4 PLATES

(This table of contents just added for convenience to understand the new chapter structure, to be cut in final version)
4 PLATES ................................................................................................................................................................1
4.1 INTRODUCTION...........................................................................................................................................3
4.2 ELASTIC LOCAL BUCKLING OF FLAT PLATES ....................................................................................5
4.2.1 Uniform Compression............................................................................................................................5
4.2.2 Compression and Bending .....................................................................................................................7
4.2.3 Shear ......................................................................................................................................................9
4.2.4 Shear and Compression .......................................................................................................................11
4.2.5 Shear and Bending ...............................................................................................................................12
4.2.6 Biaxially Compressed ..........................................................................................................................14
4.2.7 Longitudinally varying Compression ..................................................................................................14
4.2.8 Plates on one-way foundations ............................................................................................................15
4.2.9 Singly curved plates.............................................................................................................................16
4.2.10 Square Plate with a Central Hole .........................................................................................................16
4.2.11 Rectangular Plate with Multiple Holes ................................................................................................17
4.3 INELASTIC BUCKLING, POST-BUCKLING, AND STRENGTH OF FLAT PLATES ..........................21
4.3.1 Inelastic buckling.................................................................................................................................22
4.3.2 Post-buckling .......................................................................................................................................22
4.3.3 Strength and Effective Width...............................................................................................................24
4.3.4 Strength in shear and combined loadings with shear ...........................................................................32
4.3.5 Strength of biaxially compressed plates...............................................................................................33
4.3.6 Average stress and strength of aluminum plates..................................................................................34
4.3.7 Yield-line analysis and plastic buckling of plates................................................................................35
4.3.8 Energy dissipation (steel plate shear wall)...........................................................................................36
4.4 BUCKLING, POST-BUCKLING AND STRENGTH OF STIFFENED PLATES......................................38
4.4.1 Compression: Buckling........................................................................................................................38
4.4.2 Compression: Post-buckling and Strength...........................................................................................43
4.4.3 Compression and Shear: Buckling.......................................................................................................47
4.4.4 Compresion and Shear: Post-buckling and Strength............................................................................49
4.4.5 Laterally Loaded plates in compression...............................................................................................49
4.5 BUCKLING OF ORTHOTROPIC PLATES................................................................................................51
4.5.1 Compression ........................................................................................................................................52
4.5.2 Biaxial compression.............................................................................................................................53
4.5.3 Shear ....................................................................................................................................................55
4.6 INTERACTION BETWEEN PLATE ELEMENTS .....................................................................................58
4.6.1 Buckling Modes of a Plate Assembly ..................................................................................................58
4.6.2 Local Buckling of a Plate Assembly....................................................................................................60
4.6.3 Post-buckling of a Plate Assembly ......................................................................................................62
4.7 REFERENCES..............................................................................................................................................63
List of Figures
Fig. 4.1 Compression or flexural members....................................................................................................................3
Fig. 4.2 Local plate buckling coefficient, k of Eq. 4.1, for plates in compression with varied boundary conditions ...6
Fig. 4.3 Plate buckling coefficient, k, as a function of normalized plate length (β=a/b) for different boundary
conditions, m=number of buckled half-waves along the length of the plate (after Yu and Schafer (2007)) ...7
Fig. 4.4 Plate buckling coefficients for long plates under compression and bending (Brockenbrough and Johnston,
1974; Bijlaard, 1957).......................................................................................................................................8
Fig. 4.5 Plate buckling coefficients for unstiffened elements in compression and bending, after (Bambach and
Rasmussen 2004a)...........................................................................................................................................9
Fig. 4.6 Plate buckling coefficients for plates in pure shear. (Side b is the short side.)...............................................10
Fig. 4.7 Interaction curve for buckling of flat plates under shear and uniform compression. .....................................12
Fig. 4.8 Buckling coefficients for plates in combined bending and shear. ..................................................................13
Fig. 4.9 Interaction curve for buckling of flat plates under shear, compression, and bending. ...................................14
Fig. 4.10 Simply supported plate subject to a longitudinal stress gradient (note shear stresses are required for
equilibrium, after (Yu and Schafer 2007)) ....................................................................................................15
Fig. 4.11 Simply supported and clamped plates with hole under shear loading (Rockey et al., 1969)........................17
Fig. 4.12 Illustration of unstiffened strips adjacent to a hole.......................................................................................18
Fig. 4.13 Critical shear stress for webs with holes (Michael, 1960)............................................................................19
Fig. 4.14 Perforated rectangular plate under combined action of shear and bending. .................................................20
Fig. 4.15 Simple model for post-buckling of a flat plate in uniform compression (after AISI 2007)..........................23
Fig. 4.16 Post-buckling stiffness of plates having simply supported edges (Allen and Bulson, 1980) .......................24
Fig. 4.17 Longitudinal stress after buckling and definition of effective width............................................................25
Fig. 4.18 Chart for determining σe/σy ..........................................................................................................................28
Fig. 4.19 Nondimensional buckling curves for plates under uniform edge compression (adapted from
Brockenbrough and Johnston, 1974).............................................................................................................29
Fig. 4.20 Distributing effective width in plates under a stress gradient, after (Schafer and Peköz 1999) ...................31
Fig. 4.21 Shear and tension fieelds (square plate). ......................................................................................................33
Fig. 4.22 Equations for buckling stress and ultimate strength of plates used in Aluminum Association specifications
(AA, 1994). ...................................................................................................................................................35
Fig. 4.23 Hysteretic response of a steel plate shear wall under cyclic loading (Lubell et al. 2000) ............................36
Fig. 4.24 Hysteretic response of steel plate shear walls with cutouts (Hitaka and Matsui 2003) ................................37
Fig. 4.25 Stiffened plate panels; (a) panel with longitudinal stiffeners; (b) panel with transverse stiffeners. .............38
Fig. 4.26 Buckling of columns with elastic supports...................................................................................................42
Fig. 4.27 Initial and final buckled shapes ....................................................................................................................44
Fig. 4.28 Column curve for stiffened panels................................................................................................................45
Fig. 4.29 Column chart by Gerard and Becker (1957/1958). ......................................................................................45
Fig. 4.30 Longitudinally stiffened plate under combined compression and shear stresses..........................................48
Fig. 4.31 Analytical interaction relations for infinitely wide stiffened plate under combined compression and
shearing stresses. ...........................................................................................................................................48
Fig. 4.32 Analytically predicted elastic and inelastic interaction curves for an integrally stiffened panel. .................49
Fig. 4.33 Idealized beam-column. ...............................................................................................................................50
Fig. 4.34 Plate subjected to axial and shear stress. ......................................................................................................51
Fig. 4.35 Function f1 in Eq. 4.5....................................................................................................................................52
Fig. 4.36 Functions f2 and f3 in Eq. 4.52. .....................................................................................................................53
Fig. 4.37 Buckling coefficients for orthotropic plates. ................................................................................................54
Fig. 4.38 Interaction curve for orthotropic plates. .......................................................................................................55
Fig. 4.39 Shear buckling coefficients for orthotropic plates. [Adapted from Johns (1971), printed with the permission
of Her Majesty's Stationery Office.]..............................................................................................................56
Fig. 4.40 Stability response of an I-beam in major axis bending using the finite strip method (a) half-wavelength vs.
buckling stress, (b) buckling mode shapes, (c) plate buckling coefficient (Hancock 1978)..........................59
Fig. 4.41 Plate buckling coefficients for plate assemblies (Kroll et al., 1943) ............................................................61

