You are on page 1of 67

British Standard

A single copy of this


British Standard is licensed to
Unis Librarie
on August 17, 2000
This is an uncontrolled copy.
Ensure use of the most current
version of this standard by
searching British Standards Online
at bsonline.techindex.co.uk
DRAFT FOR DEVELOPMENT
DD ENV
1991-1:1996
Eurocode 1:
Basis of design and
actions on structures
Part 1: Basis of design
(together with United Kingdom
National Application Document)
ICS 91.040
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
DD ENV 1991-1:1996
This Draft for Development,
having been prepared under
the direction of the Sector
Board for Building and Civil
Engineering, was published
under the authority of the
Standards Board and comes
into effect on
15 September 1996
BSI 04-2000
The following BSI reference
relates to the work on this
Draft for Development:
Committee reference B/525/1
ISBN 0 580 25895 5
Committees responsible for this
Draft for Development
The preparation of this National Application Document for use in the UK with
ENV 1991-1:1996 was entrusted by Technical Committee B/525, Building and
civil engineering structures, to Subcommittee B/525/1, Actions (loadings) and
basis of design, upon which the following bodies were represented:
British Constructional Steelwork Association
British Iron and Steel Producers Association
British Masonry Society
Concrete Society
Department of the Environment (Building Research Establishment)
Department of the Environment (Property and Buildings Directorate)
Highways Agency
Institution of Structural Engineers
National House Building Council
Royal Institute of British Architects
Steel Construction Institute
Amendments issued since publication
Amd. No. Date Comments
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
DD ENV 1991-1:1996
BSI 04-2000 i
Contents
Page
Committees responsible Inside front cover
National foreword ii
Text of National Application Document v
Foreword 2
Text of ENV 1991-1 7
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
DD ENV 1991-1:1996
ii
BSI 04-2000
National foreword
This publication has been prepared by Subcommittee B/525/1. It comprises the
English language version of ENV 1991-1:1994 Eurocode 1: Basis of design and
actions on structures Part 1: Basis of design, as published by the European
Committee for Standardization (CEN), together with the corresponding National
Application Document (NAD). The NAD has been prepared for use in the design
of buildings and civil engineering works to be constructed in the United Kingdom.
ENV 1991-1 results from a programme of work sponsored by the European
Commission to make available a common set of rules for the structural and
geotechnical design of buildings and civil engineering works. The full range of
codes covers the basis of design and actions, the design of structures in concrete,
steel, composite construction, aluminium alloy, timber and masonry and also
geotechnical and seismic design.
An ENV, or European Prestandard, is made available for provisional application,
but does not have the status of a European Standard. The aim is to use the
experience gained to modify the ENV so that it can be adopted as a European
Standard.
The values of certain parameters in the ENV Eurocodes may be set by CEN
Members so as to meet requirements in national regulations. The numerical
values of these parameters in the ENV are indicated as boxed or by [ ].
During the ENV period, reference should be made to the supporting documents
listed in the National Application Documents (NADs).
Generally, the purpose of the NADs in DD ENV Eurocodes is to provide essential
information, particularly in relation to safety, to enable the corresponding ENVs
to be used for the design of buildings and civil engineering works to be
constructed in the UK. The requirements of the NADs take precedence in the UK
over the corresponding provisions in the ENVs. However, the purpose of the NAD
to Eurocode 1: Part 1 is somewhat different, as discussed in the text below.
There is no equivalent British design code to DD ENV 1991-1. Unlike other
design codes for various structural materials which contain detailed
recommendations for design, ENV 1991-1 contains only general structural
criteria which are material independent. However, the Basis of design sections
of ENV 1992 to ENV 1996 contain material common to ENV 1991-1; these
Eurocodes (ENV 1992 to ENV 1996) with their UK NADs contain sufficient
information to enable designs to be effected without recourse to ENV 1991-1.
ENV 1997-1 Geotechnical design does require recourse to ENV 1991-1, but only
for the definition of some terms and symbols, and not for quantitative values.
This situation will change when the ENV Eurocode Prestandards are converted
into full EN Eurocode Standards. Material currently in ENV 1992 to ENV 1997
which is also covered by ENV 1991-1 will then be removed and the corresponding
relevant provisions of ENV 1991-1 will apply. One implication is that the same
partial factors for loads and the same load combination factors will apply to all
structural materials, which is not currently the case.
Numerical values of coefficients in the UK NADs to ENV 1992 to ENV 1997 have
been calibrated to give an acceptable degree of conformity to current British
Standards. By contrast, the coefficients given in the UK NAD to ENV 1991-1 are
not always in accordance with current British Standards, nor with the UK NADs
to ENV 1992 to ENV 1997, although this is not expected to result in other than
minor differences in most circumstances. The differences between the loading
codes referred to in DD ENV 1991-1 and those referred to in DD ENV 1992 to
DD ENV 1997 may result in more significant differences. Where differences
remain, the Basis of design sections of ENV 1992 to ENV 1997 (as modified by
the applicable UK NAD) should take precedence over DD ENV 1991-1 during the
ENV trial application stage.
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
DD ENV 1991-1:1996
BSI 04-2000 iii
This NAD also provides clarification to certain clauses which were considered
ambiguous; these clarifications are not intended to change the original intention
of the drafters of ENV 1991-1.
The main reasons for publishing the UK NAD to Eurocode 1: Part 1 are as follows.
a) For essentially informative purposes, to enable structural designers in the
UK to familiarize themselves with the contents of ENV 1991-1, which, as
referred to above, has no existing equivalent British Standard.
b) Following from a), to enable UK comments on ENV Eurocode 1: Part 1 to be
obtained during its ENV period, so that these can be considered during the
conversion to an EN.
c) To provide information in cases where the Basis of design sections of
DD ENV 1992 to DD ENV 1997 require amplification, for example under
unusual circumstances not adequately covered by those Eurocodes.
d) To provide definitions for certain terms and symbols used in
DD ENV 1997-1.
e) To enable comparative designs to be performed which compare the approach
of ENV 1991-1 with those of DD ENV Eurocodes predating ENV 1991-1.
Compliance with DD ENV 1991-1:1996 does not of itself confer immunity
from legal obligations.
For consideration of the conversion of ENV 1991-1 into a full European Standard,
it is important to get as much feedback as possible from practising engineers.
Such feedback is therefore strongly encouraged, and users of this document are
invited to comment on its technical content, ease of use and any ambiguities or
anomalies. These comments will be taken into account when preparing the UK
national response to CEN on the question of whether the ENV can be converted
into an EN.
Comments should be made in writing to the Secretary of Subcommittee B/525/1,
BSI, 389 Chiswick High Road, London W4 4AL, quoting this document, the
reference to the relevant clause and, if possible, a proposed revision.
Summary of pages
This document comprises a front cover, an inside front cover, pages i to x,
the ENV title page, pages 2 to 52 and a back cover.
This standard has been updated (see copyright date) and may have had
amendments incorporated. This will be indicated in the amendment table on the
inside front cover.
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
iv
blank
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
DD ENV 1991-1:1996
BSI 04-2000 v
National Application
Document
for use in the UK with
ENV 1991-1:1994
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
DD ENV 1991-1:1996
vi
BSI 04-2000
Contents of
National Application Document
Page
Introduction vii
1 Scope vii
2 Informative references vii
3 Partial load factors, combination factors and other values vii
4 Loading codes vii
5 Reference standards vii
6 Additional recommendations vii
Table 1 Table and equation substitutions vii
Table 2.1 Notional classification of design working life viii
List of references ix
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
DD ENV 1991-1:1996
BSI 04-2000 vii
Introduction
This National Application Document (NAD) has
been prepared by Subcommittee B/525/1. It has
been developed from:
a) a textual examination of ENV 1991-1;
b) a comparison with the material independent
sections of the following DD ENV Eurocodes.
DD ENV 1992-1-1:1992
DD ENV 1993-1-1:1992
DD ENV 1994-1-1:1994
DD ENV 1995-1-1:1995
DD ENV 1996-1-1:1995
DD ENV 1997-1:1995
1 Scope
This National Application Document is issued to
enable the use of ENV 1991-1 in the circumstances
discussed in the national foreword, for conditions
pertinent to the UK.
2 Informative references
This National Application Document refers to other
publications that provide information or guidance.
Editions of these publications current at the time of
issue of this standard are listed on page ix, but
reference should be made to the latest editions.
3 Partial load factors, combination
factors and other values
The tables and equations of ENV 1991-1:1994 listed
below should be replaced with tables and equations
in this NAD, as is shown in Table 1. In all other
cases, the boxed values (see item 25 of the
foreword to ENV 1991-1:1994) should be used.
Table 1 Table and equation substitutions
4 Loading codes
Where punished, the UK national implementation
of the appropriate Part of Eurocode 1
(e.g. DD ENV 1991-2-3) should be used for
applications within the scope specified. Where such
DD ENVs are not available, the appropriate
equivalent standards listed in the NADs to
DD ENV 1992 to DD ENV 1997 should be adopted.
5 Reference standards
6 Additional recommendations
6.1 Sub-clause 1.5 Definitions
a) 1.5.4.3 A new definition should be added:
nominal value of a material property: a
characteristic value established from an
appropriate document such as a European
Standard or Prestandard.
b) 1.5.5.2 A new note should be added:
NOTE The design value of a geometrical property is
generally equal to the characteristic value. However, it may
differ in cases where the limit state under consideration is
very sensitive to the value of the geometrical property, for
example when considering the effect of geometrical
imperfections on buckling. In such cases, the design value will
normally be established as a value specified directly, for
example in an appropriate European Standard or
Prestandard. Alternatively, it can be established on a
statistical basis, with a value corresponding to a more
extreme fractile (i.e. a rarer value) than applies to the
characteristic value.
6.2 Clause 8. Design by testing
a) 8.1 A new paragraph should be added (5) This
section includes for fatigue within its scope..
b) 8.3 The final paragraph of (2) should be
replaced with: The field of application of the
partial factor used in method a) should be similar
to the tests under consideration..
ENV 1991-1 This NAD
Table 2.1: Design
working life
classification
Table 1: Notional
classification of design
working life
Equation 9.10a
and 9.10b: special
combination rules for
ultimate limit state
Equation 9.10a
and 9.10b should not be
used, pending
calibration work
Standard referred to in
ENV 1991-1
Equivalent standard to be
used in the UK
ISO 2631 ISO 2631
ISO 8930:1987 ISO 8930:1987
ISO 6707-1:1989 ISO 6707-1:1989
ISO 3898:1987 ISO 3898:1987
ENV Eurocode
Prestandard
Equivalent DD ENV
(see also clause 4 above)
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
DD ENV 1991-1:1996
viii
BSI 04-2000
6.3 Clause 9. Verification by the partial factor
method
a) 9.3.2 A new note should be added to (2) as
follows:
NOTE Information on appropriate values of *
Sd
for the
analysis of bridges is given in BS 5400-1..
b) (3)b) The phrase: the partial factor is applied
should be replaced by the phrase: the partial
factor *
F
is applied..
c) 9.4.2 In (3)a) equations (9.10a) and (9.10b)
should not be used, pending calibration work.
In (6) a note should be added at the end of the
paragraph:
NOTE For example, in the design of a section subject to
both bending moment and axial force due to a single action,
the axial force may be reduced by 20 % if it is a favourable
action effect..
d) 9.4.3 The second sentence of (2)P should be
replaced by the following:
NOTE Examples of where this may apply are as follows.
i) When considering Case A of Table 2 for the static
equilibrium of balanced cantilevers.
ii) When considering Case B of Table 2 for the bending
strength needed within a span of a multispan beam which
has adjacent span lengths that differ greatly.
6.4 Annex A to Annex D (informative)
Annex A to Annex D need further development work
before they can be considered adequately validated
for design purposes.
6.5 Table 2.1
Table 2.1 should be replaced by the one listed below.
Table 2.1 Notional classification of design working life
Class
Notional design
working life (years)
Examples
1 15 Temporary structures
2 25 Replaceable structural parts, e.g. gantry girders, bearings
3 50 Buildings and other common structures, other than those listed below
4 100 Monumental buildings, and other special or important structures
5 120 Bridges
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
DD ENV 1991-1:1996
BSI 04-2000 ix
List of references (see clause 2)
Informative reference
BSI standards publication
BRITISH STANDARDS INSTITUTION, London
BS 5400, Steel, concrete and composite bridges.
BS 5400-1:1988, General statement.
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
x
blank
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
EUROPEAN PRESTANDARD
PRNORME EUROPENNE
EUROPISCHE VORNORM
ENV 1991-1
September 1994
ICS 91.040.00
Descriptors: Buildings, civil engineering, structures, building codes, design, safety, reliability, mechanical strength, verification
English version
Eurocode 1 Basis of design and actions on structures
Part 1: Basis of design
Eurocode 1 Bases du calcul et actions sur les
structures
Partie 1: Bases du calcul
Eurocode 1 Grundlagen der
Tragwerksplanung und Einwirkungen
auf Tragwerke
Teil 1: Grundlagen der Tragwerksplanung
This European Prestandard (ENV) was approved by CEN on 1993-05-28 as a
prospective standard for provisional application. The period of validity of this
ENV is limited initially to three years. After two years the members of CEN
will be requested to submit their comments, particularly on the question of
whether the ENV can be converted into a European Standard (EN).
CEN members are required to announce the existence of this ENV in the same
way as for an EN and to make the ENV available promptly at national level in
an appropriate form. It is permissible to keep conflicting national standards in
force (in parallel to the ENV) until the final decision about the possible
conversion of the ENV into an EN is reached.
CEN members are the national standards bodies of Austria, Belgium,
Denmark, Finland, France, Germany, Greece, Iceland, Ireland, Italy,
Luxembourg, Netherlands, Norway, Portugal, Spain, Sweden, Switzerland and
United Kingdom.
CEN
European Committee for Standardization
Comit Europen de Normalisation
Europisches Komitee fr Normung
Central Secretariat: rue de Stassart 36, B-1050 Brussels
1994 Copyright reserved to CEN members
Ref. No. ENV 1991-1:1994 E
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
BSI 04-2000
2
Foreword
Objectives of the Eurocodes
(1) The Structural Eurocodes comprise a group of
standards for the structural and geotechnical design
of buildings and civil engineering works.
(2) They cover execution and control only to the
extent that is necessary to indicate the quality of the
construction products, and the standard of the
workmanship, needed to comply with the
assumptions of the design rules.
(3) Until the necessary set of harmonized technical
specifications for products and for methods of
testing their performance are available, some of the
Structural Eurocodes cover some of these aspects in
informative annexes.
Background to the Eurocode Programme
(4) The Commission of the European Communities
(CEC) initiated the work of establishing a set of
harmonized technical rules for the design of
building and civil engineering works which would
initially serve as an alternative to the different rules
in force in the various member states and would
ultimately replace them. These technical rules
became known as the Structural Eurocodes.
(5) In 1990, after consulting their respective
member states, the CEC transferred the work of
further development, issue and updating of the
Structural Eurocodes to CEN, and the EFTA
secretariat agreed to support the CEN work.
(6) CEN Technical Committee CEN/TC 250 is
responsible for all Structural Eurocodes.
Eurocode Programme
(7) Work is in hand on the following Structural
Eurocodes, each generally consisting of a number of
parts:
EN 1991, Eurocode 1: Basis of design and actions
on structures.
EN 1992, Eurocode 2: Design of concrete
structures.
EN 1993, Eurocode 3: Design of steel structures.
EN 1994, Eurocode 4: Design of composite steel
and concrete structures.
EN 1995, Eurocode 5: Design of timber
structures.
EN 1996, Eurocode 6: Design of masonry
structures.
EN 1997, Eurocode 7: Geotechnical design.
EN 1998, Eurocode 8: Design of structures for
earthquake resistance.
EN 1999, Eurocode 9: Design of aluminium alloy
structures.
(8) Separate subcommittees have been formed by
CEN/TC 250 for the various Eurocodes listed above.
(9) This Part of ENV 1991 is intended to develop for
a broader field of application the rules already
published in sections 1 and 2 of Parts 1.1 of
ENVs 1992, 1993 and 1994. It is being published as
European Prestandard ENV 1991-1.
(10) This prestandard is intended for experimental
application and for the submission of comments.
(11) After approximately two years CEN members
will be invited to submit formal comments to be
taken into account in determining future actions.
(12) Meanwhile feedback and comments on this
prestandard should be sent to the secretariat of
CEN/TC 250 at the following address:
BSI
British Standards House
389 Chiswick High Road
London W4
England
or to your national standards organization.
Purpose of this Part of Eurocode 1
Technical objectives
(13) This Part of Eurocode 1 describes the principles
and requirements for safety, serviceability and
durability of structures. It is based on the limit state
concept used in conjunction with a partial factor
method. Regarding modifications of the proposed
method, see (24) of the foreword.
(14) For the design of new structures, this Part is
intended to be used, for direct application, together
with:
the other Parts of ENV 1991;
the design Eurocodes (ENVs 1992 to 1999).
NOTE The above mentioned European Prestandards are either
published or in preparation.
(15) This Part also gives guidelines for the aspects of
structural reliability relating to safety,
serviceability and durability:
for design cases not covered by
ENVs 1991 to 1999 (other actions, structures not
treated, other materials);
to serve as a reference document for other
CEN TCs concerned with structural aspects.
(16) It is intended that the material-independent
clauses in section 2 of the design Eurocodes will be
superseded by this Part of ENV 1991 at a future
stage (EN stage).
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
BSI 04-2000 3
Intended users
(17) This prestandard is intended for the
consideration of more categories of users, than are
