0,176 0,882 1,542 1,944
42 0,261 0,991 1,686 2,113 91 0,175 0,881 1,541 1,942
43 0,257 0,987 1,681 2,106 92 0,174 0,880 1,539 1,941
44 0,254 0,982 1,675 2,100 93 0,173 0,878 1,538 1,939
45 0,251 0,978 1,669 2,093 94 0,172 0,877 1,536 1,937
46 0,248 0,975 1,664 2,087 95 0,172 0,876 1,535 1,936
47 0,246 0,971 1,660 2,082 96 0,171 0,875 1,533 1,934
48 0,243 0,968 1,655 2,076 97 0,170 0,874 1,532 1,932
49 0,241 0,964 1,651 2,071 98 0,169 0,872 1,530 1,930
50 0,238 0,961 1,646 2,065 99 0,168 0,871 1,529 1,929
100 0,167 0,870 1,527 1,927

Table A8. k1 for one-sided statistical tolerance, standard deviation: unknown and confidence
level = 95 %

NB-CPD/SG10/12/091 Page 37 of 74
n fractile : p n fractile : p
0,50 0,75 0,90 0,95 0,50 0,75 0,90 0,95
2 0,755 0,282 1,823 2,164 51 0,678 1,156 1,653 1,969
3 0,727 1,238 1,766 2,100 52 0,678 1,156 1,653 1,969
4 0,714 1,216 1,737 2,067 53 0,678 1,156 1,653 1,969
5 0,706 1,203 1,719 2,046 54 0,678 1,156 1,653 1,969
6 0,701 1,195 1,707 2,033 55 0,678 1,156 1,653 1,969
7 0,697 1,188 1,698 2,023 56 0,678 1,155 1,652 1,968
8 0,694 1,184 1,692 2,015 57 0,678 1,155 1,652 1,968
9 0,692 1,180 1,686 2,009 58 0,678 1,155 1,652 1,968
10 0,690 1,177 1,682 2,004 59 0,678 1,155 1,652 1,968
11 0,689 1,175 1,679 2,000 60 0,678 1,155 1,652 1,968
12 0,688 1,173 1,676 1,997 61 0,678 1,155 1,652 1,968
13 0,687 1,171 1,674 1,994 62 0,678 1,155 1,652 1,968
14 0,686 1,170 1,672 1,992 63 0,678 1,155 1,652 1,968
15 0,685 1,168 1,670 1,990 64 0,678 1,155 1,652 1,968
16 0,685 1,167 1,669 1,988 65 0,678 1,155 1,652 1,968
17 0,684 1,166 1,667 1,986 66 0,677 1,155 1,651 1,967
18 0,684 1,165 1,666 1,985 67 0,677 1,155 1,651 1,967
19 0,683 1,165 1,665 1,984 68 0,677 1,155 1,651 1,967
20 0,683 1,164 1,664 1,983 69 0,677 1,155 1,651 1,967
21 0,683 1,164 1,663 1,982 70 0,677 1,155 1,651 1,967
22 0,682 1,163 1,662 1,981 71 0,677 1,155 1,651 1,967
23 0,682 1,163 1,662 1,980 72 0,677 1,155 1,651 1,967
24 0,681 1,162 1,661 1,979 73 0,677 1,155 1,651 1,967
25 0,681 1,162 1,661 1,978 74 0,677 1,155 1,651 1,967
26 0,681 1,161 1,660 1,977 75 0,677 1,155 1,651 1,967
27 0,681 1,161 1,660 1,977 76 0,677 1,154 1,650 1,966
28 0,680 1,160 1,659 1,976 77 0,677 1,154 1,650 1,966
29 0,680 1,160 1,659 1,976 78 0,677 1,154 1,650 1,966
30 0,680 1,160 1,658 1,975 79 0,677 1,154 1,650 1,966
31 0,680 1,160 1,658 1,975 80 0,677 1,154 1,650 1,966
32 0,680 1,159 1,657 1,974 81 0,677 1,154 1,650 1,966
33 0,679 1,159 1,657 1,974 82 0,677 1,154 1,650 1,966
34 0,679 1,158 1,656 1,973 83 0,677 1,154 1,650 1,966
35 0,679 1,158 1,656 1,973 84 0,677 1,154 1,650 1,966
36 0,679 1,158 1,656 1,973 85 0,677 1,154 1,650 1,966
37 0,679 1,158 1,656 1,973 86 0,677 1,154 1,650 1,965
38 0,679 1,157 1,655 1,972 87 0,677 1,154 1,650 1,965
39 0,679 1,157 1,655 1,972 88 0,677 1,154 1,650 1,965
40 0,679 1,157 1,655 1,972 89 0,677 1,154 1,650 1,965
41 0,679 1,157 1,655 1,972 90 0,677 1,154 1,650 1,965
42 0,679 1,157 1,655 1,971 91 0,677 1,154 1,650 1,965
43 0,678 1,157 1,654 1,971 92 0,677 1,154 1,650 1,965
44 0,678 1,157 1,654 1,970 93 0,677 1,154 1,650 1,965
45 0,678 1,157 1,654 1,970 94 0,677 1,154 1,650 1,965
46 0,678 1,157 1,654 1,970 95 0,677 1,154 1,650 1,965
47 0,678 1,157 1,654 1,970 96 0,677 1,153 1,649 1,965
48 0,678 1,156 1,653 1,969 97 0,677 1,153 1,649 1,965
49 0,678 1,156 1,653 1,969 98 0,677 1,153 1,649 1,965
50 0,678 1,156 1,653 1,969 99 0,677 1,153 1,649 1,965
100 0,677 1,153 1,649 1,965

Table A9. k2 for two-sided statistical tolerance, standard deviation: known and confidence
level = 50 %

NB-CPD/SG10/12/091 Page 38 of 74
n fractile : p n fractile : p
0,50 0,75 0,90 0,95 0,50 0,75 0,90 0,95
2 0,919 1,520 2,106 2,464 51 0,684 1,166 1,667 1,986
3 0,834 1,402 1,971 2,323 52 0,684 1,166 1,666 1,985
4 0,792 1,340 1,897 2,244 53 0,683 1,165 1,666 1,985
5 0,768 1,303 1,850 2,194 54 0,683 1,165 1,665 1,984
6 0,752 1,278 1,818 2,158 55 0,683 1,165 1,665 1,984
7 0,741 1,260 1,794 2,132 56 0,683 1,165 1,665 1,984
8 0,732 1,246 1,776 2,112 57 0,683 1,165 1,664 1,983
9 0,726 1,236 1,762 2,096 58 0,682 1,164 1,664 1,983
10 0,721 1,227 1,751 2,083 59 0,682 1,164 1,663 1,982
11 0,716 1,220 1,742 2,073 60 0,682 1,164 1,663 1,982
12 0,713 1,214 1,734 2,064 61 0,682 1,164 1,663 1,982
13 0,710 1,209 1,727 2,056 62 0,682 1,164 1,663 1,981
14 0,707 1,205 1,722 2,050 63 0,682 1,163 1,662 1,981
15 0,705 1,202 1,717 2,044 64 0,682 1,163 1,662 1,981
16 0,703 1,198 1,712 2,039 65 0,682 1,163 1,662 1,981
17 0,702 1,196 1,708 2,034 66 0,681 1,163 1,662 1,980
18 0,700 1,193 1,705 2,030 67 0,681 1,163 1,662 1,980
19 0,699 1,191 1,702 2,027 68 0,681 1,162 1,661 1,980
20 0,698 1,189 1,699 2,024 69 0,681 1,162 1,661 1,979
21 0,697 1,187 1,697 2,021 70 0,681 1,162 1,661 1,979
22 0,695 1,185 1,694 2,018 71 0,681 1,162 1,661 1,979
23 0,695 1,184 1,692 2,016 72 0,681 1,162 1,661 1,979
24 0,694 1,183 1,690 2,013 73 0,681 1,161 1,660 1,978
25 0,693 1,182 1,684 2,011 74 0,681 1,161 1,660 1,978
26 0,692 1,180 1,678 2,009 75 0,681 1,161 1,660 1,978
27 0,692 1,179 1,681 2,008 76 0,681 1,161 1,660 1,978
28 0,691 1,178 1,684 2,006 77 0,681 1,161 1,660 1,978
29 0,691 1,177 1,683 2,005 78 0,681 1,160 1,659 1,977
30 0,690 1,176 1,681 2,003 79 0,681 1,160 1,659 1,977
31 0,690 1,175 1,680 2,002 80 0,681 1,16 1,659 1,977
32 0,689 1,175 1,679 2,001 81 0,681 1,160 1,659 1,977
33 0,689 1,174 1,678 1,999 82 0,681 1,160 1,659 1,977
34 0,688 1,174 1,677 1,998 83 0,681 1,160 1,658 1,976
35 0,688 1,173 1,676 1,997 84 0,681 1,160 1,658 1,976
36 0,688 1,172 1,675 1,996 85 0,681 1,160 1,658 1,976
37 0,687 1,172 1,674 1,995 86 0,680 1,159 1,658 1,976
38 0,687 1,171 1,674 1,994 87 0,680 1,159 1,658 1,976
39 0,686 1,171 1,673 1,993 88 0,680 1,159 1,657 1,975
40 0,686 1,170 1,672 1,992 89 0,680 1,159 1,657 1,975
41 0,686 1,170 1,671 1,991 90 0,68 1,159 1,657 1,975
42 0,686 1,169 1,671 1,991 91 0,680 1,159 1,657 1,975
43 0,685 1,169 1,670 1,990 92 0,680 1,159 1,657 1,975
44 0,685 1,168 1,670 1,990 93 0,680 1,159 1,657 1,974
45 0,685 1,168 1,669 1,989 94 0,680 1,159 1,657 1,974
46 0,685 1,168 1,669 1,988 95 0,680 1,159 1,657 1,974
47 0,685 1,167 1,668 1,988 96 0,679 1,158 1,656 1,974
48 0,684 1,167 1,668 1,987 97 0,679 1,158 1,656 1,974
49 0,684 1,166 1,667 1,987 98 0,679 1,158 1,656 1,973
50 0,684 1,166 1,667 1,986 99 0,679 1,158 1,656 1,973
100 0,679 1,158 1,656 1,973

Table A10. k2 for two-sided statistical tolerance, standard deviation: known and confidence
level = 75 %

NB-CPD/SG10/12/091 Page 39 of 74
n fractile : p n fractile : p
0,50 0,75 0,90 0,95 0,50 0,75 0,90 0,95
2 1,187 1,842 2,446 2,809 51 0,693 1,182 1,688 2,011
3 1,013 1,640 1,236 2,597 52 0,692 1,181 1,688 2,010
4 0,924 1,527 2,114 2,473 53 0,692 1,181 1,687 2,010
5 0,872 1,456 2,034 2,390 54 0,692 1,180 1,686 2,009
6 0,837 1,407 1,977 2,330 55 0,692 1,180 1,686 2,008
7 0,813 1,371 1,935 2,285 56 0,691 1,179 1,685 2,007
8 0,795 1,344 1,902 2,250 57 0,691 1,179 1,684 2,006
9 0,781 1,323 1,875 2,222 58 0,691 1,178 1,683 2,006
10 0,770 1,306 1,854 2,198 59 0,690 1,178 1,683 2,005
11 0,761 1,292 1,836 2,179 60 0,690 1,177 1,682 2,004
12 0,754 1,281 1,821 2,162 61 0,690 1,177 1,682 2,003
13 0,758 1,271 1,809 2,148 62 0,690 1,176 1,681 2,003
14 0,742 1,262 1,797 2,136 63 0,689 1,176 1,681 2,002
15 0,738 1,255 1,788 2,125 64 0,689 1,175 1,680 2,002
16 0,734 1,248 1,779 2,115 65 0,689 1,175 1,680 2,001
17 0,730 1,243 1,772 2,107 66 0,689 1,175 1,679 2,000
18 0,727 1,237 1,765 2,099 67 0,689 1,174 1,679 2,000
19 0,724 1,233 1,759 2,092 68 0,688 1,174 1,678 1,999
20 0,722 1,229 1,753 2,086 69 0,688 1,173 1,678 1,999
21 0,720 1,226 1,749 2,081 70 0,688 1,173 1,677 1,998
22 0,717 1,222 1,744 2,075 71 0,688 1,173 1,677 1,998
23 0,716 1,219 1,740 2,071 72 0,688 1,172 1,676 1,997
24 0,714 1,216 1,736 2,066 73 0,687 1,172 1,676 1,997
25 0,713 1,214 1,733 2,062 74 0,687 1,172 1,675 1,996
26 0,711 1,211 1,729 2,058 75 0,687 1,172 1,675 1,996
27 0,710 1,209 1,726 2,055 76 0,687 1,171 1,675 1,995
28 0,708 1,207 1,723 2,052 77 0,687 1,171 1,674 1,995
29 0,707 1,205 1,721 2,049 78 0,686 1,171 1,674 1,994
30 0,706 1,203 1,718 2,046 79 0,686 1,170 1,673 1,994
31 0,705 1,201 1,716 2,044 80 0,686 1,170 1,673 1,993
32 0,704 1,200 1,714 2,041 81 0,686 1,170 1,673 1,993
33 0,703 1,198 1,712 2,039 82 0,686 1,170 1,672 1,992
34 0,702 1,197 1,710 2,036 83 0,686 1,169 1,672 1,992
35 0,701 1,195 1,708 2,034 84 0,686 1,169 1,672 1,992
36 0,700 1,194 1,706 2,032 85 0,686 1,169 1,672 1,992
37 0,700 1,193 1,705 2,030 86 0,685 1,169 1,671 1,991
38 0,699 1,192 1,703 2,029 87 0,685 1,169 1,671 1,991
39 0,699 1,191 1,702 2,027 88 0,685 1,168 1,671 1,991
40 0,698 1,190 1,700 2,025 89 0,685 1,168 1,670 1,990
41 0,697 1,189 1,699 2,024 90 0,685 1,168 1,670 1,990
42 0,697 1,188 1,698 2,022 91 0,685 1,168 1,670 1,990
43 0,696 1,187 1,696 2,021 92 0,685 1,168 1,669 1,989
44 0,696 1,186 1,695 2,019 93 0,685 1,167 1,669 1,989
45 0,695 1,185 1,694 2,018 94 0,685 1,167 1,669 1,989
46 0,695 1,184 1,693 2,017 95 0,685 1,167 1,669 1,989
47 0,694 1,184 1,692 2,016 96 0,684 1,167 1,668 1,988
48 0,694 1,183 1,691 2,014 97 0,684 1,167 1,668 1,988
49 0,693 1,183 1,690 2,013 98 0,684 1,166 1,668 1,988
50 0,693 1,182 1,689 2,012 99 0,684 1,166 1,667 1,987
100 0,684 1,166 1,667 1,987

Table A11. k2 for two-sided statistical tolerance, standard deviation: known and confidence
level = 90 %

