Structures Congress 2011 © ASCE 2011 1425

Load Combination Requirements in ASCE Standard 7-10: New

Developments

Therese P. McAllister,1 M. ASCE

Bruce R. Ellingwood,2 Dist. M. ASCE

Abstract

The fundamental load combination requirements for strength design (or load and
Downloaded from ascelibrary.org by University of Leeds on 07/26/13. Copyright ASCE. For personal use only; all rights reserved.

resistance factor design) appearing in Section 2.3 of ASCE Standard 7-10

incorporate major revisions, among them the treatment of self-straining actions,
fluid and soil loads, load combinations for extraordinary events, and probability-
based load requirements for non-specified loads. The last item is based on
suggested reliability levels for the new performance-based design procedures in
the Section 1.3.1 Commentary. The Commentary to Section 2 of ASCE Standard
7-10 presents the technical basis for these changes. Since the basis of the
traditional allowable stress design combinations was purely judgmental, the ASD
combinations in Section 2.4 have been brought into alignment with the strength
combinations in Section 2.3, which have a rational reliability basis.

Introduction

The fundamental load combination requirements for strength design (or load and
resistance factor design) appearing in Section 2.3 of ASCE Standard 7-10 were
first implemented in 1982 (Galambos et al., 1982; Ellingwood et al., 1982). These
load combinations were developed using modern probabilistic models of
structural loads and load combinations, statistical data on load intensities, and
structural reliability analysis. They utilize a “principal action – companion
action” format, which is based on the notion that the maximum combined load (or
load effect) during a service period (commonly taken as 50 years) occurs when
one time-varying load attains its maximum value while the remaining time-
varying loads are at their frequent (or arbitrary-point-in-time) values. Research
has demonstrated conclusively that this model best represents how time-varying
loads found in other sections of ASCE 7 combine on structural members,
components and systems (Turkstra and Madsen, 1980).

The new load combination requirements for strength design in ASCE Standard 7-
10 incorporate a number of significant revisions. The self straining force, T, fluid
load, F, and soil (lateral earth pressure) load, H, are no longer shown in the basic
combinations. The reliability analysis leading to ASCE 7-05, Section 2.3.2, was
supported by an extensive statistical database on dead, live, snow, wind and
earthquake loads. However, the self-straining force, fluid load, and soil load do
not have the same statistical support as the primary loads. Over the years, they
were added to the combinations primarily on the basis of judgment, in order to be
responsive to users of ASCE Standard 7 and the other design documents that refer
to ASCE Standard 7. The treatment of these loads now is addressed separately in

1
Research Structural Engineer, Building and Fire Research Laboratory, National Institute of
Standards and Technology, Gaithersburg, MD 20899 (therese.mcallister@nist.gov)
2
The Raymond Allen Jones Chair in Civil Engineering, School of Civil and Environmental
Engineering, Georgia Institute of Technology, Atlanta, GA 30332 (ellingwood@gatech.edu)

Structures Congress 2011

 Structures Congress 2011 © ASCE 2011 1426

text following the basic combinations to remain responsive to the needs of the
design community and, at the same time, to distinguish load combinations
technically supported by engineering principles and observed data from those
based on judgment. The wind load factors were revised to accommodate the
changes in the wind load criteria. The load combinations for extraordinary events,
which formerly appeared in a Commentary section, now appear in the Standard,
where they serve as a point of reference for codes, standards and guidelines for
design against disproportionate collapse, fire and other extreme events. In
anticipation of the demands for “beyond-code” design from performance-based
Downloaded from ascelibrary.org by University of Leeds on 07/26/13. Copyright ASCE. For personal use only; all rights reserved.

engineering and to accommodate design of structural systems for loads that are
not included in ASCE Standard 7, a new standardized method for calculating load
and resistance factors has been provided, along with an extended commentary on
its proper application. The load 0.6D in the ASD combinations for counteracting
loads now is permitted to be increased to 0.9D in certain instances. Since the basis
of the traditional allowable stress design combinations was purely judgmental, the
ASD combinations in Section 2.4 have been brought into alignment with the
strength combinations, which have a rational reliability basis. Additional
guidance on the use of LRFD and ASD methods for structural design is provided
in the Commentary. Finally, the Commentary to Section 2 of ASCE Standard 7-10
has been extensively revised to explain the technical basis for these changes.

The following sections describe the basis for these changes in more detail.

Self-Straining Force

The self-straining force, T, was removed from the basic combinations in Sections
2.3.2 and 2.4.1 and now is treated in Sections 2.3.5 and 2.4.4, respectively, with
an extended commentary. The self-straining force covers temperature effects,
creep, shrinkage, differential settlement and similar effects. The presence of T in
the basic combinations was problematic because its effect depends on the material
of construction and conditions of constraint. This load situation is more a
potential issue for concrete/masonry than it is for steel. Cracking may relieve the
stress in concrete/masonry structures; however, it may be detrimental to the
performance and serviceability of structures, such as tanks and containment
structures.