the other Eurocodes. The categories include:
code drafting committees;
clients (e.g. for the formulation of their specific
requirements on reliability level and durability);
designers and contractors, as for other
Eurocodes;
public authorities.
Intended uses
(18) This prestandard is intended for the design of
structures within the scope of the Eurocodes.
(19) As a guidance document, for the design of
structures outside the scope of the Eurocodes, this
prestandard may be used when relevant for:
assessing other actions and their
combinations;
modelling material and structural behaviour;
assessing numerical values of the reliability
format.
(20) Numerical values for safety factors and other
safety elements are given as indications. Together
with the material-dependent indicative values given
in the design Eurocodes, they provide an acceptable
degree of reliability, assuming that an appropriate
level of workmanship and of quality assurance is
achieved. Therefore, if this Part is used as a
reference document by other CEN/TCs the same
indicative values should be taken.
Division into main text and annexes
(21) Because of the various categories of use
mentioned above, this Part is divided into a main
text and a series of annexes. This division also takes
into account the development expected during the
ENV period.
(22) The main text includes most of the principal
and operational rules necessary for direct
application for designs in the field covered by
ENV 1991, and ENVs 1992 to 1999. The principal
provisions for bridges are also included.
(23) The annexes are informative only. Other
background information and items for further
development during the ENV period may be
published separately in a CEN report.
National Application Documents (NADs)
(24) It is intended that, during the ENV period, this
prestandard is used for design purposes, in
conjunction with the particular National
Application Document valid in the country where
the designed structures are to be located.
The National Application Documents are intended
to authorize experimental use of the Eurocodes as
prestandards for design during the ENV period,
with due consideration for the current regulations
and codes relevant in individual countries, and to
facilitate these uses. The NADs may also introduce
modifications of the partial factor method in this
prestandard. Establishing the NAD is the
responsibility of the national competent authorities.
In particular each NAD may specify whether the
annexes can be used fully or partly in connection
with the main text and what are then the specific
conditions for their application, e.g. the application
of 3.4(3), and of 8.3(1) together with Annex A.
(25) In particular, for this prestandard attention
should be paid to:
confirming or amending the numerical values
identified as boxed or by [ ]; it is recommended
that modifications are introduced only where
considered to be necessary; however, for those
countries in which reliability differentiation
measures are already codified there is no
objection to numerical amendments intended to
supplement this Eurocode by such operational
measures;
considering the variety of intended users and
uses of this prestandard [see (17) above], with
regard to the existing national professional
organizations and the respective responsibilities
of each category of user.
Intended future developments of this Part
(26) The objective of this Part is to ensure the
consistency of design rules for a wide set of
construction works made of various materials. It
should be understood that this is a long-term
objective which will be reached progressively. At the
present stage the objective is limited to:
ensuring the consistency between the
Eurocodes already published or in preparation,
without contradicting them;
covering the structures treated in the same
Eurocodes in less detail for those for which Parts
of Eurocodes are in preparation, e.g. for bridges,
silos, etc. Therefore it should be understood that
by publication of the present version of this Part
it is not intended to inhibit the work of
development and improvement of the reliability
format.
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
4
BSI 04-2000
In parallel with the publication of new Parts of
Eurocodes during the ENV period, it is envisaged
that some developments may be made to some
items, e.g:
more precise definition of differentiated
reliability levels;
numerical revision, probabilistic justification
of numerical values of partial factors and possibly
supplementing this method with a probabilistic
approach;
more precise consideration of various types of
limit state equations, soil-structure interaction,
non-linear analysis, dynamic actions and the
associated analysis and reliability verification
format;
assessment and re-design of existing
structures.
Contents
Page
Foreword 2
1 General
1.1 Scope 7
1.2 Normative references 7
1.3 Assumptions 8
1.4 Distinction between principles
and application rules 8
1.5 Definitions 8
1.5.1 Common terms used in the Structural
Eurocodes (ENVs 1991-1999) 8
1.5.2 Special terms relating to design
in general 9
1.5.3 Terms relating to actions 10
1.5.4 Terms relating to material properties 12
1.5.5 Terms relating to geometric data 12
1.6 Symbols 13
2 Requirements
2.1 Fundamental requirements 15
2.2 Reliability differentiation 15
2.3 Design situations 16
2.4 Design working life 16
2.5 Durability 16
2.6 Quality assurance 17
3 Limit states
3.1 General 18
3.2 Ultimate limit states 18
3.3 Serviceability limit states 18
3.4 Limit state design 18
Page
4 Actions and environmental influences
4.1 Principal classifications 20
4.2 Characteristic values of actions 20
4.3 Other representative values of
variable and accidental actions 21
4.4 Environmental influences 22
5 Material Properties
6 Geometrical data
7 Modelling for structural analysis
and resistance
7.1 General 25
7.2 Modelling in the case of static actions 25
7.3 Modelling in the case of dynamic
actions 25
7.4 Modelling for fire actions 25
8 Design assisted by testing
8.1 General 26
8.2 Types of tests 26
8.3 Derivation of design values 26
9 Verification by the partial factor
method
9.1 Introduction 28
9.2 Limitations and simplifications 28
9.3 Design values 28
9.3.1 Design values of actions 28
9.3.2 Design values of the effects of actions 29
9.3.3 Design values of material properties 29
9.3.4 Design values of geometric data 29
9.3.5 Design resistance 30
9.4 Ultimate limit states 30
9.4.1 Verification of static equilibrium
and strength 30
9.4.2 Combination of actions 31
9.4.3 Partial factors 32
9.4.4 ? factors 34
9.4.5 Simplified verification for
building structures 34
9.4.6 Partial safety factors for materials 34
9.5 Serviceability limit states 34
9.5.1 Verifications of serviceability 34
9.5.2 Combination of actions 35
9.5.3 Partial factors 35
9.5.4 ? factors 35
9.5.5 Simplified verification for
building structures 35
9.5.6 Partial factors for materials 36
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
BSI 04-2000 5
Page
Annex A (informative) Partial factor design 37
Annex B (informative) Fatigue 43
Annex C (informative) Serviceability
limit state: verification of structures
susceptible to vibrations 44
Annex D (informative) Design assisted
by testing 46
Figure A.1 Overview of reliability methods 38
Figure A.2 Design point definition
according to first order reliability
methods (FORM) 40
Table 2.1 Design working life classification 16
Table 9.1 Design values of actions for
use in the combination of actions 31
Table 9.2 Partial factors: ultimate limit
states for buildings 33
Table 9.3 ? factors for buildings 34
Table 9.4 Design values of actions for
use in the combination of actions 35
Table A.1 Relation between " and P
1
38
Table A.2 Indicative values for the
target reliability index ". 39
Table A.3 Design values for various
distribution functions 40
Table A.4 Expression for ?
o
42
Table D.1 Values of k
n
for the 5 %
characteristic value 49
Table D.2 Values of k
n
for the ULS
design value, if X is dominating
(P{X < X
d
} = 0,1 %) 50
Table D.3 Values of k
n
for the ULS
design value, if X is non-dominating
(P{X < X
d
} = 10 %) 50
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
6
blank
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
BSI 04-2000 7
Section 1. General
1.1 Scope
(1) This Part 1 of ENV 1991 establishes the principles and requirements for safety and serviceability of
structures, describes the basis for design and verification and gives guidelines for related aspects of
structural reliability.
(2)P Part 1 of ENV 1991 provides the basis and general principles for the structural design of buildings and
civil engineering works including geotechnical aspects and shall be used in conjunction with the other parts
of ENV 1991 and ENVs 1992 to 1999. Part 1 relates to all circumstances in which a structure is required
to give adequate performance, including fire and seismic events.
(3) Part 1 of ENV 1991 may also be used as a basis for the design of structures not covered in
ENVs 1992 to 1999 and where other materials or other actions outside the scope of ENV 1991 are involved.
(4)P Part 1 of ENV 1991 is also applicable to structural design for the execution stage and structural design
for temporary structures, provided that appropriate adjustments outside the scope of ENV 1991 are made.
(5) Part 1 of ENV 1991 also gives some simplified methods of verification which are applicable to buildings
and other common construction works.
(6) Design procedures and data relevant to the design of bridges and other construction works which are
not completely covered in this Part may be obtained from other Parts of Eurocode 1 and other relevant
Eurocodes.
(7) Part 1 of ENV 1991 is not directly intended for the structural appraisal of existing construction in
developing the design of repairs and alterations or assessing changes of use but may be so used where
applicable.
(8) Part 1 of ENV 1991 does not completely cover the design of special construction works which require
unusual reliability considerations, such as nuclear structures, for which specific design procedures should
be used.
(9) Part 1 of ENV 1991 does not completely cover the design of structures where deformations modify direct
actions.
1.2 Normative references
This European Prestandard incorporates by dated or undated reference, provisions from other standards.
These normative references are cited at the appropriate places in the text and publications listed hereafter.
ISO 2631, Evaluation of human exposure to whole-body vibration.
ISO 8930:1987, General principles on reliability for structures List of equivalent terms.
ISO 6707-1:1989, Building/civil engineering Vocabulary Part 1: General terms.
ISO 3898:1987, Basis of design for structures Notations General symbols.
NOTE The following European Prestandard which are published or in preparation are cited at the appropriate places in the text
and publications listed hereafter.
ENV 1991-1, Eurocode 1: Basis of design and actions on structures Part 1: Basis of design.
ENV 1991-2-1, Eurocode 1: Basis of design and actions on structures Part 2.1: Densities, self-weight and imposed loads.
ENV 1991-2-2, Eurocode 1: Basis of design and actions on structures Part 2.2: Actions on structures exposed to fire.
ENV 1991-2-3, Eurocode 1: Basis of design and actions on structures Part 2.3: Snow loads.
ENV 1991-2-4, Eurocode 1: Basis of design and actions on structures Part 2.4: Wind loads.
ENV 1991-2-5, Eurocode 1: Basis of design and actions on structures Part 2.5: Thermal actions.
ENV 1991-2-6, Eurocode 1: Basis of design and actions on structures Loads and deformations imposed during execution.
ENV 1991-2-7, Eurocode 1: Basis of design and actions on structures Part 2.7: Accidental actions.
ENV 1991-3, Eurocode 1: Basis of design and actions on structures Part 3: Traffic loads on bridges.
ENV 1991-4, Eurocode 1: Basis of design and actions on structures Part 4: Actions in silos and tanks.
ENV 1991-5, Eurocode 1: Basis of design and actions on structures Part 5: Actions induced by cranes and machinery.
ENV 1992, Eurocode 2: Design of concrete structures.
ENV 1993, Eurocode 3: Design of steel structures.
ENV 1994, Eurocode 4: Design of composite steel and concrete structures.
ENV 1995, Eurocode 5: Design of timber structures.
ENV 1996, Eurocode 6: Design of masonry structures.
ENV 1997, Eurocode 7: Geotechnical design.
ENV 1998, Eurocode 8: Earthquake resistant design of structures.
ENV 1999, Eurocode 9: Design of aluminium alloy structures.
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
8
BSI 04-2000
1.3 Assumptions
The following assumptions apply:
The choice of the structural system and the design of a structure is made by appropriately qualified
and experienced personnel.
Execution is carried out by personnel having the appropriate skill and experience.
Adequate supervision and quality control is provided during execution of the work, i.e. in design
offices, factories, plants, and on site.
The construction materials and products are used as specified in this Eurocode or in
ENVs 1992 to 1999 or in the relevant supporting material or product specifications.
The structure will be adequately maintained.
The structure will be used in accordance with the design assumptions.
Design procedures are valid only when the requirements for the materials, execution and
workmanship given in ENVs 1992 to 1996 and 1999 are also complied with.
1.4 Distinction between principles and application rules
(1)P Depending on the character of the individual clauses, distinction is made in this Part 1 of ENV 1991
between principles and application rules.
(2)P The principles comprise:
general statements and definitions for which there is no alternative;
requirements and analytical models for which no alternative is permitted unless specifically stated.
(3) The principles are identified by the letter P following the paragraph number.
(4)P The application rules are generally recognized rules which follow the principles and satisfy their
requirements. It is permissible to use alternative rules to the application rules given in this Eurocode,
provided that it is shown that the alternative rules accord with the relevant principles and have at least
the same reliability.
(5) In this Part of ENV 1991 the application rules have only a paragraph number, e.g. as this paragraph.
1.5 Definitions
For the purposes of this prestandard, the following definitions apply.
NOTE Most definitions are reproduced from ISO 8930:1987.
1.5.1 Common terms used in the Structural Eurocodes (ENVs 1991 to 1999)
1.5.1.1
construction works
everything that is constructed or results from construction operations
NOTE This definition accords with ISO 6707-1. The term covers both building and civil engineering works. It refers to the complete
construction worlds comprising structural, non-structural and geotechnical elements.
1.5.1.2
type of building or civil engineering works
type of construction works designating its intended purpose, e.g. dwelling house, retaining wall, industrial
building, road bridge
1.5.1.3
type of construction
indication of principal structural material, e.g. reinforced concrete construction, steel construction, timber
construction, masonry construction, composite steel and concrete construction
1.5.1.4
method of construction
manner in which the execution will be carried out, e.g. cast in place, prefabricated, cantilevered
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
BSI 04-2000 9
1.5.1.5
construction material
material used in construction work, e.g. concrete, steel, timber, masonry
1.5.1.6
structure
organized combination of connected parts designed to provide some measure of rigidity
NOTE ISO 6707-1 gives the same definition but adds or a construction works having such an arrangement. In the Structural
Eurocodes this addition is not used in order to facilitate unambiguous translation.
1.5.1.7
form of structure
the arrangement of structural elements, such as beam, column, arch, foundation piles
NOTE Forms of structure are, for example, frames, suspension bridges.
1.5.1.8
structural system
the load-bearing elements of a building or civil engineering works and the way in which these elements
function together
1.5.1.9
structural model
the idealization of the structural system used for the purposes of analysis and design
1.5.1.10
execution
the activity of creating a building or civil engineering works
NOTE The term covers work on site; it may also signify the fabrication of components off site and their subsequent erection on site.
1.5.2 Special terms relating to design in general
1.5.2.1
design criteria
the quantitative formulations which describe for each limit state the conditions to be fulfilled
1.5.2.2
design situations
those sets of physical conditions representing a certain time interval for which the design will demonstrate
that relevant limit states are not exceeded
1.5.2.3
transient design situation
design situation which is relevant during a period much shorter than the design working life of the
structure and which has a high probability of occurrence
NOTE It refers to temporary conditions of the structure, of use, or exposure, e.g. during construction or repair.
1.5.2.4
persistent design situation
design situation which is relevant during a period of the same order as the design working life of the
structure
NOTE Generally it refers to conditions of normal use.
1.5.2.5
accidental design situation
design situation involving exceptional conditions of the structure or its exposure, e.g. fire, explosion, impact
or local failure
1.5.2.6
design working life
the assumed period for which a structure is to be used for its intended purpose with anticipated
maintenance but without substantial repair being necessary
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
10
BSI 04-2000
1.5.2.7
hazard
exceptionally unusual and severe event, e.g. an abnormal action or environmental influence, insufficient
strength or resistance, or excessive deviation from intended dimensions
1.5.2.8
load arrangement
identification of the position, magnitude and direction of a free action
1.5.2.9
load case
compatible load arrangements, sets of deformations and imperfections considered simultaneously with
fixed variable actions and permanent actions for a particular verification
1.5.2.10
limit states
states beyond which the structure no longer satisfies the design performance requirements
1.5.2.11
ultimate limit states
states associated with collapse, or with other similar forms of structural failure
NOTE They generally correspond to the maximum load-carrying resistance of a structure or structural part.
1.5.2.12 serviceability limit states
States which correspond to conditions beyond which specified service requirements for a structure or
structural element are no longer met.
1.5.2.12.1
irreversible serviceability limit states
limit states which will remain permanently exceeded when the responsible actions are removed
1.5.2.12.2
reversible serviceability limit states
limit states which will not be exceeded when the responsible actions are removed
1.5.2.13
resistance
mechanical property of a component, a cross-section, or a number of a structure, e.g. bending resistance,
buckling resistance
1.5.2.14
maintenance
the total set of activities performed during the working life of the structure to preserve its function
1.5.2.15
strength
mechanical property of a material, usually given in units of stress
1.5.2.16
reliability
reliability covers safety, serviceability and durability of a structure
1.5.3 Terms relating to actions
1.5.3.1
action
a) Force (load) applied to the structure (direct action)
b) An imposed or constrained deformation or an imposed acceleration caused for example, by
temperature changes, moisture variation, uneven settlement or earthquakes (indirect action).
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
BSI 04-2000 11
1.5.3.2
action effect
the effect of actions on structural members, e.g. internal force, moment, stress, strain
1.5.3.3
permanent action (G)
action which is likely to act throughout a given design situation and for which the variation in magnitude
with time is negligible in relation to the mean value, or for which the variation is always in the same