NB-CPD/SG10/12/091 Page 40 of 74
n fractile : p n fractile : p
0,50 0,75 0,90 0,95 0,50 0,75 0,90 0,95
2 1,393 2,062 2,668 3,031 51 0,701 1,194 1,707 2,032
3 1,160 1,812 2,415 2,777 52 0,700 1,194 1,706 2,031
4 1,036 1,668 1,265 2,627 53 0,700 1,193 1,705 2,030
5 0,960 1,574 2,165 2,525 54 0,699 1,192 1,704 2,029
6 0,910 1,509 2,093 2,451 55 0,699 1,192 1,703 2,028
7 0,875 1,460 2,039 2,395 56 0,699 1,191 1,701 2,026
8 0,894 0,423 1,996 2,350 57 0,698 1,190 1,700 2,025
9 0,828 1,394 1,961 2,313 58 0,698 1,189 1,699 2,024
10 0,812 1,370 1,933 2,283 59 0,697 1,189 1,698 2,023
11 0,799 1,351 1,909 2,258 60 0,697 1,188 1,697 2,022
12 0,788 1,334 1,889 2,236 61 0,697 1,187 1,696 2,021
13 0,779 1,320 1,872 2,218 62 0,696 1,187 1,696 2,020
14 0,772 1,308 1,857 2,201 63 0,696 1,186 1,695 2,019
15 0,765 1,298 1,844 2,187 64 0,696 1,186 1,694 2,018
16 0,759 1,289 1,832 2,174 65 0,696 1,185 1,694 2,018
17 0,754 1,281 1,822 2,163 66 0,695 1,184 1,693 2,017
18 0,749 1,274 1,812 2,152 67 0,695 1,184 1,692 2,016
19 0,745 1,267 1,804 2,143 68 0,695 1,183 1,691 2,015
20 0,742 1,261 1,797 2,135 69 0,694 1,183 1,691 2,014
21 0,739 1,256 1,790 2,128 70 0,694 1,182 1,690 2,013
22 0,736 1,251 1,783 2,120 71 0,694 1,182 1,689 2,012
23 0,733 1,247 1,778 2,114 72 0,693 1,181 1,689 2,012
24 0,730 1,243 1,772 2,108 73 0,693 1,181 1,688 2,011
25 0,728 1,240 1,768 2,103 74 0,693 1,180 1,688 2,011
26 0,726 1,236 1,763 2,097 75 0,693 1,180 1,687 2,010
27 0,724 1,233 1,759 2,093 76 0,692 1,180 1,686 2,009
28 0,722 1,230 1,755 2,088 77 0,692 1,179 1,686 2,009
29 0,721 1,228 1,752 2,084 78 0,692 1,179 1,685 2,008
30 0,719 1,225 1,748 2,080 79 0,691 1,178 1,685 2,008
31 0,718 1,223 1,745 2,077 80 0,691 1,178 1,684 2,007
32 0,717 1,221 1,742 2,073 81 0,691 1,178 1,684 2,007
33 0,715 1,218 1,739 2,070 82 0,691 1,177 1,683 2,006
34 0,714 1,216 1,736 2,066 83 0,690 1,177 1,683 2,006
35 0,713 1,214 1,733 2,063 84 0,690 1,177 1,682 2,005
36 0,712 1,212 1,731 2,061 85 0,690 1,177 1,682 2,005
37 0,711 1,211 1,729 2,058 86 0,690 1,176 1,682 2,004
38 0,710 1,209 1,727 2,056 87 0,690 1,176 1,681 2,004
39 0,709 1,208 1,725 2,053 88 0,689 1,176 1,681 2,003
40 0,708 1,206 1,723 2,051 89 0,689 1,175 1,680 2,003
41 0,707 1,205 1,721 2,049 90 0,689 1,175 1,680 2,002
42 0,706 1,204 1,719 2,047 91 0,689 1,175 1,680 2,002
43 0,706 1,202 1,718 2,045 92 0,689 1,175 1,679 2,001
44 0,705 1,201 1,716 2,043 93 0,689 1,174 1,679 2,001
45 0,704 1,200 1,714 2,041 94 0,689 1,174 1,679 2,000
46 0,703 1,199 1,713 2,039 95 0,689 1,174 1,679 2,000
47 0,703 1,198 1,712 2,038 96 0,688 1,174 1,678 2,000
48 0,702 1,197 1,710 2,036 97 0,688 1,174 1,678 1,999
49 0,702 1,196 1,709 2,035 98 0,688 1,173 1,678 1,999
50 0,701 1,195 1,708 2,033 99 0,688 1,173 1,677 1,998
100 0,688 1,173 1,677 1,998

Table A12. k2 for two-sided statistical tolerance, standard deviation: known and confidence
level = 95 %

NB-CPD/SG10/12/091 Page 41 of 74
n fractile : p n fractile : p
0,50 0,75 0,90 0,95 0,50 0,75 0,90 0,95
2 1,243 2,057 2,870 3,376 51 0,686 1,170 1,673 1,993
3 0,943 1,582 2,229 2,635 52 0,686 1,169 1,672 1,992
4 0,853 1,441 2,040 2,416 53 0,685 1,169 1,672 1,992
5 0,809 1,370 1,946 2,308 54 0,685 1,169 1,671 1,991
6 0,782 1,328 1,889 2,243 55 0,685 1,169 1,671 1,991
7 0,765 1,300 1,851 2,199 56 0,685 1,168 1,670 1,990
8 0,752 1,279 1,823 2,168 57 0,685 1,168 1,670 1,990
9 0,743 1,264 1,802 2,143 58 0,684 1,168 1,669 1,989
10 0,735 1,252 1,786 2,124 59 0,684 1,167 1,669 1,989
11 0,730 1,242 1,772 2,109 60 0,684 1,167 1,668 1,988
12 0,725 1,234 1,761 2,096 61 0,684 1,167 1,668 1,988
13 0,721 1,227 1,752 2,086 62 0,684 1,167 1,667 1,987
14 0,717 1,222 1,744 2,077 63 0,684 1,166 1,667 1,987
15 0,714 1,217 1,738 2,069 64 0,684 1,166 1,667 1,986
16 0,712 1,212 1,732 2,062 65 0,684 1,166 1,667 1,986
17 0,709 1,209 1,727 2,056 66 0,683 1,166 1,666 1,986
18 0,707 1,205 1,722 2,051 67 0,683 1,166 1,666 1,985
19 0,706 1,202 1,718 2,046 68 0,683 1,165 1,666 1,985
20 0,704 1,200 1,714 2,042 69 0,683 1,165 1,665 1,984
21 0,703 1,198 1,711 2,038 70 0,683 1,165 1,665 1,984
22 0,701 1,195 1,708 2,034 71 0,683 1,165 1,665 1,984
23 0,700 1,193 1,706 2,031 72 0,683 1,165 1,664 1,983
24 0,699 1,191 1,703 2,028 73 0,683 1,164 1,664 1,983
25 0,698 1,190 1,701 2,026 74 0,683 1,164 1,664 1,983
26 0,697 1,188 1,698 2,023 75 0,683 1,164 1,664 1,983
27 0,697 1,187 1,696 2,021 76 0,682 1,164 1,663 1,982
28 0,696 1,186 1,694 2,018 77 0,682 1,164 1,663 1,982
29 0,695 1,185 1,693 2,016 78 0,682 1,163 1,663 1,982
30 0,694 1,183 1,691 2,014 79 0,682 1,163 1,662 1,981
31 0,693 1,182 1,690 2,013 80 0,682 1,163 1,662 1,981
32 0,693 1,181 1,689 2,011 81 0,682 1,163 1,662 1,981
33 0,692 1,181 1,687 2,010 82 0,682 1,163 1,662 1,981
34 0,692 1,180 1,686 2,008 83 0,682 1,163 1,662 1,980
35 0,691 1,179 1,685 2,007 84 0,682 1,163 1,662 1,980
36 0,691 1,178 1,684 2,006 85 0,682 1,163 1,662 1,980
37 0,690 1,177 1,683 2,005 86 0,681 1,162 1,661 1,980
38 0,690 1,177 1,682 2,003 87 0,681 1,162 1,661 1,980
39 0,689 1,176 1,681 2,002 88 0,681 1,162 1,661 1,979
40 0,689 1,175 1,680 2,001 89 0,681 1,162 1,661 1,979
41 0,689 1,174 1,679 2,000 90 0,681 1,162 1,661 1,979
42 0,689 1,174 1,678 1,999 91 0,681 1,162 1,661 1,979
43 0,688 1,173 1,678 1,999 92 0,681 1,162 1,661 1,979
44 0,688 1,173 1,677 1,998 93 0,681 1,161 1,660 1,978
45 0,688 1,172 1,676 1,997 94 0,681 1,161 1,660 1,978
46 0,688 1,172 1,675 1,996 95 0,681 1,161 1,660 1,978
47 0,687 1,171 1,675 1,995 96 0,681 1,161 1,660 1,978
48 0,687 1,171 1,674 1,995 97 0,681 1,161 1,660 1,978
49 0,686 1,170 1,674 1,994 98 0,681 1,160 1,659 1,977
50 0,686 1,170 1,673 1,993 99 0,681 1,160 1,659 1,977
100 0,681 1,160 1,659 1,977

Table A13. k2 for two-sided statistical tolerance, standard deviation: unknown and
confidence level = 50 %

NB-CPD/SG10/12/091 Page 42 of 74
n fractile : p n fractile : p
0,50 0,75 0,90 0,95 0,50 0,75 0,90 0,95
2 2,674 4,394 6,109 7,178 51 0,736 1,255 1,793 2,137
3 1,492 2,487 3,489 4,117 52 0,736 1,254 1,792 2,135
4 1,211 2,036 2,872 3,397 53 0,735 1,252 1,790 2,133
5 1,083 1,829 2,590 3,069 54 0,734 1,251 1,789 2,131
6 1,009 1,709 2,425 2,877 55 0,734 1,250 1,787 2,129
7 0,961 1,630 2,316 2,750 56 0,733 1,249 1,785 2,127
8 1,926 1,573 2,238 2,659 57 0,732 1,248 1,784 2,125
9 0,900 1,530 2,179 2,590 58 0,731 1,246 1,782 2,123
10 0,880 1,497 2,133 2,536 59 0,731 1,245 1,781 2,121
11 0,864 1,469 2,095 2,492 60 0,730 1,244 1,779 2,119
12 0,850 1,447 2,064 2,456 61 0,730 1,243 1,778 2,118
13 0,839 1,428 2,038 2,425 62 0,729 1,242 1,776 2,116
14 0,829 1,412 2,015 2,399 63 0,729 1,242 1,775 2,115
15 0,821 1,398 1,996 2,376 64 0,728 1,241 1,774 2,113
16 0,814 1,386 1,979 2,356 65 0,728 1,240 1,773 2,112
17 0,807 1,375 1,964 2,338 66 0,727 1,239 1,771 2,111
18 0,802 1,366 1,950 2,322 67 0,727 1,238 1,770 2,109
19 0,797 1,357 1,938 2,308 68 0,726 1,238 1,769 2,108
20 0,792 1,349 1,927 2,295 69 0,726 1,237 1,767 2,106
21 0,788 1,343 1,918 2,284 70 0,725 1,236 1,766 2,105
22 0,784 1,336 1,908 2,273 71 0,725 1,235 1,765 2,104
23 0,781 1,331 1,900 2,264 72 0,724 1,235 1,764 2,103
24 0,777 1,325 1,892 2,254 73 0,724 1,234 1,763 2,101
25 0,774 1,320 1,886 2,246 74 0,723 1,233 1,762 2,100
26 0,771 1,315 1,879 2,238 75 0,723 1,233 1,762 2,099
27 0,769 1,311 1,873 2,231 76 0,723 1,232 1,761 2,098
28 0,766 1,306 1,867 2,224 77 0,722 1,231 1,760 2,097
29 0,764 1,303 1,862 2,218 78 0,722 1,230 1,759 2,095
30 0,762 1,299 1,857 2,211 79 0,721 1,230 1,758 2,094
31 0,760 1,296 1,853 2,206 80 0,721 1,229 1,757 2,093
32 0,758 1,293 1,848 2,201 81 0,721 1,228 1,756 2,092
33 0,757 1,290 1,844 2,196 82 0,720 1,228 1,755 2,091
34 0,755 1,287 1,839 2,191 83 0,720 1,227 1,755 2,090
35 0,753 1,284 1,835 2,186 84 0,720 1,227 1,754 2,089
36 0,752 1,282 1,832 2,182 85 0,720 1,226 1,753 2,089
37 0,751 1,280 1,829 2,178 86 0,719 1,225 1,752 2,088
38 0,749 1,277 1,825 2,175 87 0,719 1,225 1,751 2,087
39 0,748 1,275 1,822 2,171 88 0,719 1,224 1,751 2,086
40 0,747 1,273 1,819 2,167 89 0,718 1,224 1,750 2,085
41 0,746 1,271 1,816 2,164 90 0,718 1,223 1,749 2,084
42 0,745 1,269 1,814 2,161 91 0,718 1,223 1,748 2,083
43 0,743 1,267 1,811 2,158 92 0,717 1,222 1,748 2,082
44 0,742 1,265 1,809 2,155 93 0,717 1,222 1,747 2,082
45 0,741 1,263 1,806 2,152 94 0,717 1,221 1,746 2,081
46 0,740 1,262 1,804 2,149 95 0,717 1,221 1,746 2,080
47 0,739 1,260 1,802 2,147 96 0,716 1,221 1,745 2,079
48 0,739 1,259 1,799 2,144 97 0,716 1,220 1,744 2,078
49 0,738 1,257 1,797 2,142 98 0,716 1,220 1,743 2,078
50 0,737 1,256 1,795 2,139 99 0,715 1,219 1,743 2,077
100 0,715 1,219 1,742 2,076

Table A14. k2 for two-sided statistical tolerance, standard deviation: unknown and
confidence level = 75 %

NB-CPD/SG10/12/091 Page 43 of 74
n fractile : p n fractile : p
0,50 0,75 0,90 0,95 0,50 0,75 0,90 0,95
2 6,809 11,166 15,513 18,221 51 0,786 1,340 1,916 2,282
3 2,492 4,135 5,789 6,824 52 0,785 1,338 1,913 2,279
4 1,766 2,954 4,158 4,913 53 0,783 1,336 1,910 2,275
5 1,473 2,478 3,500 4,143 54 0,782 1,334 1,907 2,272
6 1,314 2,218 3,141 3,723 55 0,781 1,332 1,904 2,268
7 1,213 2,053 2,913 3,456 56 0,780 1,329 1,901 2,264
8 1,144 1,939 2,755 3,270 57 0,779 1,327 1,898 2,261
9 1,093 1,854 2,637 3,133 58 0,777 1,325 1,895 2,257
10 1,053 1,789 2,546 3,026 59 0,776 1,323 1,892 2,254
11 1,022 1,737 2,474 2,941 60 0,775 1,321 1,889 2,250
12 0,996 1,694 2,414 2,871 61 0,774 1,320 1,887 2,247
13 0,975 1,659 2,365 2,813 62 0,773 1,318 1,885 2,245
14 0,957 1,628 2,322 2,763 63 0,772 1,317 1,882 2,242
15 0,941 1,602 2,286 2,720 64 0,771 1,315 1,880 2,240
16 0,928 1,580 2,254 2,683 65 0,771 1,314 1,878 2,237
17 0,916 1,560 2,226 2,650 66 0,770 1,312 1,876 2,234
18 0,905 1,542 2,201 2,620 67 0,769 1,311 1,874 2,232
19 0,896 1,526 2,179 2,594 68 0,768 1,309 1,871 2,229
20 0,887 1,512 2,159 2,570 69 0,767 1,308 1,869 2,227
21 0,880 1,500 2,142 2,550 70 0,766 1,306 1,867 2,224
22 0,873 1,487 2,124 2,529 71 0,765 1,305 1,865 2,222
23 0,867 1,477 2,110 2,512 72 0,765 1,304 1,863 2,220
24 0,861 1,466 2,095 2,494 73 0,764 1,302 1,862 2,218
25 0,856 1,458 2,083 2,480 74 0,763 1,301 1,860 2,216
26 0,850 1,449 2,070 2,465 75 0,763 1,300 1,858 2,214
27 0,846 1,442 2,059 2,452 76 0,762 1,299 1,856 2,211
28 0,841 1,434 2,048 2,439 77 0,761 1,298 1,854 2,209
29 0,837 1,427 2,039 2,428 78 0,760 1,296 1,853 2,207
30 0,833 1,420 2,029 2,417 79 0,760 1,295 1,851 2,205
31 0,830 1,415 2,021 2,408 80 0,759 1,294 1,849 2,203
32 0,827 1,409 2,014 2,399 81 0,758 1,293 1,848 2,201
33 0,823 1,404 2,006 2,390 82 0,758 1,292 1,846 2,200
34 0,820 1,398 1,999 2,381 83 0,757 1,291 1,845 2,198
35 0,817 1,393 1,991 2,372 84 0,757 1,290 1,843 2,197
36 0,815 1,389 1,985 2,365 85 0,756 1,289 1,842 2,195
37 0,812 1,385 1,979 2,358 86 0,755 1,288 1,841 2,193
38 0,810 1,380 1,974 2,351 87 0,755 1,287 1,839 2,192
39 0,807 1,376 1,968 2,344 88 0,754 1,286 1,838 2,190
40 0,805 1,372 1,962 2,337 89 0,754 1,285 1,836 2,189
41 0,803 1,369 1,957 2,331 90 0,753 1,284 1,835 2,187
42 0,801 1,366 1,952 2,326 91 0,753 1,283 1,834 2,186
43 0,799 1,362 1,948 2,320 92 0,752 1,282 1,833 2,184
44 0,797 1,359 1,943 2,315 93 0,752 1,282 1,832 2,183
45 0,795 1,356 1,938 2,309 94 0,751 1,281 1,831 2,181
46 0,793 1,353 1,934 2,304 95 0,751 1,280 1,830 2,180
47 0,792 1,350 1,930 2,300 96 0,750 1,279 1,828 2,179
48 0,790 1,348 1,927 2,295 97 0,750 1,278 1,827 2,177
49 0,789 1,345 1,923 2,291 98 0,749 1,278 1,826 2,176
50 0,787 1,342 1,919 2,286 99 0,749 1,277 1,825 2,174
100 0,748 1,276 1,824 2,173