Self-straining actions should be calculated based on a realistic assessment of the

most probable values rather than the upper bound values. The most probable value
is the value that can be expected at any “arbitrary point-in-time”.

When self straining loads are combined with dead loads as the principal action, a
load factor of 1.2 may be used. However, when self straining loads are considered
as a companion load, the load factor may be reduced to no less than 1.0, if it is
unlikely that the principal and companion loads will attain their maximum values
at the same time.

If only limited data are available to define the magnitude and frequency
distribution of the self-straining load, then its value must be estimated carefully.
Estimating the uncertainty in the self straining load may be complicated by
variability in material stiffness of a member or structure under consideration.

Structures Congress 2011

 Structures Congress 2011 © ASCE 2011 1427

Fluid Loads

Fluid load, F, defines structural actions in structural supports, framework, or

foundations of a storage tank, vessel, or similar container due to stored liquid
products. The fluid load of a product in a storage tank shares characteristics of
both dead and live loads. It is similar to a dead load in that its self-weight and
hydrostatic loads have a maximum calculated value, and that the magnitude of the
actual load may have a relatively small dispersion. However, the fluid load is not
permanent. Emptying and filling causes fluctuating forces in the structure, the
Downloaded from ascelibrary.org by University of Leeds on 07/26/13. Copyright ASCE. For personal use only; all rights reserved.

maximum load may be exceeded by overfilling, and densities of stored products in

a specific tank may vary.

The fluid load, F, was removed in the basic load combinations. There are no
statistical data on fluid loads that can be used to address its treatment in load
combination analysis or to support a particular value of load factor. However,
when fluid loads are considered in design, they are included with the same load
factor as dead load D in combinations 1 through 5, when additive to other loads,
and in combination 7 when it provides resistance to uplift forces. To make it clear
that the fluid weight in a tank can be used to resist uplift, F was added to load
combination 7 where it will be treated as dead load only when F counteracts
earthquake load, E. The mass of the fluid also is included in the inertial effect due
to E (see 15.4.3) and the base shear calculations for tanks (15.7). Note that the
fluid mass effects on stabilization depend on the degree to which the tank is filled.
F is not included in combination 6 because the wind load can be present, whether
the tank is full or empty, so the governing load case in combination 6 is when F is
zero.

Loads due to Lateral Earth Pressure, Ground Water Pressure and Bulk
Materials

Loads due to lateral earth pressure, ground water pressure and pressure of bulk
materials, H, were removed from the basic load combinations. Similar to fluid
loads, there are no statistical data on H. However, where they are present, they are
included in the load combinations as follows:

1. where the effect of H adds to the primary variable load effect, include H
with a load factor of 1.6;
2. where the effect of H resists the primary variable load effect, include H
with a load factor of 0.9 where the soil load is permanent or a load factor
of 0 for all other conditions.

Uncertainties in lateral forces from bulk materials, included in H, are higher than
those in fluids, particularly when dynamic effects are introduced as the bulk
material is set in motion by filling or emptying operations. Accordingly, the load
factor for such loads is set equal to 1.6.

Where H acts as a resistance, a factor of 0.9 is suggested if the passive resistance

is computed with a conservative bias. The intent is that soil resistance be
computed for a deformation limit appropriate for the structure being designed, not
at the ultimate passive resistance. Thus an at-rest lateral pressure, as defined in

Structures Congress 2011

 Structures Congress 2011 © ASCE 2011 1428

the technical literature, would be sufficiently conservative. Higher resistances

than at-rest lateral pressure are possible, given appropriate soil conditions. Fully
passive resistance seldom would be appropriate because the necessary
deformations in the soil would cause the structure to be compromised.
Furthermore, there is great uncertainty in the nominal value of passive resistance,
which would also argue for a lower factor on H.

Load Factors on Wind Loads

Downloaded from ascelibrary.org by University of Leeds on 07/26/13. Copyright ASCE. For personal use only; all rights reserved.