direction (monotonic) until the action attains a certain limit value
1.5.3.4
variable action (Q)
action which is unlikely to act throughout a given design situation or for which the variation in magnitude
with time is neither negligible in relation to the mean value nor monotonic
1.5.3.5
accidental action (A)
action, usually of short duration, which is unlikely to occur with a significant magnitude over the period of
time under consideration during the design working life
NOTE An accidental action can be expected in many cases to cause severe consequences unless special measures are taken.
1.5.3.6
seismic action (A
E
)
action which arises due to earthquake ground motions
1.5.3.7
fixed action
action which has a fixed distribution over the structure such that the magnitude and direction of the action
are determined unambiguously for the whole structure if this magnitude and direction are determined at
one point on the structure
1.5.3.8
free action
action which may have any spatial distribution over the structure within given limits
1.5.3.9
single action
action that can be assumed to be statistically independent in time and space of any other action acting on
the structure
1.5.3.10
static action
action which does not cause significant acceleration of the structure or structural members
1.5.3.11
dynamic action
action which causes significant acceleration of the structure or structural members
1.5.3.12
quasi-static action
dynamic action that can be described by static models in which the dynamic effects are included
1.5.3.13
representative value of an action
value used for the verification of a limit state
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
12
BSI 04-2000
1.5.3.14
characteristic value of an action
the principal representative value of an action. In so far as this characteristic value can be fixed on
statistical bases, it is chosen so as to correspond to a prescribed probability of not being exceeded on the
unfavourable side during a reference period taking into account the design working life of the structure
and the duration of the design situation
1.5.3.15
reference period
See 1.5.3.14.
1.5.3.16
combination values
values associated with the use of combinations of actions (see 1.5.3.20) to take account of a reduced
probability of simultaneous occurrence of the most unfavourable values of several independent actions
1.5.3.17
frequent value of a variable action
the value determined such that:
the total time, within a chosen period of time, during which it is exceeded for a specified part, or
the frequency with which it is exceeded,
is limited to a given value.
1.5.3.18
quasi-permanent value of a variable action
the value determined such that the total time, within a chosen period of time, during which it is exceeded
is a considerable part of the chosen period of time
1.5.3.19
design value of an action F
d
the value obtained by multiplying the representative value by the partial safety factor *
F
1.5.3.20
combination of actions
set of design values used for the verification of the structural reliability for a limit state under the
simultaneous influence of different actions
1.5.4 Terms relating to material properties
1.5.4.1
characteristic value X
k
the value of a material property having a prescribed probability of not being attained in a hypothetical
unlimited test series. This value generally corresponds to a specified fractile of the assumed statistical
distribution of the particular property of the material. A nominal value is used as the characteristic value
in some circumstances
1.5.4.2
design value of a material property X
d
value obtained by dividing the characteristic value by a partial factor *
M
or, in special circumstances, by
direct determination
1.5.5 Terms relating to geometrical data
1.5.5.1
characteristic value of a geometrical property a
k
the value usually corresponding to the dimensions specified in the design. Where relevant, values of
geometrical quantities may correspond to some prescribed fractile of the statistical distribution
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
BSI 04-2000 13
1.5.5.2
design value of a geometrical property a
d
generally a nominal value. Where relevant, values of geometrical quantities may correspond to some
prescribed fractile of the statistical distribution
1.6 Symbols
For the purposes of this prestandard, the following symbols apply.
NOTE The notation used is based on ISO 3898:1987
Latin upper case letters
A Accidental action
A
d
Design value of an accidental action
A
Ed
Design value of seismic action
A
Ek
Characteristic seismic action
A
k
Characteristic value of an accidental action
C
d
Nominal value, or a function of certain design properties of materials
E Effect of an action
E
d
Design value of effects of actions
E
d,dst
Design effect of destabilizing actions
E
d,stb
Design effect of stabilizing actions
F Action
F
d
Design value of an action
F
k
Characteristic value of an action
F
rep
Representative value of an action
G Permanent action
G
d
Design value of a permanent action
G
d,inf
Lower design value of a permanent action
G
id
Characteristic value of permanent action j
G
d,sup
Upper design value of a permanent action
G
ind
Indirect permanent action
G
k
Characteristic value of a permanent action
G
k,inf
Lower characteristic value of a permanent action
G
k,sup
Upper characteristic value of a permanent action
P Prestressing action
P
d
Design value of a prestressing action
P
k
Characteristic value of a prestressing action
Q Variable action
Q
d
Design value of a variable action
Q
ind
Indirect variable action
Q
k
Characteristic value of a single variable action
Q
k1
Characteristic value of the dominant variable action
Q
id
Characteristic value of the non-dominant variable action i
R Resistance
R
d
Design value of the resistance
R
k
Characteristic resistance
X Material property
X
d
Design value of a material property
X
k
Characteristic value of a material property
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
14
BSI 04-2000
Latin lower case letters
a
d
Design value of geometrical data
a
k
Characteristic dimension
a
nom
Nominal value of geometrical data
Greek upper case letters
%
a
Change made to nominal geometrical data for particular design purposes, e.g. assessment of
effects of imperfections
Greek lower case letters
* Partial factor (safety or serviceability)
*
A
Partial factor for accidental actions
*
F
Partial factor for actions, also accounting for model uncertainties and dimensional variations
*
G
Partial factor for permanent actions
*
GA
As *
G
but for accidental design situations
*
GAj
As *
Gj
but for accidental design situations
*
G,inf
Partial factor for permanent actions in calculating lower design values
*
Gj
Partial factor for permanent action j
*
G,sup
Partial factor for permanent actions in calculating upper design values
*
I
Importance factor
*
m
Partial factor for a material property
*
M
Partial factor for a material property, also accounting for model uncertainties and dimensional
variations
*
P
Partial factor for prestressing actions
*
PA
As *
p
but for accidental design situations
*
Q
Partial factor for variable actions
*
Qi
Partial factor for variable action i
*
rd
Partial factor associated with the uncertainty of the resistance model and the dimensional
variations
*
R
Partial factor for the resistance, including uncertainties in material properties, model
uncertainties and dimensional variations
*
Rd
Partial factor associated with the uncertainty of the resistance model
*
Sd
Partial factor associated with the uncertainty of the action and/or action effect model
) Conversion factor
K Reduction factor
?
0
Coefficient for combination value of a variable action
?
1
Coefficient for frequent value of a variable action
?
2
Coefficient for quasi-permanent value of a variable action
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
BSI 04-2000 15
Section 2. Requirements
2.1 Fundamental requirements
(1)P A structure shall be designed and executed in such a way that it will, during its intended life with
appropriate degrees of reliability and in an economic way:
remain fit for the use for which it is required; and
sustain all actions and influences likely to occur during execution and use.
(2) Design according to 2.1(1) implies that due regard is given to structural safety and serviceability,
including durability, in both cases.
(3)P A structure shall also be designed and executed in such a way that it will not be damaged by events
like fire, explosion, impact or consequences of human errors, to an extent disproportionate to the original
cause.
(4)P The potential damage shall be avoided or limited by appropriate choice of one or more of the following:
avoiding, eliminating or reducing the hazards which the structure may sustain;
selecting a structural form which has low sensitivity to the hazards considered;
selecting a structural form and design that can survive adequately the accidental removal of an
individual element or a limited part of the structure, or the occurrence of acceptable localized damage;
avoiding as far as possible structural systems which may collapse without warning;
tying the structure together.
(5)P The above requirements shall be met by the choice of suitable materials, by appropriate design and
detailing, and by specifying control procedures for design, production, execution and use relevant to the
particular project.
2.2 Reliability differentiation
(1)P The reliability required for the majority of structures shall be obtained by design and execution
according to ENVs 1991-1999, and appropriate quality assurance measures.
(2) A different level of reliability may be generally adopted:
for structural safety;
for serviceability;
(3) A different level of reliability may depend on:
the cause and mode of failure;
the possible consequences of failure in terms of risk to life, injury, potential economic losses and the
level of social inconvenience;
the expense and procedures necessary to reduce the risk of failure;
different degrees of reliability required at national, regional or local level.
(4) Differentiation of the required levels of reliability in relation to structural safety and serviceability may
be obtained by the classification of whole structures or by the classification of structural components.
(5) The required reliability relating to structural safety or serviceability may be achieved by suitable
combinations of the following measures:
a) Measures relating to design:
serviceability requirements;
representative values of actions;
the choice of partial factors or appropriate quantities in design calculations;
consideration of durability;
consideration of the degree of robustness (structural integrity);
the amount and quality of preliminary investigations of soils and possible environmental
influences;
the accuracy of the mechanical models used;
the stringency of the detailing rules.
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
16
BSI 04-2000
b) Measures relating to quality assurance to reduce the risk of hazards in:
gross human errors;
design;
execution.
(6) Within individual reliability levels, the procedures to reduce risks associated with various potential
causes of failure may, in certain circumstances, be interchanged to a limited extent. An increase of effort
within one type of measure may be considered to compensate for a reduction of effort within another type.
2.3 Design situations
(1)P The circumstances in which the structure may be required to fulfil its function shall be considered and
the relevant design situations selected. The selected design situations shall be sufficiently severe and so
varied as to encompass all conditions which can reasonably be foreseen to occur during the execution and
use of the structure.
(2)P Design situations are classified as follows:
persistent situations, which refer to the conditions of normal use;
transient situations, which refer to temporary conditions applicable to the structure, e.g. during
execution or repair;
accidental situations, which refer to exceptional conditions applicable to the structure or to its
exposure, e.g. to fire, explosion, impact;
seismic situations, which refer to exceptional conditions applicable to the structure when subjected to
seismic events.
(3) Information for specific situations for each class is given in other Parts of ENV 1991 and in
ENVs 1992 to 1999.
2.4 Design working life
(1)P The design working life is the assumed period for which a structure is to be used for its intended
purpose with anticipated maintenance but without major repair being necessary.
(2) An indication of the required design working life is given in Table 2.1.
Table 2.1 Design working life classification
2.5 Durability
(1) It is an assumption in design that the durability of a structure or part of it in its environment is such
that it remains fit for use during the design working life given appropriate maintenance.
(2) The structure should be designed in such a way that deterioration should not impair the durability and
performance of the structure having due regard to the anticipated level of maintenance.
(3)P The following interrelated factors shall be considered to ensure an adequately durable structure:
the intended and possible future use of the structure;
the required performance criteria;
the expected environmental influences;
the composition, properties and performance of the materials;
the choice of the structural system;
the shape of members and the structural detailing;
Class
Required Design
working life (years)
Example
1 [15] Temporary structures
2 [25] Replaceable structural parts, e.g. gantry girders, bearings
3 [50] Building structures and other common structures
4 [100] Monumental building structures, bridges, and other civil engineering
structures
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
BSI 04-2000 17
the quality of workmanship, and level of control;
the particular protective measures;
the maintenance during the intended life.
(4) The relevant ENVs 1992-1999 specify the appropriate measures.
(5)P The environmental conditions shall be appraised at the design stage to assess their significance in
relation to durability and to enable adequate provisions to be made for protection of the materials and
products.
(6) The degree of deterioration may be estimated on the basis of calculations, experimental investigation,
experience from earlier constructions, or a combination of these considerations.
2.6 Quality assurance
(1) It is assumed that appropriate quality assurance measures are taken in order to provide a structure
which corresponds to the requirements and to the assumptions made in the design. These measures
comprise definition of the reliability requirements, organizational measures and controls at the stages of
design, execution, use and maintenance.
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
18
BSI 04-2000
Section 3. Limit states
3.1 General
(1)P Limit states are states beyond which the structure no longer satisfies the design performance
requirements.
(2) In general, a distinction is made between ultimate limit states and serviceability limit states.
NOTE Verification of one of the two limit states may be omitted if sufficient information is available to prove that the requirements
of one limit state are met by the other.
(3) Limit states may relate to persistent, transient or accidental design situations.
3.2 Ultimate limit states
(1)P Ultimate limit states are those associated with collapse or with other similar forms of structural
failure.
(2) States prior to structural collapse, which, for simplicity, are considered in place of the collapse itself are
also treated as ultimate limit states.
(3)P Ultimate limit states concern:
the safety of the structure and its contents;
the safety of people.
(4) Ultimate limit states which may require consideration include:
loss of equilibrium of the structure or any part of it, considered as a rigid body;
failure by excessive deformation, transformation of the structure or any part of it into a mechanism,
rupture, loss of stability of the structure or any part of it, including supports and foundations;
failure caused by fatigue or other time-dependent effects.
3.3 Serviceability limit states
(1)P Serviceability limit states correspond to conditions beyond which specified service requirements for a
structure or structural element are no longer met.
(2)P The serviceability requirements concern:
the functioning of the construction works or parts of them;
the comfort of people;
the appearance.
(3)P A distinction shall be made, if relevant, between reversible and irreversible serviceability limit states.
(4) Unless specified otherwise, the serviceability requirements should be determined in contracts and/or in
the design.
(5) Serviceability limit states which may require consideration include:
deformations and displacements which affect the appearance or effective use of the structure
(including the functioning of machines or services) or cause damage to finishes or non-structural
elements;
vibrations which cause discomfort to people, damage to the structure or to the materials it supports,
or which limit its functional effectiveness;.
damage (including cracking) which is likely to affect appearance, durability or the function of the
structure adversely;
observable damage caused by fatigue and other time-dependent effects.
3.4 Limit state design
(1)P Limit state design shall be carried out by:
setting up structural and load models for relevant ultimate and serviceability limit states to be
considered in the various design situations and load cases;
verifying that the limit states are not exceeded when design values for actions, material properties
and geometrical data are used in the models.
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
BSI 04-2000 19
(2) Design values are generally obtained by using the characteristic or representative values (as defined in
sections 4 to 6 and specified in ENVs 1991-1999) in combination with partial and other factors as defined
in section 9 and ENV 1991 to 1999.
(3) In exceptional cases, it may be appropriate to determine design values directly. The values should be
chosen cautiously and should correspond to at least the same degree of reliability for the various limit
states as implied in the partial factors in this code (see also section 8).
NOTE 1 Partial factor design is discussed in Annex A.
NOTE 2 Principles and application rules for verification are given in section 9.
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
20
BSI 04-2000
Section 4. Actions and environmental influences
4.1 Principal classifications
(1)P An action (F) is:
a direct action, i.e. force (load) applied to the structure; or
an indirect action, i.e. an imposed or constrained deformation or an imposed acceleration caused, for
example, by temperature changes, moisture variation, uneven settlement or earthquakes.
(2)P Actions are classified:
a) by their variation in time:
permanent actions (G), e.g. self-weight of structures, fixed equipment and road surfacings;
variable actions (Q), e.g. imposed loads, wind loads or snow loads;
accidental actions (A), e.g. explosions, or impact from vehicles.
b) by their spatial variation:
fixed actions, e.g. self-weight;
free actions, e.g. movable imposed loads, wind loads, snow loads.
c) by their nature and/or the structural response:
static actions, which do not cause significant acceleration of the structure or structural member;
dynamic actions, which cause significant acceleration of the structure or structural member.
(3) In many cases, dynamic effects of actions may be calculated from quasi-static actions by increasing the
magnitude of the static actions or by the introduction of an equivalent static action (see 7.3).
(4) Some actions, for example seismic actions and snow loads, can be considered as either accidental
and/or variable actions, depending on the site location (see other Parts of ENV 1991).
(5) Prestressing (P) is a permanent action. Detailed information is given in ENVs 1992, 1993, and 1994.
(6) Indirect actions are either permanent G
ind,
(e.g. settlement of support), or variable Q
ind,