Table A15. k2 for two-sided statistical tolerance, standard deviation: unknown and
confidence level = 90 %

NB-CPD/SG10/12/091 Page 44 of 74
n fractile : p n fractile : p
0,50 0,75 0,90 0,95 0,50 0,75 0,90 0,95
2 13,652 22,383 31,093 36,520 51 0,819 1,396 1,996 2,377
3 3,585 5,938 8,306 9,789 52 0,818 1,393 1,992 2,373
4 2,288 3,819 5,369 6,342 53 0,816 1,391 1,988 2,368
5 1,812 3,041 4,291 5,077 54 0,814 1,388 1,984 2,364
6 1,566 2,639 3,733 4,423 55 0,813 1,385 1,980 2,359
7 1,416 2,392 3,390 4,020 56 0,811 1,382 1,976 2,354
8 1,314 2,224 3,157 3,746 57 0,809 1,379 1,972 2,350
9 1,240 2,101 2,987 3,546 58 0,807 1,377 1,968 2,345
10 1,183 2,008 2,857 3,394 59 0,806 1,374 1,964 2,341
11 1,139 1,935 2,754 3,273 60 0,804 1,371 1,960 2,336
12 1,103 1,875 2,671 3,175 61 0,803 1,369 1,957 2,333
13 1,074 1,825 2,602 3,094 62 0,802 1,367 1,954 2,329
14 1,049 1,784 2,543 3,025 63 0,800 1,365 1,951 2,326
15 1,027 1,748 2,493 2,965 64 0,799 1,363 1,948 2,322
16 1,009 1,717 2,449 2,914 65 0,798 1,361 1,946 2,319
17 0,992 1,689 2,411 2,869 66 0,797 1,359 1,943 2,315
18 0,978 1,665 2,377 2,829 67 0,796 1,357 1,940 2,312
19 0,965 1,644 2,347 2,793 68 0,794 1,355 1,937 2,308
20 0,954 1,625 2,319 2,761 69 0,793 1,353 1,934 2,305
21 0,944 1,608 2,296 2,733 70 0,792 1,351 1,931 2,301
22 0,934 1,591 2,272 2,705 71 0,791 1,349 1,929 2,298
23 0,926 1,577 2,253 2,682 72 0,790 1,348 1,927 2,296
24 0,918 1,563 2,233 2,659 73 0,789 1,346 1,924 2,293
25 0,911 1,552 2,217 2,639 74 0,788 1,345 1,922 2,290
26 0,904 1,540 2,200 2,619 75 0,788 1,343 1,920 2,288
27 0,898 1,530 2,186 2,602 76 0,787 1,341 1,918 2,285
28 0,892 1,519 2,171 2,585 77 0,786 1,340 1,916 2,282
29 0,887 1,511 2,159 2,570 78 0,785 1,338 1,913 2,279
30 0,881 1,502 2,146 2,555 79 0,784 1,337 1,911 2,277
31 0,877 1,495 2,136 2,543 80 0,783 1,335 1,909 2,274
32 0,873 1,488 2,126 2,531 81 0,782 1,334 1,907 2,272
33 0,868 1,480 2,115 2,519 82 0,782 1,332 1,905 2,270
34 0,864 1,473 2,105 2,507 83 0,781 1,331 1,903 2,267
35 0,860 1,466 2,095 2,495 84 0,780 1,330 1,901 2,265
36 0,857 1,460 2,087 2,486 85 0,780 1,329 1,900 2,263
37 0,854 1,455 2,079 2,477 86 0,779 1,327 1,898 2,261
38 0,850 1,449 2,072 2,467 87 0,778 1,326 1,896 2,259
39 0,847 1,444 2,064 2,458 88 0,777 1,325 1,894 2,256
40 0,844 1,438 2,056 2,449 89 0,777 1,323 1,892 2,254
41 0,841 1,434 2,050 2,442 90 0,776 1,322 1,890 2,252
42 0,839 1,430 2,044 2,434 91 0,775 1,321 1,889 2,250
43 0,836 1,425 2,037 2,427 92 0,775 1,320 1,887 2,248
44 0,834 1,421 2,031 2,419 93 0,774 1,319 1,886 2,247
45 0,831 1,417 2,025 2,412 94 0,773 1,318 1,884 2,245
46 0,829 1,413 2,020 2,406 95 0,773 1,317 1,883 2,243
47 0,827 1,410 2,015 2,400 96 0,772 1,316 1,881 2,241
48 0,825 1,406 2,010 2,394 97 0,771 1,315 1,880 2,239
49 0,823 1,403 2,005 2,388 98 0,770 1,314 1,878 2,238
50 0,821 1,399 2,000 2,382 99 0,770 1,313 1,877 2,236
100 0,769 1,312 1,875 2,234

Table A16. k2 for two-sided statistical tolerance, standard deviation: unknown and
confidence level = 95 %

NB-CPD/SG10/12/091 Page 45 of 74
Annex B Examples of statistical evaluation

Example 1

Example of statistical analysis of compressive strength using batch control.


The fractile p = 50%
The confidence level = 95%
The number of series of inspection lots is l = 1.
One-sided tolerance interval, lower limit
The declared mean compressive strength is 15 N/mm

For the first and the following inspection lots a sample size of 6 samples are taken and tested and
evaluated inspection lot by inspection lot (xm (equation 1), standard deviation ss (equation 2) and xest
(equation 3) according to clause 5.6.6 using k1,u taken from Annex A Table A8 (p: 50 % and : 95
%)).

For the first 42 samples (1 7 inspection lot), the standard deviation of the population is considered
to be unknown and the k1,u factor taken from Annex A Table A8 (p: 50 % and : 95 %) is 0,823. For
the inspection lots 8 20 the standard deviation can be considered as known, but the used
acceptance coefficient is corrected (kc). The acceptance coefficient for the known standard
deviation k1,k is taken from Annex A Table A4 (p: 50 % and : 95 %) and is 0,672. The corrected
acceptance coefficient kc is calculated by a linear interpolation between the acceptance coefficient
k1,u and k1,k taking into account the considered inspection lot. The known standard deviation is
calculated based on the first 42 test results.

From inspection lot 21 and so on 6 samples are taken from each next inspection lot and the test
results are evaluated inspection lot by inspection lot (xm (equation 1), known standard deviation
and xest (equation 6) according to clause 5.6.6 using k1,k taken from Annex A Table A4 (p: 50 % and
: 95 %)).

After each evaluation the result has to be compared with the lower limit value (e.g. the declared
value) decided by the manufacturer.

If there is a non-conformity due to great differences between the test results, the estimated value is
highlighted by a red signal at the right side (batch 21). A non-conforming inspection lot has to be
treated separately as described in the text.

NB-CPD/SG10/12/091 Page 46 of 74
EXAMPLE1
ONE SIDED TOLERANCE INTERVAL-lower limit METHODA:useatleast6testresultsperinspectionlot

fractilep 50 Startcorrection 7 Seriesofinspectionlots 1 DeclaredValue 15


confidencelevel 95 Endcorrection 20 1,409 1,858

Equation
Inspectionlot test1 test2 test3 test4 test5 test6 n Xm Ss k1,u kc k1,k Xest Xest Xest
OK?
1 18,1 17,9 18,3 19,4 17,7 19,2 6 18,43 0,70 0,823 17,85
2 16,1 18,4 18,8 17,6 15,7 17 6 17,27 1,24 0,823 16,25
3 17,7 19,1 17,9 18,1 15,7 18,1 6 17,77 1,12 0,823 16,84
4 21,4 20,8 19,6 18,5 18,6 18,1 6 19,50 1,35 0,823 18,39
5 19,5 20,8 19,7 21,1 19,4 18,6 6 19,85 0,94 0,823 19,08
6 19,9 19,3 20,1 18,8 21,1 20,6 6 19,97 0,84 0,823 19,28
7 17,2 19,2 17,0 18,2 19,4 17,1 6 18,02 1,09 0,823 0,823 0,672 1,409 17,12 16,86
8 20,2 17,8 16,3 19,9 21,9 22,6 6 19,78 2,39 0,811 18,64
9 21,0 14,7 19,0 18,9 18,8 19,3 6 18,62 2,09 0,800 17,49
10 20,2 17,3 22,3 21,3 22,4 22,6 6 21,02 2,03 0,788 19,91
11 20,0 22,8 21,2 19,9 20,1 22,6 6 21,10 1,33 0,777 20,01
12 23,8 20,6 20,8 17,1 21 15,5 6 19,80 3,00 0,765 18,72
13 21,1 21,0 20,1 19,1 23,4 21,5 6 21,03 1,44 0,753 19,97
14 21,0 16,5 17,2 15,8 21,7 19,5 6 18,62 2,47 0,742 17,57
15 18,8 20,5 20,3 20,1 21,7 17,1 6 19,75 1,59 0,730 18,72
16 19,7 19,5 20,0 18,1 20,3 17,2 6 19,13 1,21 0,718 18,12
17 20,4 19,6 20,1 20,8 14,5 18,2 6 18,93 2,35 0,707 17,94
18 20,5 21,4 20,4 22,4 19,1 18,6 6 20,40 1,41 0,695 19,42
19 19,8 20,5 18,1 19,2 19,1 18,8 6 19,25 0,83 0,684 18,29
20 19,7 17,6 16,9 20,9 18,4 18 6 18,58 1,47 0,672 0,672 1,858 17,33 17,33 OK
21 16,5 14,2 15,3 16,4 14,3 16,4 6 15,52 1,08 0,672 14,27 OK
22 21,3 21,4 21,3 18,9 21,2 20,8 6 20,82 0,96 0,672 19,57 OK
23 19,3 17,5 18,3 17,2 19,1 16,7 6 18,02 1,06 0,672 16,77 OK
24 18,5 21,8 19,4 17,2 21,3 17,4 6 19,27 1,94 0,672 18,02 OK

NB-CPD/SG10/12/091 Page 47 of 74
Example 2

Example of statistical analysis of compressive strength using batch control.

This example is similar to example 1. The only difference is that the declared compressive strength
is a 5 % characteristic value.
The fractile p = 95%
The confidence level = 95%
The number of series of inspection lots is l = 1.
One-sided tolerance interval, lower limit
The declared 5 % characteristic compressive strength is 10 N/mm

For the first and the following inspection lots a sample size of 6 samples are taken and tested and
evaluated inspection lot by inspection lot (xm (equation 1), standard deviation ss (equation 2) and xest
(equation 3) according to clause 5.6.6 using k1,u taken from Annex A Table A8 (p: 95 % and : 95
%)).

For the first 42 samples (1 7 inspection lot), the standard deviation of the population is considered
to be unknown and the k1,u factor taken from Annex A Table A8 (p: 95 % and : 95 %) is 3,708. For
the inspection lots 8 20 the standard deviation can be considered as known, but the used
acceptance coefficient is corrected (kc). The acceptance coefficient for the known standard
deviation k1,k is taken from Annex A Table A4 (p: 95 % and : 95 %) and is 2,317. The corrected
acceptance coefficient kc is calculated by a linear interpolation between the acceptance coefficient
k1,u and k1,k taking into account the considered inspection lot. The known standard deviation is
calculated based on the first 42 test results.

From inspection lot 21 and so on 6 samples are taken from each next inspection lot and the test
results are evaluated inspection lot by inspection lot (xm (equation 1), known standard deviation
and xest (equation 6) according to clause 5.6.6 using k1,k taken from Annex A Table A4 (p: 95 % and
: 95 %)).

After each evaluation the result has to be compared with the lower limit value (e.g. the declared
value) decided by the manufacturer.

If there is a non-conformity due to great differences between the test results, the estimated value is
highlighted by a red signal at the right side. A non-conforming inspection lot has to be treated
separately as described in the text.

NB-CPD/SG10/12/091 Page 48 of 74
EXAMPLE2
ONE SIDED TOLERANCE INTERVAL-lower limit METHODA:useatleast6testresultsperinspectionlot

fractilep 95 Startcorrection 7 Seriesofinspectionlots 1 DeclaredValue 10


confidencelevel 95 Endcorrection 20 1,409 1,858

Equation
Inspectionlot test1 test2 test3 test4 test5 test6 n Xm Ss k1,u kc k1,k Xest Xest Xest
OK?
1 18,1 17,9 18,3 19,4 17,7 19,2 6 18,43 0,70 3,708 15,83
2 16,1 18,4 18,8 17,6 15,7 17 6 17,27 1,24 3,708 12,69
3 17,7 19,1 17,9 18,1 15,7 18,1 6 17,77 1,12 3,708 13,61
4 21,4 20,8 19,6 18,5 18,6 18,1 6 19,50 1,35 3,708 14,50
5 19,5 20,8 19,7 21,1 19,4 18,6 6 19,85 0,94 3,708 16,38
6 19,9 19,3 20,1 18,8 21,1 20,6 6 19,97 0,84 3,708 16,86
7 17,2 19,2 17,0 18,2 19,4 17,1 6 18,02 1,09 3,708 3,708 2,317 1,409 13,99 12,79
8 20,2 17,8 16,3 19,9 21,9 22,6 6 19,78 2,39 3,601 14,71
9 21,0 14,7 19,0 18,9 18,8 19,3 6 18,62 2,09 3,494 13,69
10 20,2 17,3 22,3 21,3 22,4 22,6 6 21,02 2,03 3,387 16,24
11 20,0 22,8 21,2 19,9 20,1 22,6 6 21,10 1,33 3,280 16,48
12 23,8 20,6 20,8 17,1 21 15,5 6 19,80 3,00 3,173 15,33
13 21,1 21,0 20,1 19,1 23,4 21,5 6 21,03 1,44 3,066 16,71
14 21,0 16,5 17,2 15,8 21,7 19,5 6 18,62 2,47 2,959 14,45
15 18,8 20,5 20,3 20,1 21,7 17,1 6 19,75 1,59 2,852 15,73
16 19,7 19,5 20,0 18,1 20,3 17,2 6 19,13 1,21 2,745 15,26
17 20,4 19,6 20,1 20,8 14,5 18,2 6 18,93 2,35 2,638 15,22
18 20,5 21,4 20,4 22,4 19,1 18,6 6 20,40 1,41 2,531 16,83
19 19,8 20,5 18,1 19,2 19,1 18,8 6 19,25 0,83 2,424 15,83
20 19,7 17,6 16,9 20,9 18,4 18 6 18,58 1,47 2,317 2,317 1,858 14,28 14,28 OK
21 16,5 14,2 15,3 16,4 14,3 16,4 6 15,52 1,08 2,317 11,21 OK
22 21,3 21,4 21,3 18,9 21,2 20,8 6 20,82 0,96 2,317 16,51 OK
23 19,3 17,5 18,3 17,2 19,1 16,7 6 18,02 1,06 2,317 13,71 OK
24 18,5 21,8 19,4 17,2 21,3 17,4 6 19,27 1,94 2,317 14,96 OK

NB-CPD/SG10/12/091 Page 49 of 74
Example 3

Example of statistical analysis of compressive strength using batch control.