ASCE Standard 7-10 contains new wind speed maps in Chapter 26. The wind
loads, W, for ordinary buildings and other structures in the previous edition of
ASCE Standard 7 were based on mapped 3-s gust wind speeds with a 50-year
mean recurrence interval (MRI) [the MRIs were 100 years for critical facilities.]
and the load factor applied to the nominal wind load was 1.6. The wind
climatology in interior and in hurricane-prone coastal areas of the United States is
different, and using the 50-yr MRI as a basis led to differences in facility risk in
those areas. ASCE Standard 7-10 now maps the “design-basis” wind speed
directly (MRI of 700 years for ordinary buildings; 1700 years for critical facilities),
and applies a load factor of 1.0 to the resulting wind load. This approach ensures
that the risk is essentially uniform and better aligns the manner in which ASCE 7
treats the wind and earthquake hazards. At the same time, the wind load factor in
Section 2.3 was reduced from 1.6 to 1.0 in combinations where it was the
principal load to maintain the reliability of the load combinations.

Counteracting Loads in ASD

Beginning with the 1998 edition of ASCE Standard 7, the load factor on the
nominal dead load in ASD combinations (7) and (8), which govern counteracting
loads, was reduced to 0.6 to achieve consistency between ASD and LRFD. This
factored load, 0.6D, has been viewed as problematic and has drawn criticism from
segments of the structural engineering community. Ellingwood and Li (2009)
have confirmed that this reduction was warranted and necessitated by the large
variability in W and E; failure to reduce the gravity load for design (i.e., not using
0.6 D) may lead to inadequate safety in many counteracting load situations.
However, in the design of certain non-building structures, foundations and in
special reinforced masonry shear walls, adjustments to ASD combinations (6) and
(7) may be justified on other grounds. These adjustments are reflected in the
Exceptions to the ASD combinations in Section 2.4.1.

The first of these situations is defined by Exception 2 of Section 2.4.1 for non-
building industrial facilities, which permits the wind load factor in the
counteracting load combination to be modified for foundation design of non-
building structures. When wind forces act on a structure, the structural action
causing uplift at the structure-foundation interface is less than would be computed
from the peak lateral force, due to area-averaging. Accordingly, the wind effects
can be reduced when checking global stability and bearing or uplift at the soil-
foundation interface. Accordingly, for situations where a foundation element
exceeds 1000 ft2 (93 m2), on either a vertical or a horizontal plane, W can be
replaced with 0.9W in combination (7), excluding anchorage of the structure to
the foundation. Exception (2) applies to the design of the foundation and to self-
anchored ground-supported tanks. For different reasons, a similar approach is

Structures Congress 2011

 Structures Congress 2011 © ASCE 2011 1429

already provided for seismic actions by ASCE 7-05 Section 12.13.4 and in Section
12.4.2.2, exception 2.

The second of these situations is defined by Exception 3 for Special Reinforced

Masonry Walls. This Exception is based upon three factors, all of which yield a
conservative design for overturning resistance under the seismic load
combination:

1. The basic allowable stress for reinforcing steel is 40% of the

Downloaded from ascelibrary.org by University of Leeds on 07/26/13. Copyright ASCE. For personal use only; all rights reserved.

specified yield, fy;

2. The minimum reinforcement required in the vertical direction
provides protection against the circumstance where the dead and
seismic loads result in a very small demand for tension
reinforcement; and
3. The maximum reinforcement limit prevents compression failure
under overturning.

Of these three, the low allowable stress in the reinforcing steel is the most
significant. Exception 3 will be reconsidered if the allowable stress in tension
reinforcement in the standard for design of masonry structures is increased above
0.4fy.

Load Combinations for Extraordinary Events

Section 2.5, Load Combinations for Extraordinary Events, provides load

combinations for low-probability events such as fire, explosions and vehicular
impact. Since the 1995 edition of ASCE Standard 7, Commentary C2.5 has
provided a set of load combinations that were derived using a probabilistic basis
similar to that used to develop the load combination requirements for ordinary
loads in Section 2.3 (Ellingwood and Leyendecker, 1978; Ellingwood and
Dusenberry, 2005). In recent years, social and political events have led to an
increasing desire on the part of architects and structural engineers, project
developers and regulatory authorities to enhance design and construction practices
for certain buildings by providing additional structural robustness and to lessen
the likelihood of disproportionate collapse if an extreme event were to occur.
Several Federal, state and local agencies have adopted policies that require new
buildings and structures to be constructed with such enhancements of structural
robustness [GSA 2003, DOD 2009]. Concurrently, advances in structural
engineering for fire conditions (e.g., AISC 2005 Specification, Appendix 4) raise
the prospect that new structural design requirements for fire safety will
supplement the existing deemed-to-satisfy provisions in the next several years.
To meet these needs, the load combinations for extraordinary events, which
formerly appeared in the Commentary, were moved up to Section 2.5 Standard,
where they now serve as a point of reference for codes, standards and guidelines
for design against disproportionate collapse, fire and other extreme events.