(e.g. temperature effect), and should be treated accordingly.
(7) An action is described by a model, its magnitude being represented in the most common cases by one
scalar which may take on several representative values. For some actions (multi-component actions) and
some verifications (e.g. for static equilibrium) the magnitude is represented by several values. For fatigue
verifications and dynamic analysis a more complex representation of the magnitudes of some actions may
be necessary.
4.2 Characteristic values of actions
(1)P The characteristic value of an action is its main representative value.
(2)P Characteristic values of actions F
k
shall be specified:
in the relevant parts of ENV 1991, as a mean value, an upper or lower value, or a nominal value
(which does not refer to a known statistical distribution);
in the design, provided that the provisions, specified in ENV 1991 are observed.
NOTE The provisions may be specified by the relevant competent authority.
(3)P The characteristic value of a permanent action shall be determined as follows:
if the variability of G is small, one single value G
k
may be used;
if the variability of G is not small, two values have to be used; an upper value G
k,sup
and a lower
value G
k,inf
.
(4) In most cases the variability of G can be assumed to be small if G does not vary significantly during the
design working life of the structure and its coefficient of variation is not greater than [0,1]. However in such
cases when the structure is very sensitive to variations in G (e.g. some types of prestressed concrete
structures), two values have to be used even if the coefficient of variation is small.
(5) The following may be assumed in most cases:
G
k
is the mean value;
G
idnf
is the [0,05] fractile, and G
ksup
is the [0,95] fractile of the statistical; distribution for G which may
be assumed to be Gaussian.
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
BSI 04-2000 21
(6) The self-weight of the structure can, in most cases, be represented by a single characteristic value and
be calculated on the basis of the nominal dimensions and mean unit masses. The values are given in
ENV 1991-2.
(7)P For variable actions the characteristic value (Q
k
) corresponds to either:
an upper value with an intended probability of not being exceeded or a lower value with an intended
probability of not falling below, during some reference period;
a nominal value which may be specified in cases where a statistical distribution is not known.
Values are given in ENVs 1991-2 and 1991-3.
(8) The following may be assumed for the time-varying part for most cases of characteristic values of
variable actions:
the intended probability is [0,98];
the reference period is [one] year.
However in some cases the character of the action makes another reference period more appropriate. In
addition, design values for other variables within the action model may have to be chosen, which may
influence the probability of being exceeded for the resulting total action.
(9) Actions caused by water should normally be based on water levels and include a geometrical parameter
to allow for fluctuation of water level. Tides, currents and waves should be taken into account where
relevant.
(10) For accidental actions the representative value is generally a characteristic value A
k
corresponding to
a specified value.
(11) Values of A
k
for explosion and for some impacts are given in ENV 1991-2-7.
(12) For accidental actions arising from fire, information is given in ENV 1991-2-2.
(13) Values of A
Ed
for seismic actions are given in ENV 1998-1.
(14) For accidental actions on bridges arising from the traffic, characteristic values to be used as design
values are given in ENV 1991-3.
(15) For multi-component actions [see 4.1(7)] the characteristic action is represented by groups of values,
to be considered alternatively in design calculations.
4.3 Other representative values of variable and accidental actions
(1)P In the most common cases the other representative values of a variable action are:
the combination value generally represented as a product: ?
0
Q
k
;
the frequent value generally represented as a product: ?
1
Q
k
;
the quasi-permanent value generally represented as a product: ?
2
Q
k
.
(2)P Combination values are associated with the use of combinations of actions, to take account of a reduced
probability of simultaneous occurrence of the most unfavourable values of several independent actions.
NOTE For methods for determining ?
0
see Annex A
(3)P The frequent value is determined such that:
the total time, within a chosen period of time, during which it is exceeded for a specified part, or
the frequency with which it is exceeded,
is limited to a given value.
(4) The part of the chosen period of time or the frequency, mentioned in 4.3(3) should be chosen with due
regard to the type of construction works considered and the purpose of the calculations. Unless other values
are specified the part may be chosen to be 0,05 or the frequency to be 300 per year for ordinary buildings.
(5)P The quasi-permanent value is so determined that the total time, within a chosen period of time, during
which it is exceeded is a considerable part of the chosen period of time.
(6) The part of the chosen period of time, mentioned in 4.3(5), may be chosen to be 0,5. The
quasi-permanent value may also be determined as the value averaged over the chosen period of time.
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
22
BSI 04-2000
(7)P These representative values and the characteristic value are used to define the design values of the
actions and the combinations of actions as explained in section 9. The combination values are used for the
verification of ultimate limit states and irreversible serviceability limit states. The frequent values and
quasi-permanent values are used for the verification of ultimate limit states involving accidental actions
and for the verification of reversible serviceability limit states. The quasi-permanent values are also used
for the calculation of long term effects of serviceability limit states. More detailed rules concerning the use
of representative values are given, for example, in ENVs 1992 to 1999.
(8) For some structures or some actions other representative values or other types of description of actions
may be required, e.g. the fatigue load and the number of cycles when fatigue is considered.
NOTE Further information concerning the specification and combination of actions is given in Annex A and other parts of
ENV 1991.
4.4 Environmental influences
The environmental influences which may affect the durability of the structure shall be considered in
the choice of structural materials, their specification, the structural concept and detailed design.
The ENVs 1992 to 1999 specify the relevant measures.
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
BSI 04-2000 23
Section 5. Material properties
(1)P Properties of materials (including soil and rock) or products are represented by characteristic values
which correspond to the value of the property having a prescribed probability of not being attained in a
hypothetical unlimited test series. They generally correspond for a particular property to a specified fractile
of the assumed statistical distribution of the property of the material in the structure.
(2) Unless otherwise stated in ENVs 1992 to 1999, the characteristic values should be defined as the 5 %
fractile for strength parameters and as the mean value for stiffness parameters.
NOTE For operational rules, see Annex D; for fatigue, information is given in Annex B
(3)P Material property values shall normally be determined from standardized tests performed under
specified conditions. A conversion factor shall be applied where it is necessary to convert the test results
into values which can be assumed to represent the behaviour of the material in the structure or the ground
(see also ENVs 1992 to 1999).
(4) A material strength may have two characteristic values, an upper and a lower. In most cases only the
lower value will need to be considered. In some cases, different values may be adopted depending on the
type of problem considered. Where an upper estimate of strength is required (e.g. for the tensile strength
of concrete for the calculation of the effects of indirect actions) a nominal upper value of the strength should
normally be taken into account.
(5) Where there is a lack of information on the statistical distribution of the property a nominal value may
be used; where the limit state equation is not significantly sensitive to its variability a mean value may be
considered as the characteristic value.
(6) Values of material properties are given in ENVs 1992 to 1999.
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
24
BSI 04-2000
Section 6. Geometrical data
(1)P Geometrical data are represented by their characteristic values, or in the case of imperfections directly
by their design values.
(2) The characteristic values usually correspond to dimensions specified in the design.
(3) Where relevant, values of geometrical quantities may correspond to some prescribed fractile of the
statistical distribution.
(4)P Tolerances for connected parts which are made from different materials shall be mutually compatible.
Imperfections which have to be taken into account in the design of structural members are given in
ENVs 1992 to 1999.
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
BSI 04-2000 25
Section 7. Modelling for structural analysis and
resistance
7.1 General
(1)P Calculations shall be performed using appropriate design models involving relevant variables.
The models shall be appropriate for predicting the structural behaviour and the limit states considered.
(2) Design models should normally be based on established engineering theory and practice, verified
experimentally if necessary.
NOTE Further information is given in Annex C and Annex D.
7.2 Modelling in the case of static actions
(1) The modelling for static actions should normally be based on an appropriate choice of the force
deformation relationships of the members and their connections.
(2) Effects of displacements and deformations should be considered in the context of ultimate limit state
verifications (including static equilibrium) if they result in an increase of the effects of actions by more
than 10 %.
(3) In general the structural analysis models for serviceability limit states and fatigue may be linear.
7.3 Modelling in the case of dynamic actions
(1) When dynamic actions may be considered as quasi-static, the dynamic parts are considered either by
including them in the static values or by applying equivalent dynamic amplification factors to the static
actions. For some equivalent dynamic amplification factors, the natural frequencies have to be determined.
(2) In some cases (e.g. for cross wind vibrations or seismic actions) the actions may be defined by provisions
for a modal analysis based on a linear material and geometric behaviour. For regular structures, where
only the fundamental mode is relevant, an explicit modal analysis may be substituted by an analysis with
equivalent static actions, depending on mode shape, natural frequency and damping.
(3) In some cases the dynamic actions may be expressed in terms of time histories or in the frequency
domain, for which the structural response may be determined by appropriate methods.
NOTE When dynamic actions may cause vibrations that may infringe serviceability limit states guidance for assessing these limit
states is given in Annex C, together with the models of some actions.
7.4 Modelling for fire actions
(1)P The structural analysis for fire design shall be performed using appropriate models for the fire
situation, involving thermal and mechanical actions, and for the structural behaviour at elevated
temperatures. The analysis may be assisted by testing.
(2) For fire design situations, see ENV 1991-2 which covers thermal actions in terms of:
nominal (standard) fire exposures; and
parametric fire exposure;
and specific rules for mechanical actions.
(3) The structural behaviour at elevated temperatures, should be assessed in accordance with
ENVs 1992 to 1996 and ENV 1999, which give thermal and structural models for analysis.
Where relevant to the specific material and the method of assessment:
thermal models may be based on the assumption of a uniform temperature within cross-sections or
may result in thermal gradients within cross-sections and along members;
structural models may be confined to an analysis of members or may account for the interaction
between members in fire exposure. The behaviour of materials or sections at elevated temperatures may
be modelled as linear-elastic, rigid-plastic or non-linear.
(4) Where tabulated data are given in ENVs 1992 to 1996 and ENV 1999, these data are mainly obtained
from test results or numerical simulation based only on the action as described by the standard fire
exposure.
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
26
BSI 04-2000
Section 8. Design assisted by testing
8.1 General
(1)P Where calculation rules or material properties given in ENVs 1991 to 1999 are not sufficient or where
economy may result from tests on prototypes, part of the design procedure may be performed on the basis
of tests.
NOTE Some of the clauses in this section may also be helpful in cases where the performance of an existing structure is to be
investigated.
(2)P Tests shall be set up and evaluated in such a way that the structure has the same level of reliability
with respect to all possible limit states and design situations as achieved by design based on calculation
procedures specified in ENVs 1991 to 1999, including this Part of ENV 1991.
(3) Sampling of test specimens and conditions during testing should be representative.
(4) Where ENVs 1991 to 1999 include implicit safety provisions related to comparable situations, these
provisions shall be taken into account in assessing the test results and may give rise to corrections. An
example is the effect of tensile strength in the bending resistance of concrete beams, which is normally
neglected during design.
8.2 Types of tests
(1) The following test types are distinguished:
a) tests to establish directly the ultimate resistance or serviceability properties of structural parts
e.g. fire tests;
b) tests to obtain specific material properties, e.g. ground testing in situ or in the laboratory, testing of
new materials;
c) tests to reduce uncertainties in parameters in load or resistance models, e.g. wind tunnel testing,
testing of full size prototypes, testing of scale models;
d) control tests to check the quality of the delivered products or the consistency of the production
characteristics, e.g. concrete cube testing;
e) tests during execution in order to take account of actual conditions experienced e.g. post-tensioning,
soil conditions;
f) control tests to check the behaviour of the actual structure or structural elements after completion,
e.g. proof loading for the ultimate or serviceability limit states.
(2) For test types a), b) and c), the test results may be available at the time of design; in those cases the
design values can be derived from the tests. For test types d), e) and f) the test results may not be available
at the time of design; in these cases the design values correspond to that part of the production that is
expected to meet the acceptance criteria at a later stage.
8.3 Derivation of design values
(1)P The derivation of the design values for a material property, a model parameter or a resistance value
from tests can be performed in either of the following two ways:
a) by assessing a characteristic value, which is divided by a partial factor and possibly multiplied by an
explicit conversion factor;
b) by direct determination of the design value, implicitly or explicitly accounting for the conversion
aspects and the total reliability required.
(2) In general method a) should be used. The derivation of a characteristic value from tests should be
performed taking account of:
1) the scatter of test data;
2) statistical uncertainty resulting from a limited number of tests;
3) implicit or explicit conversion factors resulting from influences not sufficiently covered by the tests
such as:
i) time and duration effects, not taken care of in the tests;
ii) scale, volumes and length effects;
iii) deviating environmental, loading and boundary conditions;
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
BSI 04-2000 27
iv) the way that safety factors as partial factors or additive elements are applied to get design values
(see 9.3).
The partial factor used in method a) should be chosen in such a way that there is sufficient similarity
between the tests under consideration and the usual application field of the partial factor used in numerical
verifications. (see also 3.4).
(3) When for special cases method b) is used, the determination of the design values should be carried out
by considering:
the relevant limit states;
the required level of reliability;
the statistical and model uncertainties;
the compatibility with the assumptions for the action side;
the classification of design working life of the relevant structure according to Section 2;
prior knowledge from similar cases or calculations.
(4) Further information may be found in ENVs 1992 to 1999.
NOTE see also Annex A and Annex D.
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
28
BSI 04-2000
Section 9. Verification by the partial factor method
9.1 General
(1)P In ENVs 1992 to 1999 the reliability according to the limit state concept is achieved by application of
the partial factor method. In the partial factor method, it is verified that, in all relevant design situations,
the limit states are not exceeded when design values for actions, material properties and geometrical data
are used in the design models.
(2)P In particular, it shall be verified that:
a) the effects of design actions do not exceed the design resistance of the structure at the ultimate limit
state; and
b) the effects of design actions do not exceed the performance criteria for the serviceability limit state.
Other verifications may also need to be considered for particular structures e.g. fatigue. Details are
presented in the relevant parts of ENV 1991 and in ENVs 1992 to 1999.
NOTE see also Annex A and Annex B.
(3)P The selected design situations shall be considered and critical load cases identified. For each critical
load case, the design values of the effects of actions in combination shall be determined.
(4) A load case identifies compatible load arrangements, sets of deformations and imperfections which
should be considered simultaneously for a particular verification.
(5) Rules for the combination of independent actions in design situations are given in this section. Actions
which cannot occur simultaneously, for example, due to physical reasons, should not be considered together
in combination.
(6) A load arrangement identifies the position, magnitude and direction of a free action. Rules for different
arrangements within a single action are given in ENVs 1991-2, 1991-3 and 1991-4.
(7) Possible deviations from the assumed directions or positions of actions should be considered.
(8) The design values used for different limit states may be different and are specified in this section.
9.2 Limitations and simplifications
(1) Application rules in ENV 1991-1 are limited to ultimate and serviceability limit states for structures
subject to static loading. This includes cases where the dynamic effects are assessed using equivalent
quasi-static loads and dynamic amplification factors, e.g. wind. Modifications for non-linear analysis and
fatigue are given in other parts of ENV 1991 and in ENVs 1992 to 1999.
(2) Simplified verification based on the limit state concept may be used:
by considering only limit states and load combinations which from experience or special criteria are
known to be potentially critical for the design;
by using the simplified verification for ultimate limit states and/or serviceability limit states as
specified for buildings in 9.4.5 and 9.5.5;
by specifying particular detailing rules and/or provisions to meet the safety and serviceability
requirements without calculation.
NOTE For those cases where ENVs 1991 to 1999 do not give adequate rules for the verification, for instance for new materials,
special structures, unusual limit states, guidance is given in Annex A. For those cases where the Eurocodes give adequate rules,
Annex A can be considered as background information.
9.3 Design values
9.3.1 Design values of actions
(1)P The design value F
d
of an action is expressed in general terms as:
where:
F
d
= *
F
F
rep
(9.1)
*
F
is the partial factor for the action considered taking account of:
the possibility of unfavourable deviations of the actions;
the possibility of inaccurate modelling of the actions;
uncertainties in the assessment of effects of actions.
F
rep
is the representative value of the action.
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
BSI 04-2000 29
(2) Depending on the type of verification and combination procedures, design values for particular actions
are expressed as follows:
(3)P Where distinction has to be made between favourable and unfavourable effects of permanent actions,
two different partial factors shall be used.
(4) For seismic actions the design value may depend on the structural behaviour characteristics
(see ENV 1998).
9.3.2 Design values of the effects of actions
(1) The effects of actions (E) are responses (for example internal forces and moments, stresses, strains and
displacements) of the structure to the actions. For a specific load case the design value of the effect of
actions (E
d
) is determined from the design values of the actions, geometrical data and material properties
when relevant:
where:
F
d1
, ..., a
d1
, ... and X
d1
, ... are chosen according to 9.3.1, 9.3.3 and 9.3.4 respectively.
(2) In some cases, in particular for non-linear analysis, the effect of the uncertainties in the models used in
the calculations should be considered explicitly. This may lead to the application of a coefficient of model
uncertainty, *
Sd
applied either to the actions or to the action effects, whichever is the more conservative.
The factor *
Sd
may refer to uncertainties in the action model and/or the action effect model.
(3) For non-linear analysis, i.e. when the effect is not proportional to the action, the following simplified
rules may be considered in the case of a single predominant action.
a) When the effect increases more than the action, the partial factor is applied to the representative
value of the action.
b) When the effect increases less than the action, the partial factor is applied to the action effect of the
representative value of the action.
In other cases more refined methods are necessary which are defined in the relevant Eurocodes (e.g. for
prestressed structures).
9.3.3 Design values of material properties
(1)P The design value X
d
of a material or product property is generally defined as:
where:
In some cases the conversion is implicitly taken into account by the characteristic value itself, as indicated
by the definition of ), or by *
M
.
9.3.4 Design values of geometrical data
(1)P Design values of geometrical data are generally represented by the nominal values:
G
d
= *
G
G
k
or G
k
Q
d
= *
Q
Q
k
, *
Q
?
0
Q
k
, ?
1
Q
k
, ?
2
Q
k
or Q
k
(9.2)
A
d
= *
A
A
k
or A
d
P
d
= *
P
P
k
or P
k
A
Ed
= A
Ed
E
d
= E(F
d1
, F
d2
, ... a
d1
, a
d2
, ... X
d1
, X
d2
, ...) (9.3)
X
d
= )X
k
/ *
M
or X
d
= X
k
/ *
M
(9.4)
*
M
is the partial factor for the material or product property, given in ENVs 1992 to 1999, which covers:
unfavourable deviations from the characteristic values;
inaccuracies in the conversion factors; and
uncertainties in the geometric properties and the resistance model.
) is the conversion factor taking into account the effect of the duration of the load, volume and scale
effects, effects of moisture and temperature and so on.
a
d
= a
nom
(9.5)
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
30
BSI 04-2000
Where necessary ENVs 1992 to 1999 may give further specifications.
(2)P In some cases when deviations in the geometrical data have a significant effect on the reliability of a
structure, the geometrical design values are defined by:
where %
a
takes account of the possibility of unfavourable deviations from the characteristic values.
%
a
is only introduced where the influence of deviations is critical, e.g. imperfections in buckling analysis.
Values of %
a
are given in ENVs 1992 to 1999.
9.3.5 Design resistance
(1)P Design values for the material properties, geometrical data and effects of actions, when relevant, shall
be used to determine the design resistance R
d
from:
where a
d1
, ... is defined in 9.3.4 and X
d1
, ... in 9.3.3.
(2) Operational verification formulae, based on the principle of expression (9.7), may have one of the
following forms:
where:
*
R
is a partial factor for the resistance;
*
m
is a material factor;
*
rd
covers uncertainties in the resistance model and in the geometrical properties.
NOTE For further information, see Annex A
(3) The design resistance may also be obtained directly from the characteristic value of a product
resistance, without explicit determination of design values for individual basic variables, from:
This is applicable for steel members, piles, etc. and is often used in connection with design by testing.
9.4 Ultimate limit states
9.4.1 Verifications of static equilibrium and strength
(1)P When considering a limit state of static equilibrium or of gross displacement of the structure as a rigid
body, it shall be verified that:
where:
In some cases it may be necessary to replace expression (9.8) by an interaction formula.
(2)P When considering a limit state of rupture or excessive deformation of a section, member or connection,
it shall be verified that:
where:
a
d
= a
nom
+ %
a
(9.6)
R
d
= R(a
d1
, a
d2
, ... X
d1
, X
d2
, ...) (9.7)
R
d
= R { X
k
/*
M
, a
nom
} (9.7a)
R
d
= R { X
k
, a
nom
}/*
R
(9.7b)
R
d
= R { X
k
/*
m
, a
nom
}/*
rd
(9.7c)
R
d
= R
k
/*
R
(9.7d)
E
d ,dst
k E
d ,stb
(9.8)
E
d ,dst
is the design value of the effect of destabilizing actions;
E
d ,stb
is the design value of the effect of stabilizing actions.
E
d
k R
d
(9.9)
E
d
is the design value of the effect of actions such as internal force, moment or a vector representing
several internal forces or moments;
R
d
is the corresponding design resistance, associating all structural properties with the respective
design values.
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
BSI 04-2000 31
In some cases it may be necessary to replace expression (9.9) by an interaction formula. The required load
cases are identified as described in 9.1.
9.4.2 Combination of actions
(1)P For each critical load case, the design values of the effects of actions (E
d
) should be determined by
combining the values of actions which occur simultaneously, as follows:
a) Persistent and transient situations: Design values of the dominant variable actions and the
combination design values of other actions.
b) Accidental situations: Design values of permanent actions together with the frequent value of the
dominant variable action and the quasi-permanent values of other variable actions and the design value
of one accidental action.
c) Seismic situations: Characteristic values of the permanent actions together with the quasipermanent
values of the other variable actions and the design value of the seismic actions.
(2) When the dominant action is not obvious, each variable action should be considered in turn as the
dominant action.
(3) The above combination process is represented in Table 9.1.
Table 9.1 Design values of actions for use in the combination of actions
Symbolically the combinations may be represented as follows
a) persistent and transient design situations for ultimate limit states verification other than those
relating to fatigue
NOTE This combination rule is an amalgamation of two separate load combinations:
[K] is a reduction factor for *
Gj
within the range 0.85 and 1. From the expressions (9.10a) and (9.10b) the more favourable may
be applied instead of expression (9.10) under conditions defined by the relevant National Application Document.
b) Combinations for accidental design situations
c) Combination for the seismic design situation
where:
Design situation Permanent actions G
d
Single variable actions Q
d
Accidental actions or
seismic actions A
d
Dominant Others
Persistent and transient *
G
G
k
(*
P
P
k
) *
Q1
Q
k1
*
Qi
?
Oi
Q
id
Accidental *
GA
G
k
(*
PA
P
k
) ?
11
Q
k1
?
2i
Q
id
*
A
A
k
or A
d
Seismic G
k
?
2i
Q
id
*
A
A
Ed
(9.10)
(9.10a)
(9.10b)
(9.11)
(9.12)
+ implies to be combined with;
implies the combined effect of;
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
32
BSI 04-2000
(4) Combinations for accidental design situations either involve an explicit accidental action A (e.g. fire or
impact) or refer to a situation after an accidental event (A = 0). For fire situations, apart from the
temperature effect on the material properties, A
d
refers to the design value of the indirect thermal action.
(5) Expressions (9.10) to (9.11) may refer to either actions or action effects; for non-linear analysis,
see 9.3.2(3).
(6) Where components of a vectorial force are partially correlated, the factors to any favourable component
may be reduced by [20 %].
(7) Imposed deformations should be considered where relevant.
(8) In some cases expressions (9.10) to (9.12) need modification; detailed rules are given in the relevant
parts of ENVs 1991 to 1999.
9.4.3 Partial factors
(1)P In the relevant load cases, those permanent actions that increase the effect of the variable actions
(i.e. produce unfavourable effects) shall be represented by their upper design values, those that decrease
the effect of the variable actions (i.e. produce favourable effects) by their lower design values.
(2)P Where the results of a verification may be very sensitive to variations of the magnitude of a permanent
action from place to place in the structure, the unfavourable and the favourable parts of this action shall
be considered as individual actions. This applies in particular to the verification of static equilibrium.
(3) For building structures, the partial factors for ultimate limit states in the persistent, transient and
accidental design situations are given in Table 9.2. The values have been based on theoretical
considerations, experience and back calculations on existing designs.
NOTE These values may be used for the design of silos covered in ENV-1991-4.
G
kj
is the characteristic value of permanent actions;
P
k
is the characteristic value of a prestressing action;
Q
k1
is the characteristic value of the variable action i;
Q
id
is the characteristic value of the variable actions;
A
d
is the design value of the accidental action;
A
Ed
is the design value of seismic action;
*
Gj
is the partial factor for permanent action j;
*
GAj
is the same as *
Gj
, but for accidental design situations;
*
PA
is the same as *
P
, but for accidental actions;
*
P
is the partial factor for prestressing actions;
*
Qi
is the partial factor for variable action i;
*
i
is the importance factor (see ENV 1998);
? are combination coefficients (see 4.3).
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
BSI 04-2000 33
Table 9.2 Partial factors: ultimate limit states for buildings
Case
a
Action Symbol
Situations
P/T A
Case A
Loss of static equilibrium; strength
of structural material or ground
insignificant (see 9.4.1)
Permanent actions: self weight of
structural and non-structural
components, permanent actions
caused by ground, ground-water
and free water
unfavourable *
Gsup
d
[1,10]
b
[1,00]
favourable *
Ginf
d
[0,90]
b
[1,00]
Variable actions
unfavourable *
Q
[1,50] [1,00]
Accidental actions *
A
[1,00]
Case B
e
Failure of structure or structural
elements, including those of the
footing, piles, basement walls etc.,
governed by strength of structural
material (see 9.4.1)
Permanent actions
f