This example is similar to example 1. The only difference is, that the used confidence level is 75 %.
The fractile p = 50%
The confidence level = 75%
The number of series of inspection lots is l = 1.
One-sided tolerance interval, lower limit
The declared mean compressive strength is 15 N/mm

For the first and the following inspection lots a sample size of 6 samples are taken and tested and
evaluated inspection lot by inspection lot (xm (equation 1), standard deviation ss (equation 2) and xest
(equation 3) according to clause 5.6.6 using k1,u taken from Annex A Table A6 (p: 50 % and : 75
%)).

For the first 42 samples (1 7 inspection lot), the standard deviation of the population is considered
to be unknown and the k1,u factor taken from Annex A Table A6 (p: 50 % and : 75 %) is 0,297. For
the inspection lots 8 20 the standard deviation can be considered as known, but the used
acceptance coefficient is corrected (kc). The acceptance coefficient for the known standard
deviation k1,k is taken from Annex A Table A2 (p: 50 % and : 75 %) and is 0,276. The corrected
acceptance coefficient kc is calculated by a linear interpolation between the acceptance coefficient
k1,u and k1,k taking into account the considered inspection lot. The known standard deviation is
calculated based on the first 42 test results.

From inspection lot 21 and so on 6 samples are taken from each next inspection lot and the test
results are evaluated inspection lot by inspection lot (xm (equation 1), known standard deviation
and xest (equation 6) according to clause 5.6.6 using k1,k taken from Annex A Table A2 (p: 50 % and
: 75 %)).

After each evaluation the result has to be compared with the lower limit value (e.g. the declared
value) decided by the manufacturer.

If there is a non-conformity due to great differences between the test results, the estimated value is
highlighted by a red signal at the right side. A non-conforming inspection lot has to be treated
separately as described in the text.

NB-CPD/SG10/12/091 Page 50 of 74
EXAMPLE3
ONE SIDED TOLERANCE INTERVAL-lower limit METHODA:useatleast6testresultsperinspectionlot

fractilep 50 Startcorrection 7 Seriesofinspectionlots 1 DeclaredValue 15


confidencelevel 75 Endcorrection 20 1,409 1,858

Equation
Inspectionlot test1 test2 test3 test4 test5 test6 n Xm Ss k1,u kc k1,k Xest Xest Xest
OK?
1 18,1 17,9 18,3 19,4 17,7 19,2 6 18,43 0,70 0,297 18,22
2 16,1 18,4 18,8 17,6 15,7 17 6 17,27 1,24 0,297 16,90
3 17,7 19,1 17,9 18,1 15,7 18,1 6 17,77 1,12 0,297 17,43
4 21,4 20,8 19,6 18,5 18,6 18,1 6 19,50 1,35 0,297 19,10
5 19,5 20,8 19,7 21,1 19,4 18,6 6 19,85 0,94 0,297 19,57
6 19,9 19,3 20,1 18,8 21,1 20,6 6 19,97 0,84 0,297 19,72
7 17,2 19,2 17,0 18,2 19,4 17,1 6 18,02 1,09 0,297 0,297 0,276 1,409 17,69 17,60
8 20,2 17,8 16,3 19,9 21,9 22,6 6 19,78 2,39 0,295 19,37
9 21,0 14,7 19,0 18,9 18,8 19,3 6 18,62 2,09 0,294 18,20
10 20,2 17,3 22,3 21,3 22,4 22,6 6 21,02 2,03 0,292 20,60
11 20,0 22,8 21,2 19,9 20,1 22,6 6 21,10 1,33 0,291 20,69
12 23,8 20,6 20,8 17,1 21 15,5 6 19,80 3,00 0,289 19,39
13 21,1 21,0 20,1 19,1 23,4 21,5 6 21,03 1,44 0,287 20,63
14 21,0 16,5 17,2 15,8 21,7 19,5 6 18,62 2,47 0,286 18,21
15 18,8 20,5 20,3 20,1 21,7 17,1 6 19,75 1,59 0,284 19,35
16 19,7 19,5 20,0 18,1 20,3 17,2 6 19,13 1,21 0,282 18,74
17 20,4 19,6 20,1 20,8 14,5 18,2 6 18,93 2,35 0,281 18,54
18 20,5 21,4 20,4 22,4 19,1 18,6 6 20,40 1,41 0,279 20,01
19 19,8 20,5 18,1 19,2 19,1 18,8 6 19,25 0,83 0,278 18,86
20 19,7 17,6 16,9 20,9 18,4 18 6 18,58 1,47 0,276 0,276 1,858 18,07 18,07 OK
21 16,5 14,2 15,3 16,4 14,3 16,4 6 15,52 1,08 0,276 15,00 OK
22 21,3 21,4 21,3 18,9 21,2 20,8 6 20,82 0,96 0,276 20,30 OK
23 19,3 17,5 18,3 17,2 19,1 16,7 6 18,02 1,06 0,276 17,50 OK
24 18,5 21,8 19,4 17,2 21,3 17,4 6 19,27 1,94 0,276 18,75 OK

NB-CPD/SG10/12/091 Page 51 of 74
Example 4

Example of statistical analysis of compressive strength using Rolling inspection.


The fractile p = 50%
The confidence level = 95%
The number of series of inspection lots is l = 4.
One-sided tolerance interval, lower limit
The declared mean compressive strength is 15 N/mm

For the first inspection lot a sample size of 3 samples are taken and tested and evaluated (xm
(equation 1), standard deviation ss (equation 2) and xest (equation 3) according to clause 5.6.6 using
k1,u taken from Annex A Table A8 (p: 50 % and : 95 %)). For the next and the following 2
inspection lots 3 additional samples are taken and tested and evaluated together with the ones from
the previous inspection lots (xm (equation 1), standard deviation ss (equation 2) and xest (equation 3)
according to clause 5.6.6 using k1,u taken from Annex A Table A8 (p: 50 % and : 95 %)). By doing
so the spot sample size evaluated together is gradually increased from 3 to 12 samples.

From then on, 3 additional samples are taken from each next inspection lot and evaluated together
with the ones from the previous inspection lots (xm (equation 1), standard deviation ss (equation 2)
and xest (equation 3) according to clause 5.6.6 using k1,u taken from Annex A Table A8 (p: 50 % and
: 95 %)) but the spot sample size is limited to the last 12 samples. The spot sample size continues
to be 12.

For the first 21 samples (1 7 inspection lot), the standard deviation of the population is considered
to be unknown and the k1,u factor taken from Annex A Table A8 (p: 50 % and : 95 %) is 0,519. For
the inspection lots 8 20 the standard deviation can be considered as known, but the used
acceptance coefficient is corrected (kc). The acceptance coefficient for the known standard
deviation k1,k is taken from Annex A Table A4 (p: 50 % and : 95 %) and is 0,475. The corrected
acceptance coefficient kc is calculated by a linear interpolation between the acceptance coefficient
k1,u and k1,k taking into account the considered inspection lot. The known standard deviation is
calculated based on the first 21 test results.

From inspection lot 21 and so on 3 additional samples are taken from each next inspection lot and
the test results are evaluated together with the ones from the previous inspection lots (xm (equation
1), known standard deviation and xest (equation 6) according to clause 5.6.6 using k1,k taken from
Annex A Table A4 (p: 50 % and : 95 %)) and the spot sample size is still limited to the last 12
samples.

After each evaluation the result has to be compared with the lower limit value (e.g. the declared
value) decided by the manufacturer.

Part of the evaluation is also to check that the standard deviation ss of the spot sample corresponds
to the following equation:

0,63 ss 1,37

In the last column it is indicated whether the mentioned equation fits or does not fit.

If there is a non-conformity due to great differences between the test results, the estimated value is
highlighted by a red signal at the right side (batch 1). A non-conforming inspection lot has to be
treated separately as described in the text.

NB-CPD/SG10/12/091 Page 52 of 74
EXAMPLE4
ONE SIDED TOLERANCE INTERVAL-lower limit METHODB:useatleast3testresultsperinspectionlot

fractilep 50 Startcorrection 7 Seriesofinspectionlots 4 DeclaredValue 15


confidencelevel 95 Endcorrection 20 2,325 1,762

Equation
Inspectionlot test1 test2 test3 n Xm Ss k1,u kc k1,k Xest Xest Xest
OK?
1 20,2 17,8 16,3 3 18,10 1,97 1,686 14,78
2 21,0 14,7 19,0 6 18,17 2,39 0,823 16,20
3 20,2 17,3 22,3 9 18,76 2,43 0,620 17,25
4 20,0 22,8 21,2 12 19,40 2,45 0,519 18,13
5 23,8 20,6 20,8 12 20,31 2,45 0,519 19,03
6 21,1 21,0 20,1 12 20,93 1,63 0,519 20,09
7 21,0 16,5 17,2 12 20,51 2,02 0,519 0,519 0,475 2,325 19,46 19,30
8 18,8 20,5 20,3 12 20,14 1,91 0,516 18,94
9 19,7 19,5 20,0 12 19,64 1,47 0,512 18,45
10 20,4 19,6 20,1 12 19,47 1,35 0,509 18,28
11 20,5 21,4 20,4 12 20,10 0,65 0,505 18,93
12 19,8 20,5 18,1 12 20,00 0,79 0,502 18,83
13 19,7 17,6 16,9 12 19,58 1,35 0,499 18,42
14 18,7 20,0 19,7 12 19,44 1,34 0,495 18,29
15 21,3 21,4 21,3 12 19,58 1,49 0,492 18,44
16 19,3 17,5 18,3 12 19,31 1,55 0,489 18,17
17 18,5 21,8 19,4 12 19,77 1,41 0,485 18,64
18 19,9 21,9 22,6 12 20,27 1,66 0,482 19,15
19 18,9 18,8 19,3 12 19,68 1,59 0,478 18,57
20 21,3 22,4 22,6 12 20,62 1,62 0,475 0,475 1,762 19,78 19,78 OK
21 19,9 20,1 22,6 12 20,86 1,53 0,475 20,02 OK
22 17,1 21,0 15,5 12 19,96 2,21 0,475 19,12 OK
23 19,1 23,4 21,5 12 20,54 2,37 0,475 19,70 OK
24 15,8 21,7 19,5 12 19,77 2,54 0,475 18,93 OK
25 20,1 21,7 17,1 12 19,46 2,57 0,475 18,62 OK
26 18,1 20,3 17,2 12 19,63 2,27 0,475 18,79 OK
27 20,8 14,5 18,2 12 18,75 2,32 0,475 17,91 OK
28 22,4 19,1 18,6 12 19,01 2,21 0,475 18,17 OK
29 19,2 19,1 18,8 12 18,86 1,94 0,475 18,02 OK
30 20,9 18,4 18,0 12 19,00 1,93 0,475 18,16 OK

NB-CPD/SG10/12/091 Page 53 of 74
Example 5

Example of statistical analysis of compressive strength using "Rolling" inspection.

This example is similar to example 4. The only difference is the size of the series of inspection lots.
In example 4 the series is l = 4 and in example 5 the series is l = 5.
The fractile p = 50%
The confidence level = 95%
The number of series of inspection lots is l = 5.
One-sided tolerance interval, lower limit
The declared mean compressive strength is 15 N/mm

For the first inspection lot a sample size of 3 samples are taken and tested and evaluated (xm
(equation 1), standard deviation ss (equation 2) and xest (equation 3) according to clause 5.6.6 using
k1,u taken from Annex A Table A8 (p: 50 % and : 95 %)). For the next and the following 3
inspection lots 3 additional samples are taken and tested and evaluated together with the ones from
the previous inspection lots (xm (equation 1), standard deviation ss (equation 2) and xest (equation 3)
according to clause 5.6.6 using k1,u taken from Annex A Table A8 (p: 50 % and : 95 %)). By doing
so the spot sample size evaluated together is gradually increased from 3 to 15 samples.

From then on, 3 additional samples are taken from each next inspection lot and evaluated together
with the ones from the previous inspection lots (xm (equation 1), standard deviation ss (equation 2)
and xest (equation 3) according to clause 5.6.6 using k1,u taken from Annex A Table A8 (p: 50 % and
: 95 %)) but the spot sample size is limited to the last 15 samples. The spot sample size continues
to be 15.

For the first 21 samples (1 7 inspection lot), the standard deviation of the population is considered
to be unknown and the k1,u factor taken from Annex A Table A8 (p: 50 % and : 95 %) is 0,455. For
the inspection lots 8 20 the standard deviation can be considered as known, but the used
acceptance coefficient is corrected (kc). The acceptance coefficient for the known standard
deviation k1,k is taken from Annex A Table A4 (p: 50 % and : 95 %) and is 0,425. The corrected
acceptance coefficient kc is calculated by a linear interpolation between the acceptance coefficient
k1,u and k1,k taking into account the considered inspection lot. The known standard deviation is
calculated based on the first 21 test results.

From inspection lot 21 and so on 3 additional samples are taken from each next inspection lot and
the test results are evaluated together with the ones from the previous inspection lots (xm (equation
1), known standard deviation and xest (equation 6) according to clause 5.6.6 using k1,k taken from
Annex A Table A4 (p: 50 % and : 95 %)) and the spot sample size is still limited to the last 15
samples.

After each evaluation the result has to be compared with the lower limit value (e.g. the declared
value) decided by the manufacturer.

Part of the evaluation is also to check that the standard deviation ss of the spot sample corresponds
to the following equation:

0,63 ss 1,37

In the last column it is indicated whether the mentioned equation fits or does not fit.

NB-CPD/SG10/12/091 Page 54 of 74
If there is a non-conformity due to great differences between the test results, the estimated value is
highlighted by a red signal at the right side (batch 1 and 22). A non-conforming inspection lot has
to be treated separately as described in the text.