Performance-Based Engineering

Engineers may wish to develop load criteria for strength design that are consistent
with the requirements in ASCE Standard 7 for situations that are not covered
explicitly in the Standard. They also may wish to consider loading criteria for

Structures Congress 2011

 Structures Congress 2011 © ASCE 2011 1430

special situations, as required by the client in performance-based engineering

(PBE) applications, in accordance with Section 1.3.1.3. In addition, groups
writing standards and specifications for strength design of structural systems and
elements may wish to develop resistance factors that, when used with the load
requirements in the Standard, permit a stipulated reliability to be achieved. Such
load criteria must be developed using a standardized procedure to ensure that the
resulting factored design loads/load combinations will lead to target reliabilities
(or levels of performance) that can be benchmarked against the common load
criteria in Section 2.3.2 and in existing standards and specifications governing
Downloaded from ascelibrary.org by University of Leeds on 07/26/13. Copyright ASCE. For personal use only; all rights reserved.

strength design for common construction materials. Section 2.3.6 permits, for the
first time, the engineer to develop load combinations for strength design through a
standardized method that is consistent with the methodology used to develop the
basic combinations that appear in Section 2.3.2.

Use of LRFD and ASD Methods

Section 2.1 requires the engineer to use either ASD or LRFD exclusively when
designing a structure. This requirement dates back to the introduction of the load
combinations for strength design (LRFD) in the 1982 edition of the Standard. The
General Commentary in C2.1 notes that an indiscriminate mix of the LRFD and
ASD methods is not permitted in structural design, as such an approach may lead
to unpredictable structural system performance. The reliability analyses and code
calibrations leading to the LRFD load combinations were based on member limit
states rather than system limit states.

However, designers of cold-formed steel and open web steel joists often design
(or specify) these products using ASD and, at the same time, design the structural
steel in the rest of the building or other structure using LRFD. The AISC Code of
Standard Practice states that cold-formed products and steel joists are not
considered to be structural steel. Foundations are also commonly designed using
ASD, while strength design is used for the remainder of the structure. Using
different design standards in these situations has not been shown to be a problem.
Thus, Section 2.1 is intended to permit current industry practice while, at the same
time, not permitting LRFD and ASD to be mixed indiscriminately in the design of
a structural frame.

Summary

The load combination requirements for strength design in ASCE Standard 7-10
incorporate a number of major revisions. Changes to the ASD requirements were
made for consistency with the probability-based strength design requirements:

• Self-straining force, T, was removed from the basic combinations and now
is treated in a separate section, with an extended commentary.

• Wind load factors were revised to accommodate the changes in the wind
speed map in Section 26.

• The load 0.6D in the ASD combinations for counteracting loads may be
increased to 0.9D in design of certain non-building structures and special
reinforced masonry shear walls;

Structures Congress 2011

 Structures Congress 2011 © ASCE 2011 1431

• Load combinations for extraordinary events, formerly in a Commentary

section, were moved up to the Standard, as a point of reference for codes,
standards and guidelines for design against disproportionate collapse, fire
and other extreme events.

• A standardized method for calculating load and resistance factors has been
provided, along with an extended commentary on the application of this
method.
Downloaded from ascelibrary.org by University of Leeds on 07/26/13. Copyright ASCE. For personal use only; all rights reserved.

References

DOD (2009), Design of buildings to resist progressive collapse. Unified Facilities

Criteria (UFC) 4-023-03, Washington, DC, June, 2009.
Ellingwood B.R. and E.V. Leyendecker (1978). “Approaches for design against
progressive collapse.” J. Struct. Div. ASCE 104(3):413-423.
Ellingwood, B.R., et al (1982). “Probability-based load criteria: load factors and
load combinations.” J. Struct. Div. ASCE 108(5):978-997.
Ellingwood, B. (1998). “Reliability-based performance concept for building
construction.” in Structural Engineering Worldwide 1998, Paper T178-4,
Elsevier Science Publishers (CD-ROM).
Ellingwood, B. R. and D.O. Dusenberry 2005. Building design for abnormal
loads and progressive collapse. Computer-aided Civil and Infrastruct.
Engrg. 20(5):194-205.
Ellingwood, B.R. and Y. Li (2009). “Counteracting Structural Loads: Treatment
in ASCE Standard 7-05.” J. Struct. Div. ASCE 135(1):94-97.
GSA (2003), Progressive collapse analysis and design guidelines for new federal
office buildings and major modernization projects. General Services
Administration, Washington, DC.
Galambos, T. V., et al (1982). “Probability-based load criteria: assessment of
current design practice.” J. Struct. Div. ASCE 108(5):959-977.
Turkstra, C.J. and H.O. Madsen (1980). “Load combinations in codified
structural design.” J. Struct. Div., 106(12), 2527-2543.

Structures Congress 2011