(see above)
unfavourable *
Gsup
d
[1,35]
c
[1,00]
favourable *
Ginf
d
[1,00]
c
[1,00]
Variable actions
unfavourable *
Q
[1,50] [1,00]
Accidental actions *
A
[1,00]
Case C
e
Failure in the ground
Permanent actions
(see above)
unfavourable *
Gsup
d
[1,00] [1,00]
favourable *
Ginf
d
[1,00] [1,00]
Variable actions
unfavourable *
Q
[1,30] [1,00]
Accidental actions *
A
[1,00]
P: Persistent situation T: Transient situation A: Accidental situation
a
The design should be verified for each case A, B and C separately as relevant.
b
In this verification the characteristic value of the unfavourable part of the permanent action is multiplied by the factor [1,1] and
the favourable part by the factor [0,9]. More refined rules are given in ENV 1993 and ENV 1994.
c
In the verification the characteristic values of all permanent actions from one source are multiplied by [1,35] if the total resulting
action effect is unfavourable and by [1,0] if the total resulting action effect is favourable.
d
In cases when the limit state is very sensitive to variations of permanent actions, the upper and lower characteristic values of
these actions should be taken according to 4.2(3).
e
For cases B and C the design ground properties may be different, see ENV 1997-1-1
f
Instead of using *
G
(1,35) and *
Q
(1,50) for lateral earth pressure actions the design ground properties may be introduced in
accordance with ENV 1997 and a model factor *
Sd
is applied.
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
34
BSI 04-2000
9.4.4 ? factors
(1) ? factors for buildings are given in Table 9.3. For other applications see relevant parts of ENV 1991.
Table 9.3 ? factors for buildings
9.4.5 Simplified verification for building structures
(1) The process for the persistent and transient situations described in 9.4.2 may be simplified by
considering the most unfavourable for the following combinations:
a) Design situations with only one variable action Q
k1