EXAMPLE5
ONE SIDED TOLERANCE INTERVAL-lower limit METHODB:useatleast3testresultsperinspectionlot

fractilep 50 Startcorrection 7 Seriesofinspectionlots 5 DeclaredValue 15


confidencelevel 95 Endcorrection 20 2,490 2,461

Equation
Inspectionlot test1 test2 test3 n Xm Ss k1,u kc k1,k Xest Xest Xest
OK?
1 20,2 17,8 16,3 3 18,10 1,97 1,686 14,78
2 21,0 14,7 19,0 6 18,17 2,39 0,823 16,20
3 20,2 17,3 22,3 9 18,76 2,43 0,620 17,25
4 20,0 22,8 21,2 12 19,40 2,45 0,519 18,13
5 23,8 20,6 20,8 15 19,87 2,47 0,455 18,74
6 21,1 21,0 20,1 15 20,39 2,19 0,455 19,40
7 15,4 16,5 17,2 15 20,02 2,40 0,455 0,455 0,425 2,490 18,93 18,89
8 18,8 20,5 20,3 15 20,01 2,24 0,453 18,88
9 19,7 19,5 20,0 15 19,69 2,06 0,450 18,57
10 20,4 19,6 20,1 15 19,35 1,68 0,448 18,23
11 20,5 21,4 20,4 15 19,35 1,69 0,446 18,24
12 19,8 20,5 18,1 15 19,97 0,79 0,443 18,87
13 19,7 17,6 16,9 15 19,61 1,20 0,441 18,51
14 18,7 20,0 19,7 15 19,56 1,22 0,439 18,47
15 14,7 15,8 16,2 15 18,67 2,02 0,437 17,58
16 19,3 17,5 14,4 15 17,93 1,98 0,434 16,85
17 20,8 15,3 19,4 15 17,73 2,10 0,432 16,66
18 15,2 16,1 14,5 15 17,17 2,26 0,430 16,10
19 14,1 16,1 15,4 15 16,32 2,03 0,427 15,26
20 14,4 15,5 15,7 15 16,25 2,06 0,425 0,425 2,461 15,20 15,20 OK
21 19,9 16,1 14,3 15 16,19 2,11 0,425 15,14 OK
22 14,9 16,0 15,5 15 15,58 1,38 0,425 14,53 OK
23 16,0 23,4 21,5 15 16,59 2,76 0,425 15,54 OK
24 15,8 21,7 19,5 15 17,35 2,99 0,425 16,30 OK
25 20,1 21,7 17,1 15 18,23 2,99 0,425 17,19 OK
26 18,1 20,3 17,2 15 18,59 2,75 0,425 17,54 OK
27 20,8 14,5 18,2 15 19,06 2,60 0,425 18,01 OK
28 22,4 19,1 18,6 15 19,01 2,26 0,425 17,96 OK
29 19,2 19,1 18,8 15 19,01 1,96 0,425 17,97 OK
30 20,9 18,4 18,0 15 18,91 1,82 0,425 17,86 OK

NB-CPD/SG10/12/091 Page 55 of 74
Example 6

Example of statistical analysis of compressive strength using a special type of Rolling inspection
"Progressive Sampling.
The fractile p = 50%
The confidence level = 95%
The number of series of inspection lots is l = 15.
One-sided tolerance interval, lower limit
The declared mean compressive strength is 6 N/mm

For each of the 1st to 5th inspection lots a spot size of one sample is taken and tested. These
inspection lots are evaluated together (xm (equation 1), standard deviation ss (equation 2) and xest
(equation 3) according to clause 5.6.6 using k1,u taken from Annex A Table A8 (p: 50 % and : 95
%)). For the 6th and following inspection lots 1 additional sample is taken and tested and evaluated
together with the ones from the previous inspection lots (xm (equation 1), standard deviation ss
(equation 2) and xest (equation 3) according to clause 5.6.6 using k1,u taken from Annex A Table A8
(p: 50 % and : 95 %)). The spot size is gradually increased from 5 to 15 samples.

From then on, 1 additional sample is taken from each next inspection lot and evaluated together
with the ones from the previous inspection lots (xm (equation 1), standard deviation ss (equation 2)
and xest (equation 3) according to clause 5.6.6 using k1,u taken from Annex A Table A8 (p: 50 % and
: 95 %)) but the spot sample size is limited to the last 15 samples. The spot sample size continues
to be 15.

For the first 30 samples, the standard deviation of the population is considered to be unknown and
the k1,u factor taken from Annex A Table A8 (p: 50 % and : 95 %) is 0,455.

For the inspection lots 30 60 the standard deviation can be considered as known, but the used
acceptance coefficient is corrected (kc). The acceptance coefficient for the known standard
deviation k1,k is taken from Annex A Table A4 (p: 50 % and : 95 %) and is 0,425. The corrected
acceptance coefficient kc is calculated by a linear interpolation between the acceptance coefficient
k1,u and k1,k taking into account the considered inspection lot. The known standard deviation is
calculated based on the first 21 test results.

From inspection lot 61 and so on 1 additional sample is taken from each next inspection lot and the
test results are evaluated together with the ones from the previous inspection lots (xm (equation 1),
known standard deviation and xest (equation 6) according to clause 5.6.6 using k1,k taken from
Annex A Table A4 (p: 50 % and : 95 %)) and the spot sample size is still limited to the last 15
samples.

After each evaluation the result has to be compared with the lower limit value (e.g. the declared
value) decided by the manufacturer.

Part of the evaluation is also to check that the standard deviation ss of the spot sample corresponds
to the following equation:

0,63 ss 1,37

In the last column it is indicated whether the mentioned equation fits or does not fit.

If there is a non-conformity due to great differences between the test results, the estimated value is
highlighted by a red signal at the right side (batch 2). A non-conforming inspection lot has to be
treated separately as described in the text.

NB-CPD/SG10/12/091 Page 56 of 74
EXAMPLE6
ONE SIDED TOLERANCE INTERVAL-lower limit METHODB:progressivesampling:useonly1testresultperinspectionlot

fractilep 50 Startcorrection 30 Seriesofinspectionlots 15 DeclaredValue 6


confidencelevel 95 Endcorrection 60 1,009 0,935

Equation
Inspectionlot test1 test2 test3 n Xm Ss k1,u kc k1,k Xest Xest Xest
OK?
1 6,78 1 6,78 #DEEL/0!
2 8,36 2 7,57 1,11 4,465 2,59
3 8,64 3 7,92 1,00 1,686 6,24
4 8,41 4 8,05 0,85 1,177 7,04
5 8,25 5 8,09 0,74 0,954 7,38
6 7,44 6 7,98 0,72 0,823 7,39
7 8,57 7 8,06 0,69 0,735 7,56
8 6,25 8 7,84 0,91 0,670 7,23
9 7,83 9 7,84 0,85 0,620 7,31
10 7,40 10 7,79 0,81 0,580 7,32
11 8,57 11 7,86 0,80 0,547 7,42
12 8,27 12 7,90 0,77 0,519 7,49
13 7,67 13 7,88 0,74 0,495 7,51
14 8,13 14 7,90 0,72 0,474 7,56
15 6,53 15 7,81 0,78 0,455 7,45
16 7,85 15 7,88 0,72 0,455 7,55
17 7,20 15 7,80 0,73 0,455 7,47
18 7,95 15 7,75 0,69 0,455 7,44
19 7,39 15 7,69 0,67 0,455 7,38
20 5,96 15 7,53 0,79 0,455 7,17
21 6,26 15 7,46 0,85 0,455 7,07
22 5,60 15 7,26 0,92 0,455 6,84
23 7,15 15 7,32 0,88 0,455 6,92
24 6,01 15 7,20 0,93 0,455 6,78
25 7,88 15 7,23 0,94 0,455 6,80
26 6,01 15 7,06 0,91 0,455 6,64
27 5,91 15 6,90 0,89 0,455 6,50
28 7,05 15 6,86 0,87 0,455 6,47
29 5,51 15 6,69 0,86 0,455 6,30
30 5,87 15 6,64 0,88 0,455 0,455 0,425 1,009 6,24 6,18
31 7,59 15 6,62 0,86 0,454 6,17
32 8,64 15 6,72 1,00 0,453 6,26
33 6,72 15 6,64 0,94 0,452 6,18
34 8,11 15 6,69 1,00 0,451 6,23
35 8,21 15 6,84 1,05 0,450 6,38
36 8,48 15 6,98 1,11 0,449 6,53
37 7,87 15 7,14 1,07 0,448 6,68
38 7,36 15 7,15 1,07 0,447 6,70
39 8,11 15 7,29 1,05 0,446 6,84
40 6,50 15 7,20 1,05 0,445 6,75
41 7,38 15 7,29 1,00 0,444 6,84
42 7,78 15 7,41 0,93 0,443 6,97
43 8,11 15 7,48 0,94 0,442 7,04
44 8,70 15 7,70 0,81 0,441 7,25
45 7,46 15 7,80 0,65 0,440 7,36
46 7,76 15 7,81 0,64 0,439 7,37
47 6,46 15 7,67 0,69 0,438 7,23
48 7,78 15 7,74 0,64 0,437 7,30
49 6,73 15 7,65 0,68 0,436 7,21
50 7,47 15 7,60 0,66 0,435 7,16
51 6,98 15 7,50 0,63 0,434 7,06
52 6,35 15 7,40 0,68 0,433 6,96
53 6,35 15 7,33 0,74 0,432 6,89
54 5,53 15 7,16 0,83 0,431 6,72
55 6,68 15 7,17 0,83 0,430 6,73
56 5,86 15 7,07 0,89 0,429 6,63
57 7,40 15 7,04 0,87 0,428 6,61
58 6,42 15 6,93 0,83 0,427 6,50
59 5,80 15 6,74 0,72 0,426 6,31
60 6,57 15 6,68 0,69 0,425 0,425 0,935 6,28 6,28 OK
61 5,71 15 6,54 0,67 0,425 6,14 OK
62 5,78 15 6,49 0,69 0,425 6,10 OK
63 10,46 15 6,67 1,21 0,425 6,28 OK
64 12,26 15 7,04 1,88 0,425 6,64 OK
65 8,90 15 7,14 1,94 0,425 6,74 OK

NB-CPD/SG10/12/091 Page 57 of 74
Example 7

Example of statistical analysis of compressive strength using Rolling inspection.

This example is similar to example 4. The only difference is that the declared compressive strength
is a 5 % characteristic value.
The fractile p = 95%
The confidence level = 95%
The number of series of inspection lots is l = 4.
One-sided tolerance interval, lower limit
The declared 5 % characteristic compressive strength is 10 N/mm

For the first inspection lot a sample size of 3 samples are taken and tested and evaluated (xm
(equation 1), standard deviation ss (equation 2) and xest (equation 3) according to clause 5.6.6 using
k1,u taken from Annex A Table A8 (p: 95 % and : 95 %)). For the next and the following 3
inspection lots 3 additional samples are taken and tested and evaluated together with the ones from
the previous inspection lots (xm (equation 1), standard deviation ss (equation 2) and xest (equation 3)
according to clause 5.6.6 using k1,u taken from Annex A Table A8 (p: 95 % and : 95 %)). By doing
so the spot sample size evaluated together is gradually increased from 3 to 12 samples.

From then on, 3 additional samples are taken from each next inspection lot and evaluated together
with the ones from the previous inspection lots (xm (equation 1), standard deviation ss (equation 2)
and xest (equation 3) according to clause 5.6.6 using k1,u taken from Annex A Table A8 (p: 95 % and
: 95 %)) but the spot sample size is limited to the last 12 samples. The spot sample size continues
to be 12.

For the first 21 samples (1 7 inspection lot), the standard deviation of the population is considered
to be unknown and the k1,u factor taken from Annex A Table A8 (p: 95 % and : 95 %) is 2,737. For
the inspection lots 8 20 the standard deviation can be considered as known, but the used
acceptance coefficient is corrected (kc). The acceptance coefficient for the known standard
deviation k1,k is taken from Annex A Table A4 (p: 95 % and : 95 %) and is 2,120. The corrected
acceptance coefficient kc is calculated by a linear interpolation between the acceptance coefficient
k1,u and k1,k taking into account the considered inspection lot. The known standard deviation is
calculated based on the first 21 test results.

From inspection lot 21 and so on 3 additional samples are taken from each next inspection lot and
the test results are evaluated together with the ones from the previous inspection lots (xm (equation
1), known standard deviation and xest (equation 6) according to clause 5.6.6 using k1,k taken from
Annex A Table A4 (p: 95 % and : 95 %)) and the spot sample size is still limited to the last 12
samples.

After each evaluation the result has to be compared with the lower limit value (e.g. the declared
value) decided by the manufacturer.

Part of the evaluation is also to check that the standard deviation ss of the spot sample corresponds
to the following equation:

0,63 ss 1,37

In the last column it is indicated whether the mentioned equation fits or does not fit.

NB-CPD/SG10/12/091 Page 58 of 74
If there is a non-conformity due to great differences between the test results, the estimated value is
highlighted by a red signal at the right side (batch 1 and 2). A non-conforming inspection lot has to
be treated separately as described in the text.

EXAMPLE7
ONE SIDED TOLERANCE INTERVAL-lower limit METHODB:useatleast3testresultsperinspectionlot

fractilep 95 Startcorrection 7 Seriesofinspectionlots 4 DeclaredValue 10


confidencelevel 95 Endcorrection 20 2,490 2,461

Equation
Inspectionlot test1 test2 test3 n Xm Ss k1,u kc k1,k Xest Xest Xest
OK?
1 20,2 17,8 16,3 3 18,10 1,97 7,656 3,04
2 21,0 14,7 19,0 6 18,17 2,39 3,708 9,31
3 20,2 17,3 22,3 9 18,76 2,43 3,032 11,38
4 20,0 22,8 21,2 12 19,40 2,45 2,737 12,68
5 23,8 20,6 20,8 12 20,31 2,45 2,737 13,59
6 21,1 21,0 20,1 12 20,93 1,63 2,737 16,48
7 15,4 16,5 17,2 12 20,04 2,49 2,737 2,737 2,120 2,490 13,23 13,23
8 18,8 20,5 20,3 12 19,68 2,32 2,690 12,98
9 19,7 19,5 20,0 12 19,18 1,84 2,642 12,60
10 20,4 19,6 20,1 12 19,00 1,70 2,595 12,54
11 20,5 21,4 20,4 12 20,10 0,65 2,547 13,76
12 19,8 20,5 18,1 12 20,00 0,79 2,500 13,78
13 19,7 17,6 16,9 12 19,58 1,35 2,452 13,48
14 18,7 20,0 19,7 12 19,44 1,34 2,405 13,45
15 14,7 15,8 16,2 12 18,14 1,90 2,357 12,27
16 19,3 17,5 14,4 12 17,54 1,98 2,310 11,79
17 20,8 15,3 19,4 12 17,65 2,27 2,262 12,02
18 15,2 16,1 14,5 12 16,60 2,16 2,215 11,08
19 14,1 16,1 15,4 12 16,51 2,23 2,167 11,11
20 14,4 15,5 15,7 12 16,04 2,02 2,120 2,120 2,461 10,83 10,83 OK
21 19,9 16,1 14,3 12 15,61 1,54 2,120 10,39 OK
22 14,9 16,0 15,5 12 15,66 1,51 2,120 10,44 OK
23 16,0 23,4 21,5 12 16,93 2,97 2,120 11,72 OK
24 15,8 21,7 19,5 12 17,88 3,12 2,120 12,67 OK
25 20,1 21,7 17,1 12 18,60 3,02 2,120 13,38 OK
26 18,1 20,3 17,2 12 19,37 2,50 2,120 14,15 OK
27 20,8 14,5 18,2 12 18,75 2,32 2,120 13,53 OK
28 22,4 19,1 18,6 12 19,01 2,21 2,120 13,79 OK
29 19,2 19,1 18,8 12 18,86 1,94 2,120 13,64 OK
30 20,9 18,4 18,0 12 19,00 1,93 2,120 13,78 OK

NB-CPD/SG10/12/091 Page 59 of 74
Example 8

Example of statistical analysis of compressive strength using a special type of Rolling inspection:
Progressive Sampling

This example is similar to example 6. The only difference is that the declared compressive strength
is a 5 % characteristic value.
The fractile p = 95%
The confidence level = 95%
The number of series of inspection lots is l = 15.
One-sided tolerance interval, lower limit
The declared 5 % characteristic compressive strength is 4 N/mm

For each of the 1st to 5th inspection lots a spot size of one sample is taken and tested. These
inspection lots are evaluated together (xm (equation 1), standard deviation ss (equation 2) and xest
(equation 3) according to clause 5.6.6 using k1,u taken from Annex A Table A8 (p: 95 % and : 95
%)). For the 6th and following inspection lots 1 additional sample is taken and tested and evaluated
together with the ones from the previous inspection lots (xm (equation 1), standard deviation ss
(equation 2) and xest (equation 3) according to clause 5.6.6 using k1,u taken from Annex A Table A8
(p: 95 % and : 95 %)). The spot size is gradually increased from 5 to 15 samples.

From then on, 1 additional sample is taken from each next inspection lot and evaluated together
with the ones from the previous inspection lots (xm (equation 1), standard deviation ss (equation 2)
and xest (equation 3) according to clause 5.6.6 using k1,u taken from Annex A Table A8 (p: 95 % and
: 95 %)) but the spot sample size is limited to the last 15 samples. The spot sample size continues
to be 15.

For the first 30 samples, the standard deviation of the population is considered to be unknown and
the k1,u factor taken from Annex A Table A8 (p: 95 % and : 95 %) is 2,567.

For the inspection lots 30 60 the standard deviation can be considered as known, but the used
acceptance coefficient is corrected (kc). The acceptance coefficient for the known standard
deviation k1,k is taken from Annex A Table A4 (p: 95 % and : 95 %) and is 2,070. The corrected
acceptance coefficient kc is calculated by a linear interpolation between the acceptance coefficient
k1,u and k1,k taking into account the considered inspection lot. The known standard deviation is
calculated based on the first 21 test results.

From inspection lot 61 and so on 1 additional sample is taken from each next inspection lot and the
test results are evaluated together with the ones from the previous inspection lots (xm (equation 1),
known standard deviation and xest (equation 6) according to clause 5.6.6 using k1,k taken from
Annex A Table A4 (p: 95 % and : 95 %)) and the spot sample size is still limited to the last 15
samples.

After each evaluation the result has to be compared with the lower limit value (e.g. the declared
value) decided by the manufacturer.

Part of the evaluation is also to check that the standard deviation ss of the spot sample corresponds
to the following equation:

0,63 ss 1,37

NB-CPD/SG10/12/091 Page 60 of 74
In the last column it is indicated whether the mentioned equation fits or does not fit. At batch 68
there is a non-conformity. The manufacturer has to restart or he decides to continue working with
the acceptance coefficient k1,u. This means that the inspection lots have to be treated separately.

If there is a non-conformity due to great differences between the test results, the estimated value is
highlighted by a red signal at the right side (batch 2, 3, 4, 34 and 68). A non-conforming inspection
lot has to be treated separately as described in the text.

EXAMPLE8
ONE SIDED TOLERANCE INTERVAL-lower limit METHODB:progressivesampling:useonly1testresultperinspectionlot

fractilep 95 Startcorrection 30 Seriesofinspectionlots 15 DeclaredValue 4


confidencelevel 95 Endcorrection 60 1,009 0,974

Equation
Inspectionlot test1 test2 test3 n Xm Ss k1,u kc k1,k Xest Xest Xest
OK?
1 6,78 1 6,78 #DEEL/0!
2 8,36 2 7,57 1,11 26,260 21,69
3 8,64 3 7,92 1,00 7,656 0,26
4 8,41 4 8,05 0,85 5,144 3,66
5 8,25 5 8,09 0,74 4,203 4,96
6 7,44 6 7,98 0,72 3,708 5,32
7 8,57 7 8,06 0,69 3,400 5,71
8 6,25 8 7,84 0,91 3,188 4,95
9 7,83 9 7,84 0,85 3,032 5,27
10 7,40 10 7,79 0,81 2,911 5,44
11 8,57 11 7,86 0,80 2,815 5,60
12 8,27 12 7,90 0,77 2,737 5,78
13 7,67 13 7,88 0,74 2,671 5,89
14 8,13 14 7,90 0,72 2,615 6,02
15 6,53 15 7,81 0,78 2,567 5,81
16 7,85 15 7,88 0,72 2,567 6,02
17 7,20 15 7,80 0,73 2,567 5,93
18 7,95 15 7,75 0,69 2,567 5,97
19 7,39 15 7,69 0,67 2,567 5,95
20 5,96 15 7,53 0,79 2,567 5,51
21 6,26 15 7,46 0,85 2,567 5,26
22 5,60 15 7,26 0,92 2,567 4,90
23 7,15 15 7,32 0,88 2,567 5,07
24 6,01 15 7,20 0,93 2,567 4,82
25 7,88 15 7,23 0,94 2,567 4,81
26 6,01 15 7,06 0,91 2,567 4,72
27 5,91 15 6,90 0,89 2,567 4,61
28 7,05 15 6,86 0,87 2,567 4,63
29 5,51 15 6,69 0,86 2,567 4,48
30 5,87 15 6,64 0,88 2,567 2,567 2,070 1,009 4,37 4,05
31 7,59 15 6,62 0,86 2,550 4,05
32 8,64 15 6,72 1,00 2,534 4,16
33 6,72 15 6,64 0,94 2,517 4,10
34 4,91 15 6,47 1,01 2,501 3,95
35 8,21 15 6,62 1,09 2,484 4,12
36 8,48 15 6,77 1,19 2,468 4,28
37 7,87 15 6,92 1,17 2,451 4,45
38 7,36 15 6,94 1,18 2,434 4,48
39 8,11 15 7,08 1,18 2,418 4,64
40 6,50 15 6,98 1,17 2,401 4,56
41 7,38 15 7,07 1,14 2,385 4,67
42 7,78 15 7,20 1,11 2,368 4,81
43 8,11 15 7,27 1,13 2,352 4,90
44 8,70 15 7,48 1,08 2,335 5,13
45 7,46 15 7,59 0,98 2,319 5,25
46 7,76 15 7,60 0,98 2,302 5,28
47 6,46 15 7,45 0,98 2,285 5,15
48 7,78 15 7,52 0,96 2,269 5,24
49 6,73 15 7,65 0,68 2,252 5,37
50 7,47 15 7,60 0,66 2,236 5,34
51 6,98 15 7,50 0,63 2,219 5,26
52 6,35 15 7,40 0,68 2,203 5,17
53 6,35 15 7,33 0,74 2,186 5,12
54 5,53 15 7,16 0,83 2,169 4,97
55 6,68 15 7,17 0,83 2,153 5,00
56 5,86 15 7,07 0,89 2,136 4,91
57 7,40 15 7,04 0,87 2,120 4,90
58 6,42 15 6,93 0,83 2,103 4,81
59 5,80 15 6,74 0,72 2,087 4,63
60 6,57 15 6,68 0,69 2,070 2,070 0,974 4,66 4,66 OK
61 5,71 15 6,54 0,67 2,070 4,52 OK
62 5,78 15 6,49 0,69 2,070 4,48 OK
63 10,46 15 6,67 1,21 2,070 4,66 OK
64 11,26 15 6,97 1,69 2,070 4,96 OK
65 8,90 15 7,07 1,76 2,070 5,05 OK
66 11,50 15 7,37 2,10 2,070 5,35 OK
67 5,89 15 7,34 2,12 2,070 5,32 OK
68 10,50 15 7,62 2,25 2,070 5,60 NOK

NB-CPD/SG10/12/091 Page 61 of 74
Example 9

Example of statistical analysis of compressive strength using a special type of Rolling inspection:
Progressive Sampling

This example is similar to example 8. The only difference is that the confidence level in this
example is 75 %.
The fractile p = 95%
The confidence level = 75%
The number of series of inspection lots is l = 15.
One-sided tolerance interval, lower limit
The declared 5 % characteristic compressive strength is 4 N/mm

For each of the 1st to 5th inspection lots a spot size of one sample is taken and tested. These
inspection lots are evaluated together (xm (equation 1), standard deviation ss (equation 2) and xest
(equation 3) according to clause 5.6.6 using k1,u taken from Annex A Table A6 (p: 95 % and : 75
%)). For the 6th and following inspection lots 1 additional sample is taken and tested and evaluated
together with the ones from the previous inspection lots (xm (equation 1), standard deviation ss
(equation 2) and xest (equation 3) according to clause 5.6.6 using k1,u taken from Annex A Table A6
(p: 95 % and : 75 %)). The spot size is gradually increased from 5 to 15 samples.

From then on, 1 additional sample is taken from each next inspection lot and evaluated together
with the ones from the previous inspection lots (xm (equation 1), standard deviation ss (equation 2)
and xest (equation 3) according to clause 5.6.6 using k1,u taken from Annex A Table A6 (p: 95 % and
: 75 %)) but the spot sample size is limited to the last 15 samples. The spot sample size continues
to be 15.

For the first 30 samples, the standard deviation of the population is considered to be unknown and
the k1,u factor taken from Annex A Table A6 (p: 95 % and : 75 %) is 1,991.

For the inspection lots 30 60 the standard deviation can be considered as known, but the used
acceptance coefficient is corrected (kc). The acceptance coefficient for the known standard
deviation k1,k is taken from Annex A Table A2 (p: 95 % and : 75 %) and is 1,820. The corrected
acceptance coefficient kc is calculated by a linear interpolation between the acceptance coefficient
k1,u and k1,k taking into account the considered inspection lot. The known standard deviation is
calculated based on the first 21 test results.

From inspection lot 61 and so on 1 additional sample is taken from each next inspection lot and the
test results are evaluated together with the ones from the previous inspection lots (xm (equation 1),
known standard deviation and xest (equation 6) according to clause 5.6.6 using k1,k taken from
Annex A Table A2 (p: 95 % and : 75 %)) and the spot sample size is still limited to the last 15
samples.

After each evaluation the result has to be compared with the lower limit value (e.g. the declared
value) decided by the manufacturer.

Part of the evaluation is also to check that the standard deviation ss of the spot sample corresponds
to the following equation:

0,63 ss 1,37

NB-CPD/SG10/12/091 Page 62 of 74
In the last column it is indicated whether the mentioned equation fits or does not fit. At batch 68
there is a non-conformity. The manufacturer has to restart or he decides to continue working with
the acceptance coefficient k1,u. This means that the inspection lots have to be treated separately.

If there is a non-conformity due to great differences between the test results, the estimated value is
highlighted by a red signal at the right side (batch 2 and 68). A non-conforming inspection lot has
to be treated separately as described in the text.

NB-CPD/SG10/12/091 Page 63 of 74
EXAMPLE9
ONE SIDED TOLERANCE INTERVAL-lower limit METHODB:progressivesampling:useonly1testresultperinspectionlot

fractilep 95 Startcorrection 30 Seriesofinspectionlots 15 DeclaredValue 4


confidencelevel 75 Endcorrection 60 1,009 0,974

Equation
Inspectionlot test1 test2 test3 n Xm Ss k1,u kc k1,k Xest Xest Xest
OK?
1 6,78 1 6,78 #DEEL/0!
2 8,36 2 7,57 1,11 5,122 1,86
3 8,64 3 7,92 1,00 3,152 4,77
4 8,41 4 8,05 0,85 2,681 5,76
5 8,25 5 8,09 0,74 2,464 6,25
6 7,44 6 7,98 0,72 2,336 6,31
7 8,57 7 8,06 0,69 2,251 6,51
8 6,25 8 7,84 0,91 2,189 5,86
9 7,83 9 7,84 0,85 2,142 6,02
10 7,40 10 7,79 0,81 2,104 6,09
11 8,57 11 7,86 0,80 2,074 6,20
12 8,27 12 7,90 0,77 2,048 6,31
13 7,67 13 7,88 0,74 2,026 6,37
14 8,13 14 7,90 0,72 2,008 6,45
15 6,53 15 7,81 0,78 1,991 6,26
16 7,85 15 7,88 0,72 1,991 6,44
17 7,20 15 7,80 0,73 1,991 6,35
18 7,95 15 7,75 0,69 1,991 6,37
19 7,39 15 7,69 0,67 1,991 6,34
20 5,96 15 7,53 0,79 1,991 5,96
21 6,26 15 7,46 0,85 1,991 5,76
22 5,60 15 7,26 0,92 1,991 5,43
23 7,15 15 7,32 0,88 1,991 5,57
24 6,01 15 7,20 0,93 1,991 5,35
25 7,88 15 7,23 0,94 1,991 5,35
26 6,01 15 7,06 0,91 1,991 5,24
27 5,91 15 6,90 0,89 1,991 5,13
28 7,05 15 6,86 0,87 1,991 5,13
29 5,51 15 6,69 0,86 1,991 4,98
30 5,87 15 6,64 0,88 1,991 1,991 1,820 1,009 4,88 4,63
31 7,59 15 6,62 0,86 1,985 4,62
32 8,64 15 6,72 1,00 1,980 4,72
33 6,72 15 6,64 0,94 1,974 4,65
34 4,91 15 6,47 1,01 1,968 4,49
35 8,21 15 6,62 1,09 1,963 4,64
36 8,48 15 6,77 1,19 1,957 4,80
37 7,87 15 6,92 1,17 1,951 4,95
38 7,36 15 6,94 1,18 1,945 4,97
39 8,11 15 7,08 1,18 1,940 5,12
40 6,50 15 6,98 1,17 1,934 5,03
41 7,38 15 7,07 1,14 1,928 5,13
42 7,78 15 7,20 1,11 1,923 5,26
43 8,11 15 7,27 1,13 1,917 5,34
44 8,70 15 7,48 1,08 1,911 5,55
45 7,46 15 7,59 0,98 1,906 5,67
46 7,76 15 7,60 0,98 1,900 5,68
47 6,46 15 7,45 0,98 1,894 5,54
48 7,78 15 7,52 0,96 1,888 5,62
49 6,73 15 7,65 0,68 1,883 5,75
50 7,47 15 7,60 0,66 1,877 5,70
51 6,98 15 7,50 0,63 1,871 5,61
52 6,35 15 7,40 0,68 1,866 5,51
53 6,35 15 7,33 0,74 1,860 5,45
54 5,53 15 7,16 0,83 1,854 5,29
55 6,68 15 7,17 0,83 1,849 5,30
56 5,86 15 7,07 0,89 1,843 5,21
57 7,40 15 7,04 0,87 1,837 5,19
58 6,42 15 6,93 0,83 1,831 5,08
59 5,80 15 6,74 0,72 1,826 4,89
60 6,57 15 6,68 0,69 1,820 1,820 0,974 4,90 4,90 OK
61 5,71 15 6,54 0,67 1,820 4,77 OK
62 5,78 15 6,49 0,69 1,820 4,72 OK
63 10,46 15 6,67 1,21 1,820 4,90 OK
64 11,26 15 6,97 1,69 1,820 5,20 OK
65 8,90 15 7,07 1,76 1,820 5,30 OK
66 11,50 15 7,37 2,10 1,820 5,60 OK
67 5,89 15 7,34 2,12 1,820 5,57 OK
68 10,50 15 7,62 2,25 1,820 5,84 NOK

NB-CPD/SG10/12/091 Page 64 of 74
Example10

Example of statistical analysis of gross dry density using Rolling inspection.


The fractile p = 50%
The confidence level = 90%
The number of series of inspection lots is l = 4.
One-sided tolerance interval, upper limit
The declared mean gross dry density is 700 kg/m3

For the first inspection lot a sample size of 3 samples are taken and tested and evaluated (xm
(equation 1), standard deviation ss (equation 2) and xest (equation 4) according to clause 5.6.6 using
k1,u taken from Annex A Table A7 (p: 50 % and : 90 %)). For the next and the following 2
inspection lots 3 additional samples are taken and tested and evaluated together with the ones from
the previous inspection lots (xm (equation 1), standard deviation ss (equation 2) and xest (equation 4)
according to clause 5.6.6 using k1,u taken from Annex A Table A7 (p: 50 % and : 90 %)). By doing
so the spot sample size evaluated together is gradually increased from 3 to 12 samples.

From then on, 3 additional samples are taken from each next inspection lot and evaluated together
with the ones from the previous inspection lots (xm (equation 1), standard deviation ss (equation 2)
and xest (equation 4) according to clause 5.6.6 using k1,u taken from Annex A Table A7 (p: 50 % and
: 90 %)) but the spot sample size is limited to the last 12 samples. The spot sample size continues
to be 12.