b) Design situations with two or more variable actions Q
k ,i

In this case the effect of actions should also be verified for the dominant variable actions using
expression (9.13).
(2) The *
G
values are given in Table 9.2.
9.4.6 Partial safety factors for materials
Partial safety factors for properties of materials and products are given in ENVs 1992 to 1999.
9.5 Serviceability limit states
9.5.1 Verifications of serviceability
(1)P It shall be verified that:
where:
Action ?
0
?
1
?
2
Imposed loads in buildings
a
category A: domestic, residential [0,7] [0,5] [0,3]
category B: offices [0,7] [0,5] [0,3]
category C: congregation areas [0,7] [0,7] [0,6]
category D: shopping [0,7] [0,7] [0,6]
category E: storage [1,0] [0,9] [0,8]
Traffic loads in buildings
category F: vehicle weight k 30 kN [0,7] [0,7] [0,6]
category G: 30 kN < vehicle weight k 160 kN [0,7] [0,5] [0,3]
category H: roofs [0] [0] [0]
Snow loads on buildings [0,6]
b
[0,2]
b
[0]
b
Wind loads on buildings [0,6]
b
[0,5]
b
[0]
b
Temperature (non-fire) in buildings
c
[0,6]
b
[0,5]
b
[0]
b
a
For combination of imposed loads in multistorey buildings, see ENV 1991-2-1.
b
Modification for different geographical regions may be required.
c
See ENV 1991-2-5.
(9.13)
(9.14)
E
d
k C
d
(9.15)
C
d
is a nominal value or a function of certain design properties of materials related to the design
effects of actions considered; and
E
d
is the design value of the action effect (e.g. displacement, acceleration), determined on the basis
of one of the combinations defined in 9.5.2.
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
BSI 04-2000 35
NOTE Guidance for C
d
may be found in ENV 1992 to 1999.
9.5.2 Combination of actions
(1) The combination of actions to be considered for serviceability limit states depends on the nature of the
effect of actions being checked, e.g. irreversible, reversible or long term. Three combinations designated by
the representative value of the dominant action are given in Table 9.4.
Table 9.4 Design values of actions for use in the
combination of actions
(2) Three combinations of actions for serviceability limit states are defined symbolically by the following
expressions:
a) Characteristic (rare) combination
b) Frequent combination
c) Quasi-permanent combination
where the notation is as given in 1.6 and 9.4.2.
(3) Loads due to imposed deformations should be considered where relevant.
(4) In some cases expressions (9.16) to (9.18) may require a modification; detailed rules are given in the
relevant Parts of ENVs 1991 to 1999.
9.5.3 Partial factors
The partial factors for serviceability limit states are equal to [1,0] except where specified otherwise, e.g. in
ENVs 1992 to 1999.
9.5.4 ? factors
Values of ? factors are given in Table 9.3.
9.5.5 Simplified verification for building structures
(1) For building structures the characteristic (rare) combination may be simplified to the following
expressions, which may also be used as a substitute for the frequent combination.
a) Design situations with only one variable action, Q
k1

Combination Permanent actions G
d
Variable actions Q
d
Dominant Others
Characteristic (rare) G
k
(P
k
) Q
k1
?
0i
Q
ki
Frequent G
k
(P
k
) ?
11
Q
k1
?
2i
Q
ki
Quasi-permanent G
k
(P
k
) ?
21
Q
k1
?
2i
Q
ki
NOTE For serviceability limit states, the partial factors (serviceability) *
G
and *
Q
are taken as 1,0
except where specified otherwise.
(9.16)
(9.17)
(9.18)
(9.19)
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
36
BSI 04-2000
b) Design situations with two or more variable actions, Q
k1

In this case the effect of actions should also be verified for the dominant variable action using
expression (9.19).
(2) Where simplified prescriptive rules are given for serviceability limit states, detailed calculations using
combinations of actions are not required.
9.5.6 Partial factors for materials
Partial factors for the properties of materials and products are given in ENVs 1992 to 1999.
(9.20)
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
BSI 04-2000 37
Annex A (informative)
Partial factor design
A.1 General
(1) This annex gives information and theoretical background concerning the partial factor method as
described in section 9. This annex is also an introduction to Annex D. The information in these annexes
may be used if the verification rules of ENVs 1991-1999 are not considered adequate for the case
considered.
(2) In the partial factor method it is verified that all relevant limit states are not exceeded, given design
values for actions, resistances and geometrical data. The design values are the products or quotients of the
characteristic values and the appropriate partial factors and ? values, as indicated in 9.3 to 9.5. In
general, partial factors are intended to take account of:
unfavourable deviations from the representative values;
inaccuracies in the action models and structural models;
inaccuracies in the conversion factors.
(3) The value of the partial factors should depend on the degree of uncertainty in the actions, resistances,
geometrical quantities and models, and on the type of construction works and the type of limit state.
(4) In principle there are two ways to determine numerical values for partial factors:
a) on the basis of calibration to a long and successful history of building tradition; for most of the factors
proposed in the currently available Eurocodes this is the leading principle;
b) on the basis of the statistical evaluation of experimental data and field observations; this should be
done within the framework of a probabilistic reliability theory.
(5) In practice, the two methods described in A.1(4) can also be used in combination. In particular, a mere
statistical (probabilistic) approach usually fails from a lack of sufficient data. Some reference to traditional
design methods should always be made. Where there has been a long and successful building tradition, it
is of great value to obtain a rational understanding of that success. Understanding may justify the
reduction of some factors for specified conditions, which in their turn may lead to economy. From this point
of view, the statistical methods should be considered as giving added value to the more traditional
approach.
A.2 Overview of reliability methods
(1) Figure A.1 presents an overview of the various methods for reliability verification and the interactions
between them. The probabilistic verification procedures can be subdivided into two main classes: exact
methods and first order reliability methods (FORM), sometimes referred to as level III and level II methods
respectively. In both methods the measures of reliability are failure probabilities P
1
for the failure modes
under consideration and for some appropriate reference period. These values are calculated and compared
with some preset target value P
0
. If the failure probability is larger than the target, the structure is
considered to be unsafe.
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
38
BSI 04-2000
(2) In the level II procedures one generally works with an alternative measure of safety, the so-called
reliability index ", which is related to P
1
by:
where 9 is the distribution function of the normal distribution.
Although fully equivalent to the failure probability itself, the use of the reliability index emphasizes the
formal and notional nature of the reliability analysis. The relationship between " and P
1
is presented in
Table A.1.
Table A.1 Relation between " and P
1
(3) According to Figure A.1, the safety elements of the partial factor method (level I) can be obtained in
three ways:
a) from calibration to historical and empirical design methods;
b) from calibration to probabilistic methods;
c) as a simplification of FORM via the (calibrated) design value method as described in A.3.
The present generation of Eurocodes has been primarily based on method a), with amendments based
on c) or equivalent methods, mainly in the field of design assisted by testing.
(4) Indicative target values for " in various design situations are given in Table A.2. Values are given for
the design working life (see Table 2.1 of ENV 1999-1) and for one year. Values for one year might be
relevant for transient design situations and temporary structures where human safety is of great
importance.
(5) The values in Table A.2 are intended as appropriate for most cases. For reasons related to the type
and consequences of failure and economy of building, it may be appropriate to use higher or lower values
(see 2.2). A class difference in reliability level is usually associated with differences in " values in the order
of 0,5 to 1,0. A difference of reliability level may be desired for a total building, some specific components
or some specific hazards.
NOTE 1 A given reliability level may lead to different partial factors for various material properties and loads, depending on their
variability and influence, see A.3 and A.4. This should not be confused with reliability differentiation.
NOTE 2 Choosing a different target reliability index is not the only possible measure for reliability differentiation; other measures
are related to the accuracy of calculation, the degree of quality assurance and the stringency of detailing rules.
Figure A.1 Overview of reliability methods
P
1
= 9(-") (A.1)
P
1
10
1
10
2
10
3
10
4
10
5
10
6
10
7
" 1,3 2,3 3,1 3,7 4,2 4,7 5,2
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
BSI 04-2000 39
(6) The values in Table A.2 should be considered as reasonable minimum requirements, following from
calibration calculations to design codes in various countries. In these calibrations lognormal or Weibull
distributions for resistance parameters and model uncertainties were usually used. Normal distributions
were usually taken for self weight and extreme value distributions for variable loads. It should be noted,
however, that these calibrations showed a wide scatter, depending on the code at hand, the type of
structural component, and the quantification of the various uncertainties.
(7) The value of 3.8 for the ultimate limit state target reliability index is in particular, accepted for many
applications mainly relating to resistance. This, however, does not mean that standard design according to
the Eurocodes automatically would lead to "-values equal or close to this target. In fact, up to now, the
present generation of Eurocodes have not been wholly evaluated in this way. Such an evaluation is not very
straightforward as serviceability, durability, round off effects or multimodal distribution effects may
disturb the picture in many cases. Additionally design rules in codes may also include implicit safety
differentiations depending on the type of failure, especially ductile or brittle behaviour.
(8) Finally, it should be stressed that a "-value and the corresponding failure probability are formal or
notional numbers, intended primarily as a tool for developing consistent design rules, rather than giving a
description of the structural failure frequency.
Table A.2 Indicative values for the target reliability index "
A.3 Reliability verification using design values
(1) In the design value method (method Ib in Figure A.1), design values are defined for all variables that
should be considered as uncertain (basic variables). The design is considered to be sufficient if the limit
states are not reached when design values are used in the models. In symbolic notation (see section 9):
where
Note that expression (A.2) is partly symbolic and that sometimes a more general formulation is necessary.
(2) The set of design values for the design point corresponds to the point on the failure surface having the
highest probability of occurrence (see Figure A.2). In this way the design value method is related to the
probabilistic level II method [see A.2(1)].
(3) The design value of action effects E
d
and the resistances R
d
are defined such that the probability of
having a more unfavourable value is as follows:
Limit state
Target reliability index
(design working life)
Target reliability index (one year)
Ultimate 3,8 4,7
Fatigue 1,5 to 3,8
a