For the first 21 samples (1 7 inspection lot), the standard deviation of the population is considered
to be unknown and the k1,u factor taken from Annex A Table A7 (p: 50 % and : 90 %) is 0,394. For
the inspection lots 8 20 the standard deviation can be considered as known, but the used
acceptance coefficient is corrected (kc). The acceptance coefficient for the known standard
deviation k1,k is taken from Annex A Table A3 (p: 50 % and : 90 %) and is 0,370. The corrected
acceptance coefficient kc is calculated by a linear interpolation between the acceptance coefficient
k1,u and k1,k taking into account the considered inspection lot. The known standard deviation is
calculated based on the first 21 test results.

From inspection lot 21 and so on 3 additional samples are taken from each next inspection lot and
the test results are evaluated together with the ones from the previous inspection lots (xm (equation
1), known standard deviation and xest (equation 7) according to clause 5.6.6 using k1,k taken from
Annex A Table A3 (p: 50 % and : 90 %)) and the spot sample size is still limited to the last 12
samples.

After each evaluation the result has to be compared with the lower limit value (e.g. the declared
value) decided by the manufacturer.

Part of the evaluation is also to check that the standard deviation ss of the spot sample corresponds
to the following equation:

0,63 ss 1,37

In the last column it is indicated whether the mentioned equation fits or does not fit.

If there is a non-conformity due to great differences between the test results, the estimated value is
highlighted by a red signal at the right side. A non-conforming inspection lot has to be treated
separately as described in the text.

NB-CPD/SG10/12/091 Page 65 of 74
EXAMPLE10
ONE SIDED TOLERANCE INTERVAL-upper limit METHODB:useatleast3testresultsperinspectionlot

fractilep 50 Startcorrection 7 Seriesofinspectionlots 4 DeclaredValue 700


confidencelevel 90 Endcorrection 20 20,92845 20,217

Equation
Inspectionlot test1 test2 test3 n Xm Ss k1,u kc k1,k Xest Xest Xest
OK?
1 603,00 632,0 681,0 3 638,67 39,43 1,089 681,60
2 663,00 672,0 657,0 6 651,33 28,93 0,603 668,78
3 672,00 680,0 693,0 9 661,44 27,95 0,466 674,47
4 659,00 662,0 638,0 12 659,33 24,78 0,394 669,10
5 672,00 669,0 682,0 12 668,25 14,07 0,394 673,79
6 626,00 655,0 662,0 12 664,17 18,61 0,394 671,50
7 653,00 656,0 652,0 12 657,17 14,82 0,394 0,394 0,370 20,928 663,01 665,41
8 668,00 658,0 662,0 12 659,58 13,79 0,392 667,79
9 632,00 692,0 636,0 12 654,33 17,54 0,390 662,50
10 618,00 635,0 668,0 12 652,50 19,92 0,388 660,63
11 637,00 662,0 669,0 12 653,08 21,21 0,387 661,17
12 676,00 696,0 685,0 12 658,83 26,31 0,385 666,89
13 668,00 671,0 639,0 12 660,33 23,10 0,383 668,35
14 628,00 638,0 652,0 12 660,08 21,30 0,381 668,06
15 663,00 672,0 672,0 12 663,33 20,31 0,379 671,27
16 628,00 680,0 657,0 12 655,67 18,35 0,377 663,56
17 672,00 662,0 666,0 12 657,50 17,61 0,376 665,36
18 668,00 669,0 692,0 12 666,75 15,18 0,374 674,57
19 683,00 655,0 691,0 12 668,58 17,62 0,372 676,37
20 641,00 656,0 682,0 12 669,75 15,34 0,370 0,370 20,217 677,23 677,23 OK
21 698,00 658,0 682,0 12 672,92 17,77 0,370 680,40 OK
22 665,00 692,0 688,0 12 674,25 18,35 0,370 681,73 OK
23 671,00 635,0 618,0 12 665,50 24,81 0,370 672,98 OK
24 662,00 662,0 619,0 12 662,50 26,78 0,370 669,98 OK
25 638,00 696,0 668,0 12 659,50 26,75 0,370 666,98 OK
26 673,00 671,0 667,0 12 656,67 23,83 0,370 664,15 OK
27 655,00 682,0 662,0 12 662,92 19,73 0,370 670,40 OK
28 672,00 662,0 672,0 12 668,17 14,10 0,370 675,65 OK
29 672,00 652,0 672,0 12 667,67 8,48 0,370 675,15 OK
30 662,00 662,0 669,0 12 666,17 8,47 0,370 673,65 OK
31 628,00 516,0 676,0 12 651,25 44,55 0,370 658,73 OK
32 667,00 668,0 698,0 12 653,50 46,23 0,370 660,98 OK
33 619,00 669,0 626,0 12 646,67 47,14 0,370 654,15 OK
34 652,00 685,0 652,0 12 646,33 47,68 0,370 653,81 OK

NB-CPD/SG10/12/091 Page 66 of 74
Example 11

Example of statistical analysis of net dry density using Rolling inspection.


The fractile p = 50%
The confidence level = 90%
The number of series of inspection lots is l = 4.
One-sided tolerance interval, upper limit
The declared mean net dry density is 1400 kg/m3

For the first inspection lot a sample size of 3 samples are taken and tested and evaluated (xm
(equation 1), standard deviation ss (equation 2) and xest (equation 4) according to clause 5.6.6 using
k1,u taken from Annex A Table A7 (p: 50 % and : 90 %)). For the next and the following 2
inspection lots 3 additional samples are taken and tested and evaluated together with the ones from
the previous inspection lots (xm (equation 1), standard deviation ss (equation 2) and xest (equation 4)
according to clause 5.6.6 using k1,u taken from Annex A Table A7 (p: 50 % and : 90 %)). By doing
so the spot sample size evaluated together is gradually increased from 3 to 12 samples.

From then on, 3 additional samples are taken from each next inspection lot and evaluated together
with the ones from the previous inspection lots (xm (equation 1), standard deviation ss (equation 2)
and xest (equation 4) according to clause 5.6.6 using k1,u taken from Annex A Table A7 (p: 50 % and
: 90 %)) but the spot sample size is limited to the last 12 samples. The spot sample size continues
to be 12.

For the first 21 samples (1 7 inspection lot), the standard deviation of the population is considered
to be unknown and the k1,u factor taken from Annex A Table A7 (p: 50 % and : 90 %) is 0,394. For
the inspection lots 8 20 the standard deviation can be considered as known, but the used
acceptance coefficient is corrected (kc). The acceptance coefficient for the known standard
deviation k1,k is taken from Annex A Table A3 (p: 50 % and : 90 %) and is 0,370. The corrected
acceptance coefficient kc is calculated by a linear interpolation between the acceptance coefficient
k1,u and k1,k taking into account the considered inspection lot. The known standard deviation is
calculated based on the first 21 test results.

From inspection lot 21 and so on 3 additional samples are taken from each next inspection lot and
the test results are evaluated together with the ones from the previous inspection lots (xm (equation
1), known standard deviation and xest (equation 7) according to clause 5.6.6 using k1,k taken from
Annex A Table A3 (p: 50 % and : 90 %)) and the spot sample size is still limited to the last 12
samples.

After each evaluation the result has to be compared with the lower limit value (e.g. the declared
value) decided by the manufacturer.

Part of the evaluation is also to check that the standard deviation ss of the spot sample corresponds
to the following equation:

0,63 ss 1,37

In the last column it is indicated whether the mentioned equation fits or does not fit.

If there is a non-conformity due to great differences between the test results, the estimated value is
highlighted by a red signal at the right side. A non-conforming inspection lot has to be treated
separately as described in the text.

NB-CPD/SG10/12/091 Page 67 of 74
The last column makes the link to EN 1745 Annex A to calculate the 10,dry (50/90) value of the
material.

EXAMPLE11
ONE SIDED TOLERANCE INTERVAL-upper limit METHODB:useatleast3testresultsperinspectionlot

fractilep 50 Startcorrection 7 Seriesofinspectionlots 4 DeclaredValue 1400


confidencelevel 90 Endcorrection 20 41,8569 40,433
10,dry
Equation (50/90)
Inspectionlot test1 test2 test3 n Xm Ss k1,u kc k1,k Xest Xest Xest
OK? material
concrete
1 1206,00 1264,0 1362,0 3 1277,33 78,85 1,089 1363,20 0,383
2 1326,00 1344,0 1314,0 6 1302,67 57,86 0,603 1337,56 0,357
3 1344,00 1360,0 1386,0 9 1322,89 55,90 0,466 1348,94 0,368
4 1318,00 1324,0 1276,0 12 1318,67 49,55 0,394 1338,19 0,357
5 1344,00 1338,0 1364,0 12 1336,50 28,13 0,394 1347,58 0,367
6 1252,00 1310,0 1324,0 12 1328,33 37,21 0,394 1342,99 0,362
7 1306,00 1312,0 1304,0 12 1314,33 29,64 0,394 0,394 0,370 41,857 1326,01 1330,82 0,345
8 1336,00 1316,0 1324,0 12 1319,17 27,59 0,392 1335,58 0,355
9 1264,00 1384,0 1272,0 12 1308,67 35,08 0,390 1325,00 0,344
10 1236,00 1270,0 1336,0 12 1305,00 39,83 0,388 1321,26 0,340
11 1274,00 1324,0 1338,0 12 1306,17 42,42 0,387 1322,35 0,341
12 1352,00 1392,0 1370,0 12 1317,67 52,61 0,385 1333,77 0,353
13 1336,00 1342,0 1278,0 12 1320,67 46,20 0,383 1336,69 0,356
14 1256,00 1276,0 1304,0 12 1320,17 42,59 0,381 1336,12 0,355
15 1326,00 1344,0 1344,0 12 1326,67 40,63 0,379 1342,54 0,362
16 1256,00 1360,0 1314,0 12 1311,33 36,69 0,377 1327,13 0,346
17 1344,00 1324,0 1332,0 12 1315,00 35,22 0,376 1330,72 0,350
18 1336,00 1338,0 1384,0 12 1333,50 30,37 0,374 1349,14 0,369
19 1366,00 1310,0 1382,0 12 1337,17 35,25 0,372 1352,73 0,372
20 1282,00 1312,0 1364,0 12 1339,50 30,68 0,370 0,370 40,433 1354,46 1354,46 OK 0,374
21 1396,00 1316,0 1364,0 12 1345,83 35,55 0,370 1360,79 OK 0,381
22 1330,00 1384,0 1376,0 12 1348,50 36,70 0,370 1363,46 OK 0,384
23 1342,00 1270,0 1236,0 12 1331,00 49,63 0,370 1345,96 OK 0,365
24 1324,00 1324,0 1238,0 12 1325,00 53,56 0,370 1339,96 OK 0,359
25 1276,00 1392,0 1336,0 12 1319,00 53,51 0,370 1333,96 OK 0,353
26 1346,00 1342,0 1334,0 12 1313,33 47,67 0,370 1328,29 OK 0,347
27 1310,00 1364,0 1324,0 12 1325,83 39,46 0,370 1340,79 OK 0,360
28 1344,00 1324,0 1344,0 12 1336,33 28,20 0,370 1351,29 OK 0,371
29 1344,00 1304,0 1344,0 12 1335,33 16,96 0,370 1350,29 OK 0,370
30 1324,00 1324,0 1338,0 12 1332,33 16,95 0,370 1347,29 OK 0,367
31 1256,00 1032,0 1352,0 12 1302,50 89,10 0,370 1317,46 OK 0,336
32 1334,00 1336,0 1396,0 12 1307,00 92,46 0,370 1321,96 OK 0,340
33 1238,00 1338,0 1252,0 12 1293,33 94,29 0,370 1308,29 OK 0,326

NB-CPD/SG10/12/091 Page 68 of 74
Example 12

Example of statistical analysis of net dry density using Rolling inspection.

This example is similar to example 11. The only difference is that the confidence level in this
example is 50 % and the numbers of series of inspection lots is 5.
The fractile p = 50%
The confidence level = 50%
The number of series of inspection lots is l = 5.
One-sided tolerance interval, upper limit
The declared mean net dry density is 1400 kg/m3

For the first inspection lot a sample size of 3 samples are taken and tested and evaluated (xm
(equation 1), standard deviation ss (equation 2) and xest (equation 4) according to clause 5.6.6 using
k1,u taken from Annex A Table A5 (p: 50 % and : 50 %)). For the next and the following 2
inspection lots 3 additional samples are taken and tested and evaluated together with the ones from
the previous inspection lots (xm (equation 1), standard deviation ss (equation 2) and xest (equation 4)
according to clause 5.6.6 using k1,u taken from Annex A Table A5 (p: 50 % and : 50 %)). By doing
so the spot sample size evaluated together is gradually increased from 3 to 15 samples.

From then on, 3 additional samples are taken from each next inspection lot and evaluated together
with the ones from the previous inspection lots (xm (equation 1), standard deviation ss (equation 2)
and xest (equation 4) according to clause 5.6.6 using k1,u taken from Annex A Table A5 (p: 50 % and
: 50 %)) but the spot sample size is limited to the last 15 samples. The spot sample size continues
to be 15.

For the first 21 samples (1 7 inspection lot), the standard deviation of the population is considered
to be unknown and the k1,u factor taken from Annex A Table A5 (p: 50 % and : 50 %) is 0,000,
which means that xest = xm. For the inspection lots 8 20 the standard deviation can be considered
as known. The acceptance coefficient for the known standard deviation k1,k is taken from Annex A
Table A1 (p: 50 % and : 50 %) and is 0,000, which means that xest = xm. The known standard
deviation is calculated based on the first 21 test results.

From inspection lot 21 and so on 3 additional samples are taken from each next inspection lot and
the test results are evaluated together with the ones from the previous inspection lots (xm (equation
1), known standard deviation and xest (equation 7) according to clause 5.6.6 using k1,k taken from
Annex A Table A1 (p: 50 % and : 50 %)) and the spot sample size is still limited to the last 15
samples.

After each evaluation the result has to be compared with the lower limit value (e.g. the declared
value) decided by the manufacturer.

Part of the evaluation is also to check that the standard deviation ss of the spot sample corresponds
to the following equation:

0,63 ss 1,37

In the last column it is indicated whether the mentioned equation fits or does not fit.

If there is a non-conformity due to great differences between the test results, the estimated value is
highlighted by a red signal at the right side. A non-conforming inspection lot has to be treated
separately as described in the text.

NB-CPD/SG10/12/091 Page 69 of 74
The last column makes the link to EN 1745 Annex A to calculate the 10,dry (50/50) value of the
material.

EXAMPLE12
ONE SIDED TOLERANCE INTERVAL-upper limit METHODB:useatleast3testresultsperinspectionlot

fractilep 50 Startcorrection 7 Seriesofinspectionlots 5 DeclaredValue 1400


confidencelevel 50 Endcorrection 20 41,8569 40,433
10,dry
Equation (50/50)
Inspectionlot test1 test2 test3 n Xm Ss k1,u kc k1,k Xest Xest Xest
OK? material
clay
1 1206,00 1264,0 1362,0 3 1277,33 78,85 0,000 1277,33 0,291
2 1326,00 1344,0 1314,0 6 1302,67 57,86 0,000 1302,67 0,301
3 1344,00 1360,0 1386,0 9 1322,89 55,90 0,000 1322,89 0,309
4 1318,00 1324,0 1276,0 12 1318,67 49,55 0,000 1318,67 0,308
5 1344,00 1338,0 1364,0 15 1324,67 45,94 0,000 1324,67 0,310
6 1252,00 1310,0 1324,0 15 1328,27 33,47 0,000 1328,27 0,312
7 1306,00 1312,0 1304,0 15 1324,13 34,15 0,000 0,000 0,000 41,857 1324,13 1324,13 0,310
8 1336,00 1316,0 1324,0 15 1316,53 26,93 0,000 1316,53 0,307
9 1264,00 1384,0 1272,0 15 1316,67 35,61 0,000 1316,67 0,307
10 1236,00 1270,0 1336,0 15 1303,07 38,35 0,000 1303,07 0,301
11 1274,00 1324,0 1338,0 15 1306,40 37,64 0,000 1306,40 0,303
12 1352,00 1392,0 1370,0 15 1319,20 46,90 0,000 1319,20 0,308
13 1336,00 1342,0 1278,0 15 1317,87 48,51 0,000 1317,87 0,307
14 1256,00 1276,0 1304,0 15 1312,27 45,41 0,000 1312,27 0,305
15 1326,00 1344,0 1344,0 15 1323,73 38,67 0,000 1323,73 0,310
16 1256,00 1360,0 1314,0 15 1323,33 41,62 0,000 1323,33 0,310
17 1344,00 1324,0 1332,0 15 1315,73 33,99 0,000 1315,73 0,306
18 1336,00 1338,0 1384,0 15 1322,53 36,37 0,000 1322,53 0,309
19 1366,00 1310,0 1382,0 15 1337,33 31,49 0,000 1337,33 0,315
20 1282,00 1312,0 1364,0 15 1333,60 35,73 0,000 0,000 40,433 1333,60 1333,60 OK 0,314
21 1396,00 1316,0 1364,0 15 1343,33 32,16 0,000 1343,33 OK 0,318
22 1330,00 1384,0 1376,0 15 1349,33 34,16 0,000 1349,33 OK 0,320
23 1342,00 1270,0 1236,0 15 1335,33 47,11 0,000 1335,33 OK 0,314
24 1324,00 1324,0 1238,0 15 1323,87 50,05 0,000 1323,87 OK 0,310
25 1276,00 1392,0 1336,0 15 1326,93 52,45 0,000 1326,93 OK 0,311
26 1346,00 1342,0 1334,0 15 1323,33 48,32 0,000 1323,33 OK 0,310
27 1310,00 1364,0 1324,0 15 1317,20 44,29 0,000 1317,20 OK 0,307
28 1344,00 1324,0 1344,0 15 1328,13 35,57 0,000 1328,13 OK 0,312
29 1344,00 1304,0 1344,0 15 1335,20 26,58 0,000 1335,20 OK 0,314
30 1324,00 1324,0 1338,0 15 1334,00 15,58 0,000 1334,00 OK 0,314
31 1256,00 1032,0 1352,0 15 1308,53 80,66 0,000 1308,53 OK 0,304
32 1334,00 1336,0 1396,0 15 1313,07 83,02 0,000 1313,07 OK 0,305
33 1238,00 1338,0 1252,0 15 1300,80 85,44 0,000 1300,80 OK 0,300
34 1304,00 1370,0 1304,0 15 1299,87 85,88 0,000 1299,87 OK 0,300

NB-CPD/SG10/12/091 Page 70 of 74
Example 13

Example of statistical analysis of net dry density using Rolling inspection.

This example is similar to example 11. The only difference is that the declared value is a 90 %
characteristic value and the number of series of inspection lots is l = 5.
The fractile p = 90%
The confidence level = 90%
The number of series of inspection lots is l = 5.
One-sided tolerance interval, upper limit
The declared 90 % characteristic net dry density is 1400 kg/m3

For the first inspection lot a sample size of 3 samples are taken and tested and evaluated (xm
(equation 1), standard deviation ss (equation 2) and xest (equation 4) according to clause 5.6.6 using
k1,u taken from Annex A Table A7 (p: 90 % and : 90 %)). For the next and the following 3
inspection lots 3 additional samples are taken and tested and evaluated together with the ones from
the previous inspection lots (xm (equation 1), standard deviation ss (equation 2) and xest (equation 4)
according to clause 5.6.6 using k1,u taken from Annex A Table A7 (p: 90 % and : 90 %)). By doing
so the spot sample size evaluated together is gradually increased from 3 to 15 samples.

From then on, 3 additional samples are taken from each next inspection lot and evaluated together
with the ones from the previous inspection lots (xm (equation 1), standard deviation ss (equation 2)
and xest (equation 4) according to clause 5.6.6 using k1,u taken from Annex A Table A7 (p: 90 % and
: 90 %)) but the spot sample size is limited to the last 15 samples. The spot sample size continues
to be 15.

For the first 21 samples (1 7 inspection lot), the standard deviation of the population is considered
to be unknown and the k1,u factor taken from Annex A Table A7 (p: 90 % and : 90 %) is 1,867. For
the inspection lots 8 20 the standard deviation can be considered as known, but the used
acceptance coefficient is corrected (kc). The acceptance coefficient for the known standard
deviation k1,k is taken from Annex A Table A3 (p: 90 % and : 90 %) and is 1,613. The corrected
acceptance coefficient kc is calculated by a linear interpolation between the acceptance coefficient
k1,u and k1,k taking into account the considered inspection lot. The known standard deviation is
calculated based on the first 21 test results.

From inspection lot 21 and so on 3 additional samples are taken from each next inspection lot and
the test results are evaluated together with the ones from the previous inspection lots (xm (equation
1), known standard deviation and xest (equation 7) according to clause 5.6.6 using k1,k taken from
Annex A Table A3 (p: 90 % and : 90 %)) and the spot sample size is still limited to the last 15
samples.

After each evaluation the result has to be compared with the lower limit value (e.g. the declared
value) decided by the manufacturer.

Part of the evaluation is also to check that the standard deviation ss of the spot sample corresponds
to the following equation:

0,63 ss 1,37

In the last column it is indicated whether the mentioned equation fits or does not fit.

NB-CPD/SG10/12/091 Page 71 of 74
If there is a non-conformity due to great differences between the test results, the estimated value is
highlighted by a red signal at the right side (batch 1, 2, 3, 4, 5, 7, 19, 21, 22, 23 and 29). A non-
conforming inspection lot has to be treated separately as described in the text.

The last column makes the link to EN 1745 Annex A to calculate the 10,dry (90/90) value of the
material.

EXAMPLE13
ONE SIDED TOLERANCE INTERVAL-upper limit METHODB:useatleast3testresultsperinspectionlot

fractilep 90 Startcorrection 7 Seriesofinspectionlots 5 DeclaredValue 1400


confidencelevel 90 Endcorrection 20 41,8569 40,433

10,dry
Equation
Inspectionlot test1 test2 test3 n Xm Ss k1,u kc k1,k Xest Xest Xest
OK?
(90/90)
material

1 1206,00 1264,0 1362,0 3 1277,33 78,85 4,259 1613,16 0,427


2 1326,00 1344,0 1314,0 6 1302,67 57,86 2,494 1446,98 0,360
3 1344,00 1360,0 1386,0 9 1322,89 55,90 2,133 1442,13 0,358
4 1318,00 1324,0 1276,0 12 1318,67 49,55 1,967 1416,14 0,347
5 1344,00 1338,0 1364,0 15 1324,67 45,94 1,867 1410,43 0,345
6 1252,00 1310,0 1324,0 15 1328,27 33,47 1,867 1390,76 0,337
7 1306,00 1312,0 1304,0 15 1324,13 34,15 1,867 1,867 1,613 41,857 1387,88 1402,28 0,336
8 1336,00 1316,0 1324,0 15 1316,53 26,93 1,847 1393,86 0,338
9 1264,00 1384,0 1272,0 15 1316,67 35,61 1,828 1393,18 0,338
10 1236,00 1270,0 1336,0 15 1303,07 38,35 1,808 1378,76 0,332
11 1274,00 1324,0 1338,0 15 1306,40 37,64 1,789 1381,28 0,333
12 1352,00 1392,0 1370,0 15 1319,20 46,90 1,769 1393,26 0,338
13 1336,00 1342,0 1278,0 15 1317,87 48,51 1,750 1391,11 0,337
14 1256,00 1276,0 1304,0 15 1312,27 45,41 1,730 1384,69 0,335
15 1326,00 1344,0 1344,0 15 1323,73 38,67 1,711 1395,34 0,339
16 1256,00 1360,0 1314,0 15 1323,33 41,62 1,691 1394,12 0,338
17 1344,00 1324,0 1332,0 15 1315,73 33,99 1,672 1385,70 0,335
18 1336,00 1338,0 1384,0 15 1322,53 36,37 1,652 1391,68 0,337
19 1366,00 1310,0 1382,0 15 1337,33 31,49 1,633 1405,67 0,343
20 1282,00 1312,0 1364,0 15 1333,60 35,73 1,613 1,613 40,433 1398,82 1398,82 OK 0,340
21 1396,00 1316,0 1364,0 15 1343,33 32,16 1,613 1408,55 OK 0,344
22 1330,00 1384,0 1376,0 15 1349,33 34,16 1,613 1414,55 OK 0,347
23 1342,00 1270,0 1236,0 15 1335,33 47,11 1,613 1400,55 OK 0,341
24 1324,00 1324,0 1238,0 15 1323,87 50,05 1,613 1389,09 OK 0,336
25 1276,00 1392,0 1336,0 15 1326,93 52,45 1,613 1392,15 OK 0,338
26 1346,00 1342,0 1334,0 15 1323,33 48,32 1,613 1388,55 OK 0,336
27 1310,00 1364,0 1324,0 15 1317,20 44,29 1,613 1382,42 OK 0,334
28 1344,00 1324,0 1344,0 15 1328,13 35,57 1,613 1393,35 OK 0,338
29 1344,00 1304,0 1344,0 15 1335,20 26,58 1,613 1400,42 OK 0,341
30 1324,00 1324,0 1338,0 15 1334,00 15,58 1,613 1399,22 OK 0,340
31 1256,00 1032,0 1352,0 15 1308,53 80,66 1,613 1373,75 OK 0,330
32 1334,00 1336,0 1396,0 15 1313,07 83,02 1,613 1378,29 OK 0,332
33 1238,00 1338,0 1252,0 15 1300,80 85,44 1,613 1366,02 OK 0,327

NB-CPD/SG10/12/091 Page 72 of 74
Example 14

Example of TWO-SIDED statistical analysis of dimension using Rolling inspection.


The fractile p = 50%
The confidence level = 75%
The number of series of inspection lots is l = 4.
Two-sided tolerance interval
The manufacturer wants to have the mean value of the length of the green units between 242
mm and 247 mm

For the first inspection lot a sample size of 3 samples are taken and tested and evaluated (xm
(equation 1), standard deviation ss (equation 2) and xest (equation 5) according to clause 5.6.6 using
k2,u taken from Annex A Table A14 (p: 50 % and : 75 %)). For the next and the following 3
inspection lots 3 additional samples are taken and tested and evaluated together with the ones from
the previous inspection lots (xm (equation 1), standard deviation ss (equation 2) and xest (equation 5)
according to clause 5.6.6 using k2,u taken from Annex A Table A14 (p: 50 % and : 75 %)). By doing
so the spot sample size evaluated together is gradually increased from 3 to 12 samples.

From then on, 3 additional samples are taken from each next inspection lot and evaluated together
with the ones from the previous inspection lots (xm (equation 1), standard deviation ss (equation 2)
and xest (equation 5) according to clause 5.6.6 using k2,u taken from Annex A Table A14 (p: 50 %
and : 75 %)) but the spot sample size is limited to the last 12 samples. The spot sample size
continues to be 12.

For the first 21 samples (1 7 inspection lot), the standard deviation of the population is considered
to be unknown and the k2,u factor taken from Annex A Table A14 (p: 50 % and : 75 %) is 0,821.
For the inspection lots 8 20 the standard deviation can be considered as known, but the used
acceptance coefficient is corrected (kc). The acceptance coefficient for the known standard
deviation k2,k is taken from Annex A Table A10 (p: 50 % and : 75 %) and is 0,705. The corrected
acceptance coefficient kc is calculated by a linear interpolation between the acceptance coefficient
k2,u and k2,k taking into account the considered inspection lot. The known standard deviation is
calculated based on the first 21 test results.

From inspection lot 21 and so on 3 additional samples are taken from each next inspection lot and
the test results are evaluated together with the ones from the previous inspection lots (xm (equation
1), known standard deviation and xest (equation 8) according to clause 5.6.6 using k2,k taken from
Annex A Table A10 (p: 50 % and : 75 %)) and the spot sample size is still limited to the last 12
samples.

After each evaluation the result has to be compared with the lower limit value (e.g. the declared
value) decided by the manufacturer.

Part of the evaluation is also to check that the standard deviation ss of the spot sample corresponds
to the following equation:

0,63 ss 1,37

In the last column it is indicated whether the mentioned equation fits or does not fit.

If there is a non-conformity due to great differences between the test results, the estimated value is
highlighted by a red signal at the right side (batch 5, 6, 7, 14, 18, 19, 20, 23, 24 and 25). A non-
conforming inspection lots give a warning to the manufacturer to take some corrective actions.

NB-CPD/SG10/12/091 Page 73 of 74
EXAMPLE14 2,166

TWO SIDED TOLERANCE INTERVAL-lower limit


fractilep 50 Startcorrection 7 Seriesofinspectionlots 4 DeclaredValuelowerlimit 242
confidencelevel 75 Endcorrection 20 DeclaredValueupperlimit 247 1,677

Xest Xest
Xestlower Xestlower Xestlower Xestupper Equation
Inspectionlot test1 test2 test3 n Xm Ss k1,u kc k1,k
limit
upper
limit
upper
limit limit OK?
limit limit

1 244 245 246 3 245,00 1,00 1,492 243,508 246,492


2 245 247 246 6 245,50 1,05 1,009 244,442 246,5582
3 246 247 247 9 245,89 1,05 0,9 244,940 246,8376
4 247 246 246 12 246,00 0,95 0,85 245,190 246,8104
5 242 242 241 12 245,17 2,21 0,85 243,289 247,0441
6 241 242 242 12 244,08 2,57 0,85 241,895 246,2718
7 245 246 247 12 243,92 2,43 0,85 0,850 0,713 2,166 241,852 245,9816 242,076 245,7576
8 247 246 246 12 243,92 2,43 0,839 242,099 245,7347
9 243 244 243 12 244,33 2,10 0,829 242,538 246,1286
10 244 244 243 12 244,83 1,53 0,818 243,061 246,6058
11 246 246 246 12 244,83 1,47 0,808 243,084 246,5829
12 245 245 246 12 244,58 1,24 0,797 242,857 246,3101
13 245 245 244 12 244,92 1,00 0,787 243,213 246,6206
14 246 245 245 12 245,33 0,65 0,776 243,652 247,0145
15 245 244 243 12 244,83 0,83 0,766 243,175 246,4916
16 244 245 244 12 244,58 0,79 0,755 242,948 246,2188
17 248 248 247 12 245,33 1,61 0,745 243,721 246,946
18 247 246 248 12 245,75 1,82 0,734 244,160 247,3398
19 245 245 245 12 246,00 1,54 0,724 244,433 247,567
20 244 245 245 12 246,08 1,44 0,713 0,713 1,677 244,888 247,2791 244,888 247,279 OK
21 246 245 248 12 245,75 1,29 0,713 244,554 246,946 OK
22 245 246 248 12 245,58 1,24 0,713 244,388 246,779 OK
23 245 246 247 12 245,83 1,27 0,713 244,638 247,029 OK
24 247 246 247 12 246,33 1,07 0,713 245,138 247,529 OK
25 245 244 244 12 245,83 1,27 0,713 244,638 247,029 OK
26 243 244 245 12 245,25 1,36 0,713 244,054 246,446 OK

NB-CPD/SG10/12/091 Page 74 of 74

You might also like