Serviceability
(irreversible)
1,5 3,0
a
Depends on degree of inspectability, reparability and damage tolerance.
E
d
< R
d
(A.2)
E
d
= E {F
d1
, F
d2
, ... a
d1
, a
d2
, ... F
d1
, F
d2
, ...}
R
d
= R {f
d1
, f
d2
, ... a
d1
, a
d2
, ... F
d1
, F
d2
, ...}
E is the action effect;
R is the resistance;
F is action;
f is material property;
a is geometrical property;
F model uncertainty.
P (E > E
d
) = 9 (+
E
") = 9 ( 0,7") (A.3a)
P (R < R
d
) = 9 (
R
") = 9 ( 0,8") (A.3b)
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
40
BSI 04-2000
where:
! is the FORM weight factor ( 1 k ! k + 1);
" is the target value for the reliability index (see Table A.2).
For a load ! is negative; for a resistance parameter ! is positive.
(4) The essence of the method is the setting of the !
E
and !
R
values to 0,7 and + 0,8 respectively. The
validity range for these values is limited for the case " = 3,8 (accepting a maximum deviation of 0.5) to the
ratios:
0,16 < B
E
/B
R
< 7,6
Outside this range it is recommended to use ! = 1.0 for the variable having the largest value of B.
(5) When the load or resistance model contains several basic variables (other loads, conversion factors,
more materials) expressions (A.3a) and (A.3b) only hold for the dominating variables. For non-dominating
variables:
For " = 3,8 these values correspond approximately to the 0,90 and 0,10 fractiles respectively.
(6) Table A.3 gives expressions for calculating the design values for given ! and ".
Table A.3 Design values for various distribution functions
A.4 Reliability verification formats in Eurocodes
(1) In ENVs 1991 to 1999 design values X
d
and F
d
are not introduced directly. Basic variables are first
introduced by their representative values X
k
and F
k
, which can be defined as:
values with a prescribed or intended probability of being exceeded, e.g. loads and material properties;
nominal values, e.g. geometrical properties;
P {E > E
d
} = 9 ( 0,4 0,7 ") = 9 ( 0,28 ") (A.4a)
P {R < R
d
} = 9 ( 0,4 0,8 ") = 9 ( 0,32 ") (A.4b)
Distribution Design values Remarks
Normal 4 !"B 4 is the mean,
B is the standard deviation
Lognormal 4 exp( !"V) for V = B/4 < 0,2
Gumbel u a
1
In { In9( !")} u = 4 0,577/a, a = ;/(B6)
Figure A.2 Design point definition according to first order reliability methods (FORM)
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
BSI 04-2000 41
values calibrated to reach the aimed reliability, e.g. coefficients and model factors.
Additionally there is a set of partial safety factors and load combination factors.
(2) The design values for actions F, material properties X and geometrical properties follow from:
The index k denotes characteristic values.
(3) The design values for model uncertainties normally enter the equations by partial factors *
Sd
and *
Rd
on
the total model. It follows that:
In this model:
*
f
takes account of:
the possibility of unfavourable deviations of the action values from the representative values.
*
m
takes account of:
the possibility of unfavourable deviations of the material properties from the characteristic values;
the systematic part of the conversion factor [if relevant, see also 8.3(1)];
uncertainties of the conversion factor.
%
a
takes account of:
the possibility of unfavourable deviations of the geometrical data from the characteristic (specified)
values governed by the tolerance specifications;
the importance of variations;
the cumulative effect of a simultaneous occurrence of several geometrical deviations.
*
Rd
takes account of:
the uncertainties of the resistance model if these uncertainties are not covered by the model itself.
*
Sd
takes account of the uncertainties:
in the action model;
in the action effect model.
? takes account of reductions in design values for loads, in particular:
the combination value ?
0
*
F
F
k
is determined in such a way that the probability of combined action
effect values being exceeded is approximately the same as when a single variable action only is
present. Within the context of a design value approach (A.3), operational formulae are presented in
Table A.4 for the case of two fluctuating loads.
The frequent value of a variable action ?
1
F
k
corresponds to the value which is exceeded either 5 %
of the time or 300 times per year; the highest value should be chosen.
The quasi-permanent value ?
2
F
k
corresponds to the time average or to the value with a probability
of being exceeded of 50 %.
(4) The procedure described by expressions (A.8) and (A.9) is theoretically perfect but cumbersome from a
practical point of view. Therefore the following simplifications are made.
a) On the loading side (for a single loading):
Providing E is proportional to F, a, and a model uncertainty F, i.e.
E FaF,
the value of *
F
may follow from [see expression (A.8) and (A.10)]:
F
d
= *
f
F
k
or F
d
= *
f
?F
k
(? can be ?
0
, ?
1
or ?
2
) (A.5)
X
d
= X
k
/*
m
(A.6)
a
d
= a
nom
%
a
(A.7)
E
d
= *
sd
E{*
f
F
k
, *
f
? F
k
, a
nom
%
a
...} (A.8)
R
d
= R {X
k
/*
m
, a
nom
%
a
...}/*
Rd
(A.9)
E
d
= E {*
F
F
k
, a
nom
} (A.10)
*
F
F
k
a
nom
= *
f
F
k
(a
nom
+ %
a
)*
Sd
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
42
BSI 04-2000
In addition, *
F
is highly standardized. For instance *
F
= 1,5 for all variable loads. Therefore, the
characteristic value is recommended to be adjusted when necessary.
b) On the resistance side (depending on the particular Eurocode):
Providing R is proportional to the strength X, the model uncertainty F and the geometrical property a,
i.e. R ! FaX. the following simple relations apply:
For non-linear models, or in the case of multi-variable load or resistance models, commonly encountered in
Eurocodes, these relations become more complicated.
Table A.4 Expression for ?
0
*
F
= *
f
*
Sd
(1 + %
a
/a
nom
) (A.11)
R
d
= R {X
k
/*
M
, a
nom
} (ENVs 1992 and 1995) (A.12)
R
d
= R {X
k
, a
nom
}/*
R
(ENV 1993) (A.13)
R
d
= R {X
k
/*
m
, a
nom
}/*
rd
(ENV 1994) (A.14)
*
M
= *
m
*
Rd
/{1 + %
a
/a
nom
} (ENVs 1992 and 1995) (A.15)
*
R
= *
m
*
Rd
/{1 + %
a
/a
nom
} (ENV 1993) (A.16)
*
rd
= *
Rd
/{1 + %
a
/a
nom
} (ENV 1994) (A.17)
Distribution
?
o
= F
non dom
/F
dom
General
Normal
(approximation)
Gumbel
F
s
() is the probability distribution function of the extreme value of the non dominating load in the
design period T;
9() is the standard normal distribution function;
N is the T/T
1
;
T is the design period;
T
1
is the period of an independent load variation of the slower varying load;
" is the reliability index;
V is the coefficient of variation for the non dominating load.
NOTE For intermittent loads, the parameter T
1
is equal to the duration of the load and F
s
( ) represents the unconditional
distribution function of the load intensity; so F
s
( ) is not the conditional distribution function given that the load is active.
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
BSI 04-2000 43
A.5 Closure
It is clear from A.1 to A.4 that the same level of formal reliability can be obtained in many different ways.
Some partial factors may be put equal to 1,0 and the required safety margin may be included in another
factor. High characteristic values and low safety factors may be taken or vice versa. The various safety
elements form a set of communicating vessels. For every individual design situation, however, there is
the possibility of calibrating the specific coefficients in order to obtain the required reliability level.
In the currently available set of Eurocodes, the characteristic values for loads and strength parameters and
the geometrical properties are generally taken in accordance with A.2 to A.4. ENV 1991-1 gives values for
the partial load factors and the material-related design codes give values for the partial resistance factors.
This is mainly done in a global way, partly based on probabilistic considerations, partly on a historical or
empirical motivation. Furthermore, the choice of the representative values and the corresponding values
for partial factors was done taking into account the needs for and aspects of an easy and economic
application of the verification procedure in practical design. This has led to the following requests.
For common structures the design values for actions or action effects should be independent of the
design values of the resistance.
There should be only a small set of *
F
values.
Only one constant *
M
value should be given for each material property.
Further simplifications concerning the safety and serviceability verification as well as in structural
analysis should be possible, i.e. avoiding the need to consider too many load arrangements, load cases,
and load combinations in the relevant design situations.
Annex B (informative)
Fatigue
B.1 The fatigue phenomenon
(1) Fatigue is a local material deterioration caused by repeated variations of stresses or strains.
(2) Low cycle fatigue and high cycle fatigue may be distinguished.
low cycle fatigue is associated with non-linear material and geometric behaviour, e.g. alternating plastic
strains in plastic zones. Criteria to exclude low cycle fatigue are given in ENVs 1992 to 1999.
high cycle fatigue is mainly governed by elastic behaviour. Therefore the analysis model can be elastic.
(3) Criteria for determining whether fatigue assessment is needed are given in ENVs 1992 to 1999.
B.2 Fatigue resistance
(1) Except for cases where the fatigue strength of members is determined in specific tests with a load-time
history close to the actual loading to which they are subjected, the fatigue behaviour of structural members
is generally studied for code purposes by simplified tests. In these tests the members are subjected to
constant amplitude load variations, until excessive deformations or fractures due to cracking occur.
(2) The fatigue strength of a given detail is then defined by a %B
R
N
R
relationship, which approximately
represents the 95 % fractile of survival; where %B
R
is the stress range and N
R
the number of cycles up to
failure.
This relationship may be modelled by a standardized linear, bilinear or trilinear line in double logarithmic
scale.
(3) For a range of details, a system of such equidistant %B
R
N
R
curves may be established to allow for
classification.
B.3 Determination of fatigue action effects compatible with the fatigue resistance
(1) Fatigue actions are specified in the other Parts of ENV 1991.
(2) When stress-time histories representative of the fatigue action on a given detail are available, any
stress-time history may be evaluated using the reservoir counting method or rainflow counting method.
These methods enable stress ranges and the numbers of cycles to be determined, together with the
associated mean stresses when these are relevant.
(3) The stress ranges and the number of cycles may be ordered in stress-range frequency distributions or
stress-range spectra.
(4) The stress-range frequency distributions or stress range spectra may be transformed to
fatigue-damage-equivalent constant amplitude stress-range spectra using Miners rule.
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
44
BSI 04-2000
B.4 Fatigue verification
(1) The safety verification for fatigue may be carried out by:
a damage calculation, where the damage caused by the fatigue actions is related to an ultimate
damage representing the limit state;
a fatigue life verification, where, for a representative level of stress range, a damage equivalent
number of load cycles caused by the fatigue action is related to an ultimate number of cycles representing
the limit state;
a stress range verification, where, for a representative number of stress cycles, the magnitude of the
damage-equivalent stress range, caused by the fatigue action is related to an ultimate stress-range
resistance representing the limit state.
(2) Further information is given in ENVs 1992 to 1999.
B.5 Safety concept
(1) In general the design of a structure susceptible to fatigue should be such that it is damage-tolerant. To
be damage-tolerant the structure should be capable of sustaining all loads with sufficient reliability until
cracks can be detected by regular inspections and appropriate remedial measures can be undertaken before
structural failure occurs.
(2) For structures that can be verified to be damage-tolerant, the safety factor *
M
on the fatigue resistance
side may be taken as 1,00.
(3) For structures for which the damage-tolerance cannot be verified, the safety factors have to be
chosen such that they take account of the uncertainties in defining the fatigue actions, the fatigue
action effects, and the fatigue resistances, and also the possible decrease of resistances by corrosion or
other time-dependent phenomena, having due regard for the consequences of a failure without
pre-warning.
(4) Further information on design against fatigue is given in ENVs 1992 to 1999.
Annex C (informative)
Serviceability limit state: verification of structures susceptible to vibrations
C.1 General
C.1.1 Objective
(1) This annex gives guidance for serviceability limit state verifications of structures susceptible to
vibrations.
(2) It deals with the treatment of the action side, the determination of the structural response and the
limits to be considered for the structural response to ensure that vibrations are not disturbing or harmful.
(3) Dynamic effects relating to ultimate limit states or fatigue are treated in the other Parts of ENV 1991
and therefore are not considered in this annex.
C.1.2 Sources of vibrations
(1) Vibrations may be induced by the following:
a) people, e.g. on:
pedestrian bridges;
floors where people walk;
floors for sport or dance activities;
floors with fixed seating and spectator galleries.
b) machines, e.g. on:
machine foundations and supports;
bell towers;
ground with transmitted vibrations.
c) wind, e.g. on:
buildings;
towers;
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
BSI 04-2000 45
chimneys and masts;
guyed masts;
pylons;
bridges;
cantilevered roofs.
d) traffic, e.g. on:
rail or road bridges;
buildings, such as in exhibition halls or car parks.
e) earthquakes.
C.1.3 Modelling of actions and structures
(1) For serviceability limit states the modelling of these actions and of the structure depends on how the
serviceability limits are formulated.
(2) These limits may refer to;
human comfort;
limits for the proper functioning of machines or other installations;
maximum deflection limits to avoid damages or pounding.
(3) In order to verify that these limits are not exceeded the actions may be modelled in terms of force-time
histories, for which the structural responses may then be determined as time histories of deflections or
accelerations by using appropriate integration methods.
(4) Where the structural response may significantly influence the force-time histories to be applied
(e.g. when vehicles are excited to self vibrations by the vibrations of the structure or when synchronising
effects of moving masses occur) these interactions have to be considered either in modelling a combined
load-structure vibration system or by appropriate modifications of the force-time histories.
C.2 Force-time histories
C.2.1 General
(1) The force-time histories used in the dynamic analysis should sufficiently represent the relevant loading
situations for which the serviceability limits are to be verified.
(2) The force-time histories may model:
human induced vibrations, e.g. the walking or running of a single person or a number of persons, or
dancing or motions in stadia or concert halls;
machine induced vibrations, e.g. by force vectors due to mass eccentricities and frequencies, that may
be variable with time;
wind induced vibrations;
traffic loads, e.g. fork-lift trucks, cars or heavy vehicles;
crane operations;
other dynamic actions such as wave loads or earthquake actions.
C.3 Modelling of structures
C.3.1 General
(1) The dynamic analysis model to be used for determining the action effects from force-time histories shall
be established such that all relevant structural elements, their masses and stiffness and damping ratios
are realistically considered.
(2) In the case when dynamic actions are caused by the motion of masses (e.g. by persons, machinery etc.)
these masses should be included in the analysis (e.g. when determining the eigen frequencies).
(3) For other variable actions to be combined with the self weight of the structure the quasi-static values
should be used, unless other specifications are given in the identification of the serviceability limit states.
(4) Where there is significant ground-structure interaction, the contribution of the soil may be modelled by
appropriate equivalent springs and dampers.
(5) In general the behaviour of the structure should be taken as linear, unless other specifications are given
in defining the limit states.
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
46
BSI 04-2000
(6) Damping ratios should be evaluated by using suitable experimental procedures, approved theoretical
methods and values derived from collections of reliable measurements of homogenous structural classes.
C.4 Evaluation of the structural responses
C.4.1 General
(1) The evaluation of structural responses depends on the limits that are specified for them.
(2) Limits may be expressed in terms of:
a) r.m.s (root-mean-square) values determined for a certain exposure time
where:
b) extreme values during a certain exposure time T for narrow banded stochastic responses only
where:
(3) The structural responses a
eff
or a
max
should be compared with the specified limits.
C.4.2 Limiting values for vibrations
C.4.2.1 Human comfort
(1) Where conditions for human comfort are specified, these conditions should be given in terms of an
acceptance criteria according to ISO 2631.
(2) The acceptance criteria should include the relevant acceleration (a
eff
) frequency (f
s
) line for the selected
exposure time and direction of vibration.
C.4.2.2 Functioning of machines
(1) Limits for the movements of the machines should be specified in terms of maximum deflections and
frequency (maximum deflection-frequency lines).
C.4.2.3 Other limits
(1) Limits not covered by acceleration-frequency lines or deflection-frequency lines may be:
the attainment of a maximum stress (e.g. to avoid permanent deformations);
the attainment of a maximum stress range (e.g. to avoid a limited fatigue life or accumulative
deflections);
the attainment of a maximum deformation (e.g. to avoid bumping and for continuous operations).
These limits should be given in the design specifications.
Annex D (informative)
Design assisted by testing
D.1 Scope and objectives
(1) This annex is intended to give guidance for the planning and evaluation of experiments to be carried
out in connection with structural design as indicated in Section 8, when the number of tests is sufficient
for a statistical interpretation of their results.
(C.1)
a
eff
the effective or r.m.s value or the response e.g. the effective acceleration
T is the exposure time;
a
i
is the response value (e.g. acceleration) for each time step %t
i
;
t is time;
%t
i
is the time step.
(C.2)
n is the natural frequency of the structure
a
max
is the expected maximum value of the response, e.g. the maximum acceleration
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
BSI 04-2000 47
(2) Testing may be carried out in the following circumstances:
if material properties or load parameters are not sufficiently known;
if adequate calculation models are not available;
if a large number of similar components will be used;
if the real behaviour is of special interest;
to define control checks assumed in design.
(3) The following types of tests are distinguished:
a) tests to establish directly the ultimate resistance or serviceability properties of structural parts,
e.g. fire tests;
b) tests to obtain specific material properties, e.g. ground investigations or testing of new materials;
c) tests to reduce uncertainties in load or resistance models, e.g. wind tunnel testing, testing of full-size
prototypes, testing of scale models;
d) control tests to check the quality of the delivered products or the consistency of the production
characteristics, e.g. concrete cube testing;
e) tests during execution in order to take account of actual conditions experienced e.g. post-tensioning,
soil conditions;
f) control tests to check the behaviour of the actual structure or structural elements after completion,
e.g. proof loading for the ultimate or serviceability limit states.
(4) The results may be used for a specific structure or may serve as a basis for the design of a wide range
of structures, including the development of rules in structural codes.
(5) Further information on design assisted by testing may be found in ENVs 1992 to 1999.
D.2 Planning
Prior to the execution of tests, a test plan should be agreed with the testing organization. This plan should
contain the objective of the test and all specifications necessary for the selection or production of the test
specimens, the execution of the tests and the test evaluation. In particular, the test plan should deal with
the following items.
a) Scope
The information required from the tests should be clearly stated, e.g. the required properties, the
influence of certain design parameters varied during the test and the range of validity. Limitations of
the test and required conversions should be specified.
b) Expected behaviour
It is essential to present a description of all properties and circumstances which may influence the
behaviour at the limit state under consideration, e.g. geometrical parameters and their tolerances,
material properties, parameters influenced by fabrication and erection procedures, scale effects and
environmental conditions. Modes of failure and/or calculation models with the corresponding variables
should be described. When the prediction of the critical failure modes to be expected in the tests is
extremely doubtful, the test plan should be developed on the basis of accompanying pilot tests.
c) Specification of test specimen
Properties of the test specimen should be specified; in particular dimensions, material and fabrication
of prototypes, number of test specimens, sampling procedures, restraints. Normally a representative
sample in the statistical sense should be aimed for.
d) Loading specifications
Based on item b) loading and environmental conditions in the test should be specified, in particular,
loading points, loading paths in time and space, temperatures, loading by deformation or force control,
etc. Loading paths should be selected such that they are representative for the anticipated scope of
application of the structural member. Account should be taken of possible unfavourable paths and/or of
those paths which are considered in calculations in comparable cases. Interactions with structural
response should be considered where relevant.
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
48
BSI 04-2000
Where structural properties are conditioned by one or several effects of actions which are not varied
systematically, then these effects should be specified at least by their design values. Where they are
independent of the other parameters of the loading path, design values related to estimated values of
lead combination may be adopted.
e) Testing arrangements
Special attention should be given to measures to ensure sufficient strength and stiffness of the loading
and supporting rigs, and clearance for deflections, etc.
f) Measurements
A list should be made of all relevant properties of each individual test specimen to be determined prior
to the execution of the tests. Similarly a list should be made of observation points and methods for
observation and recording, e.g. time histories of displacements, velocities, accelerations, strains, forces
and pressures, required frequency and accuracy of measurements and measuring devices. Depending
on the type of test it could be recommended to have some measurements available during the test.
g) Evaluation and reporting of the test
For specific guidance, see ENVs 1992 to 1999.
D.3 Evaluation of test results
D.3.1 General
(1) All test results should be evaluated critically. The general behaviour and failure modes should be
compared with the expected ones. When large deviations from the expectation occurs, an explanation
should be sought, involving additional tests if necessary.
(2) Where relevant, the evaluation of test results should be on the basis of statistical methods. In principle
the tests should lead to a statistical distribution for the preselected unknown variables, including the
statistical uncertainties. Based on this distribution, design values, characteristic values and partial safety
factors to be used in partial coefficient design may be derived. If possible, only the characteristic value may
be derived while the partial factor is taken from normal design procedure
(3) If the response (or the strength) of the material depends on the load duration or history, the volume or
scale, the environmental conditions, or other non-structural effects, then the calculation model should take
these items into account by use of appropriate factors (conversion) and scaling rules. Further guidance may
be found in ENVs 1991 to 1999. In particular where codes include implicit safety provisions related to
comparable situations, these provisions should also be applied when testing and may give rise to additional
safety elements in the formulae. An example is the effect of tensile strength in concrete test specimens,
which in many cases is neglected during design.
(4) The result of a test evaluation is valid for the specifications and load characteristics considered.
Extrapolation to cover other design parameters and loadings requires additional information, e.g. from
previous tests or theoretical considerations.
D.3.2 Statistical evaluation of resistance/material tests
D.3.2.1 General
(1) This clause is intended to give the operational formulae for deriving design values from the test types
a) and b) for resistance and material testing [see D.1(3)], where the characteristic value is determined from
a standardized or established distribution of the material properties. Use will be made of Bayesian
procedures with vague prior distributions.
NOTE This leads to almost the same result as classical statistics with confidence levels equal to 0.75.
(2) In 8.3 two different methods are distinguished. In method a) a characteristic value is derived first and
then divided by the relevant partial factor. In method b) a direct determination of the design value is made.
These methods are discussed in D.3.2.2 and D.3.2.3 respectively.
(3) The tables and formulae in D.3.2.2 and D.3.2.3 are based on:
the normal distribution;
a complete lack of prior knowledge for the mean;
a complete lack of prior knowledge for the coefficient of variation in the case V
x
unknown or, on the
other hand, full knowledge for the coefficient of variation in the case V
x
known.
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
BSI 04-2000 49
In practice there may be prior knowledge that the distribution type is of a more favourable nature (for
instance the lognormal distribution) and there might be partial prior knowledge, both about the mean and
about the standard deviation. This prior knowledge may be based on previous experience with similar cases
and will in general lead to more favourable design values. Further guidance is, however, outside the scope
of this annex.
D.3.2.2 Method a) Assessment via the characteristic value
Assume that a sample of n numerical test results is available. The design value of a variable X is obtained
from:
where:
The assessment of the conversion factor is strongly dependent on the type of test and the type of material.
No further guidance is given here.
The partial factor should be selected from the field of application under consideration in the test.
The value of k
n
follows from Table D.1. Table D.1 is based on the 5 % characteristic value and on the normal
distribution. Two cases are considered as follows.
i) The coefficient of variation V
x
is known from pre-knowledge; pre-knowledge might be found from the
evaluation of previous tests in comparable situations. What is comparable is determined by engineering
judgement. In that case the row V
x
known should be used.
The coefficient of variation V
x
is not known from pre-knowledge, but must be estimated from the sample:
In this case the row V
x
unknown should be used.
Table D.1 Values of k
n
for the 5 % characteristic value
D.3.2.3 Method b) Direct assessment of the design value
In method b) the design value for X follows from:
The meaning of all variables is the same in D.3.2.2, however, )
d
should now cover all uncertainties not
covered by the tests. The value of k
n
should now follow from Table D.2 or Table D.3.
If X is the dominating variable in the resistance model, k
n
may follow from Table D.2. The table is based on
the assumption that the design value corresponds to = 3,8 and ! = 0,8 (see Annex A) and that X is
normally distributed. This gives a value with about 0,1 % probability of observing a lower value.
(D.1)
*
M
is the partial factor for the design;
)
d
is the design value of the conversion factor;
X
k(n)
is the characteristic value including statistical uncertainty;
m
x is the mean of the sample results
V
x
is the coefficient of variation of X;
k
n
is the coefficient following from Table D.1.
(D.2)
V
x
= s
x
/m
x
(D.3)
n 1 2 3 4 5 6 8 10 20 30 Z
V
x
known 2,31 2,01 1,89 1,83 1,80 1,77 1,74 1,72 1,68 1,67 1,64
V
x
unknown 3,37 2,63 2,33 2,18 2,00 1,92 1,76 1,73 1,64
(D.4)
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
50
BSI 04-2000
If both a design value and a characteristic value are determined, a partial factor can be found from
*
M
= X
k
/X
d
.
If X is a non-dominating variable, then ! = 0,4 0,8 (see Annex A) and Table D.3 should be used. The
probability of observing a lower value is about 10 %.
Table D.2 Values of k
n
for the ULS design value, if X is dominating (P{X < X
d
} = 0,1 %)
Table D.3 Values of k
n
for the ULS design value, if X is non-dominating (P{X < X
d
} = 10 %)
D.3.3 Evaluation of tests for determining model factors
(1) In some cases a tentative calculation model is available, but the accuracy of the model is not known or
the uncertainty is too large for some fields of application. In those circumstances tests can be carried out
to find the statistical characteristics and design values of the model factors test type c as described
in D.1(3). This type of testing is often performed in the process of codification of design formulae. It is
assumed that the available model, although incomplete, predicts adequately the basic tendencies. In
principle the calculation model can range from simple semi-empirical formulae to advanced finite element
models.
(2) For resistance testing account should be taken of the fact that a structural member may possess a
number of fundamentally different failure modes. For example; a girder may fail by bending at midspan or
shear at the supports. It is possible that the average strength region is governed by different modes than
the low strength region. As the low strength region (e.g. mean value minus two to three standard
deviations) is most important in reliability analysis, the modelling of the member should focus on the
corresponding mode.
(3) Assume that the calculation model available is as follows:
where:
(4) Assume a series of n experiments (i = 1 ... n) is carried out, where:
the values of W have been taken equal to w
i
;
the values of X have been measured as x
i
;
the values of R have been measured as r
i
.
(5) It is recommended that the observed experimental results r
i
are plotted against the calculated values
R
t
(x
i
w
i
) according to the model and versus each of the observed basic variables. This plotting procedure is
intended to check whether the calculation models adequately account for the respective variables.
(6) If more than one failure mode is observed in the test results, it is recommended that the tests in a
number of series are repeated. In every series all modes but one should be excluded.
(7) From the test results the following set of observations for the unknown coefficient D may by derived:
n 1 2 3 4 5 6 8 10 20 30 Z
V
x
known 4,36 3,77 3,56 3,44 3,37 3,33 3,27 3,23 3,16 3,13 3,08
V
x
unknown 11,40 7,85 6,36 5,07 4,51 3,64 3,44 3,08
n 1 2 3 4 5 6 8 10 20 30 Z
V
x
known 1,81 1,57 1,48 1,43 1,40 1,38 1,36 1,34 1,31 1,30 1,28
V
x
unknown 3,77 2,18 1,83 1,68 1,56 1,51 1,45 1,36 1,33 1,28
R = D R
t
(X, W) (D.5)
X is the vector of random variables;
W is the set of measurable deterministic variables;
R
t
is the theoretical model;
R is the measurable result of the experiment;
D is the unknown coefficient, to be determined by the experiment.
d
i
= r
i
/ R
t
{x
i
, w
i
} (D.6)
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
BSI 04-2000 51
(8) It will be assumed that D has a normal distribution. It should be noted that the normal distribution may
be replaced by a lognormal distribution, provided that this can be justified from experience with similar
tests in the past.
(9) The further statistical evaluation of D is the same as in D.3.2. For cases where the deterministic
specifications W are varied and/or the random basic variables X are measured indirectly or not at all,
specialist literature should be consulted.
D.3.4 Design value for quality control tests
(1) Control tests, as defined as test type d in D.1(3), are intended to check the quality of the delivered
products or the consistency of the production characteristics.
(2) It will be assumed that the product under consideration is produced in batches. A batch is tentatively
defined as a set of units, produced by one producer, in a relatively short period, with no obvious changes in
production circumstances.
(3) For discrete products the definition of a unit is generally self-evident. For continuously produced
materials, a unit may be defined as one test specimen, e.g. a concrete test cube.
In practice, batches correspond to, for example:
a single production of concrete from the same materials and plant;
structural steel from one melt processed according to the same conditions;
foundation piles for a specific site.
(4) Quality control may be performed on every unit (total control) or on samples (batch control). Typically,
testing all units requires a non-destructive testing technique. In general a non-destructive testing
technique is not able to predict the strength with the same precision as a destructive testing technique.
Therefore some kind of measurement error has to be included. In theory there is always a measurement
error present, but this can often be ignored.
(5) If sampling is used, a random sample is usually taken. In a random sample each unit of the batch has
the same probability of being sampled.
(6) If quality control is performed on the basis of pre-defined selection rules, the control may lead to three
possible outcomes:
the batch or unit is rejected: d < 0;
the batch or unit is critical: d = 0;
the batch or unit is fully acceptable: d > 0.
Where d is a function of the test result of a single unit or of the combined test result of the units in a sample.
A common formulation for an acceptance rule is given by:
where:
From which d = m
x
2
n
s
x
X
c
The number of tests n and the parameters 2
n
and X
c
should be determined in such a way that an economical
and efficient test is obtained.
(7) In practice two requirements are often defined which should be met simultaneously. In those cases the
batch is accepted only if for example d
1
> 0 and d
2
> 0. The second requirement is often related to the lowest
observation, and could be of the type:
(8) The design value corresponding to given quality control criteria should be calculated on the basis of:
the operations characteristic of the control rules; this is the probability of some given batch being
accepted;
m
x
> X
c
+ 2
n
s
x
(D.7)
m
x
is the sample mean;
s
x
is the sample standard deviation;
X
c
is a fixed value, for instance the required characteristic value;
2
n
is a number, normally depending on n.
x
min
> X
c
(D.8)
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
ENV 1991-1:1994
52
BSI 04-2000
the production characteristic; this is the information about the batch-to-batch variation in the
uncontrolled supply.
General formulae are out of the scope of this annex.
(9) Consider by way of example the case that x has a normal distribution, a known standard deviation, that
there is no prior knowledge about the mean and that a single criterion [as in expression (D.7)] is present.
The design or characteristic value based on the critical batch (having d = 0) is then given by:
The value of k
n
follows from Table D.1, Table D.2 and Table D.3, where V
x
known must be assumed. It
should be noted that in most quality control tests there is substantial information on the mean, which leads
to more favourable values. This is as also stated in D.3.4(8) and is outside the scope of this annex.
(10) Finally for total or unit-by-unit testing, it is reasonable to expect some substantial error, as this is
normally carried out by a non-destructive testing procedure. It is assumed here that an error e is present
with mean zero and standard deviation B
e
. It is further assumed that the mean and standard deviation of
x, either of the batch or the total supply, are known:
The result for this case is also conservatively based on the critical unit and not on the arbitrary accepted
unit. The value of k
n
follows from Table D.1, Table D.2, and Table D.3, where V
x
known may be assumed.
D.3.5 Proof Loading
(1) Proof loading is a test on the actual structure, i.e. test type f in D.1(3). Special care should be taken that
the structure is not unnecessarily damaged during the test. This requires a continuous monitoring of the
load and the response.
(2) A distinction is made between:
an acceptance test and
a strength test.
(3) The acceptance test is intended to confirm that the overall structural performance complies with design
intentions. The load is raised to values between the characteristic value and the design value for the
ultimate limit state. Requirements may be set for the deformations, the degree of non-linearity and the
residual deformations after removal of the test loading.
(4) The strength test is intended to show that the structure or the structural element has at least the
strength that is assumed in the design. If an assessment for the test element only is required, it is sufficient
to raise the load to the design load for the ultimate limit state. Obviously, as already stated in D.3.5(1) care
should be taken not to damage the structure unnecessarily.
(5) If the strength test is intended to prove that other but similar elements also have the required strength,
a higher load is required. A minimum requirement in this respect would be to correct the design load for
the presence of better material properties in the tested element, compared to the design values. This means
that the material properties of the tested element have to be measured.
(6) If the relationship between the resistance and the material property is linear, the design strength R
d

corresponding to a successful test with test load F
t
is:
From the requirement R
d
U F
d
, the minimum test load can be calculated.
(7) If it is not possible to measure the material properties, the design value for the resistance can
conservatively be found from:
Here V
R
is the known coefficient of variation for the resistance of the element population under
consideration and k
n
follows from Table D.2. The case with V
x
unknown is outside the scope of this annex.
(8) It is also possible to use a combination of expressions (D.11) and (D.12), e.g. if only a part of the relevant
random variables can be measured. If V is not know by pre-knowledge, a more sophisticated analysis is
required. This is outside the scope of this annex.
X
k
or X
d
= X
c
+ (2
n
k
n
)B
x
(D.9)
(D.10)
R
d
= F
t
X
d
/ X
t
(D.11)
X
t
is the strength of material in the test.
R
d
= F
t
(1 k
n
V
R
) (D.12)
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
blank
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I
DD ENV
1991-1:1996
BSI
389 Chiswick High Road
London
W4 4AL
BSI British Standards Institution
BSI is the independent national body responsible for preparing
British Standards. It presents the UK view on standards in Europe and at the
international level. It is incorporated by Royal Charter.
Revisions
British Standards are updated by amendment or revision. Users of
British Standards should make sure that they possess the latest amendments or
editions.
It is the constant aim of BSI to improve the quality of our products and services.
We would be grateful if anyone finding an inaccuracy or ambiguity while using
this British Standard would inform the Secretary of the technical committee
responsible, the identity of which can be found on the inside front cover.
Tel: 020 8996 9000. Fax: 020 8996 7400.
BSI offers members an individual updating service called PLUS which ensures
that subscribers automatically receive the latest editions of standards.
Buying standards
Orders for all BSI, international and foreign standards publications should be
addressed to Customer Services. Tel: 020 8996 9001. Fax: 020 8996 7001.
In response to orders for international standards, it is BSI policy to supply the
BSI implementation of those that have been published as British Standards,
unless otherwise requested.
Information on standards
BSI provides a wide range of information on national, European and
international standards through its Library and its Technical Help to Exporters
Service. Various BSI electronic information services are also available which give
details on all its products and services. Contact the Information Centre.
Tel: 020 8996 7111. Fax: 020 8996 7048.
Subscribing members of BSI are kept up to date with standards developments
and receive substantial discounts on the purchase price of standards. For details
of these and other benefits contact Membership Administration.
Tel: 020 8996 7002. Fax: 020 8996 7001.
Copyright
Copyright subsists in all BSI publications. BSI also holds the copyright, in the
UK, of the publications of the international standardization bodies. Except as
permitted under the Copyright, Designs and Patents Act 1988 no extract may be
reproduced, stored in a retrieval system or transmitted in any form or by any
means electronic, photocopying, recording or otherwise without prior written
permission from BSI.
This does not preclude the free use, in the course of implementing the standard,
of necessary details such as symbols, and size, type or grade designations. If these
details are to be used for any other purpose than implementation then the prior
written permission of BSI must be obtained.
If permission is granted, the terms may include royalty payments or a licensing
agreement. Details and advice can be obtained from the Copyright Manager.
Tel: 020 8996 7070.
L
i
c
e
n
s
e
d

C
o
p
y
:

U
n
i
s

L
i
b
r
a
r
i
e
,

,

1
7
-
A
u
g
-
0
0
,

U
n
c
o
n
t
r
o
l
l
e
d

C
o
p
y
.


B
S
I