Tables
Table 4.1 Coefficients for plate buckling under longitudinal stress gradients.............................................................15
Table 4.2 Limiting Values of gamma for Transverse Stiffeners..................................................................................42
4.1 INTRODUCTION
Metal compression or flexural members (Fig. 4.1, excepting Fig. 4.1a) typically employ cross-
sectional shapes which may be idealized as a composition of elements, or flat plates. When a
plate element is subjected to direct compression, bending, shear, or a combination of these
stresses in its plane, the plate may buckle locally before the member as a whole becomes
unstable or before the yield stress of the material is reached. Such local buckling behavior is
characterized by distortion of the cross-section of the member. Contrary to the notion of buckling
being a sudden or discontinuous phenomena, the almost inevitable presence of initial out-of-
planeness (imperfections) results in a gradual growth of this cross-sectional distortion with no
sudden discontinuity in real behavior at the theoretical critical load.

Fig. 4.1 Compression or flexural members

The theoretical, or elastic critical local bucking load is not on its own a satisfactory basis
for design. Ultimate strength of plates may be less than the critical local buckling load due to
yielding, or may be in excess of the critical local buckling load due to beneficial post-buckling
reserve. For example, a slender plate loaded in uniaxial compression, with both longitudinal
edges supported, will undergo stress redistribution as well as develop transverse tensile
membrane stresses after buckling that provide post-buckling reserve. Thus, additional load may
often be applied without structural damage. Initial imperfections in such a plate may cause
deformations to begin below the buckling load, yet unlike an initially imperfect column, the plate
may sustain loads greater than the theoretical buckling load. Despite its limitations the critical
local buckling load typically forms the basis for an initial evaluation of plates and is the focus of
the first section of this Chapter.
After considering elastic local bucking of flat plates in Section 4.2, the inelastic buckling,
post-buckling, and ultimate strength of flat plates is discussed in Section 4.3. Particular attention
is paid to the effective width method, which is an approximate method for ultimate strength
determination which is still in wide use in design today. Flat plates are often augmented with
stiffeners, both longitudinally and transversally, these stiffeners improve and complicate the
response as detailed in Section 4.4. One common approximation for stiffened plates is to treat the
stiffened plate as a flat plate with orthotropic properties, this approximation is detailed in Section
4.5. Finally, in Section 4.6, the concluding section of this chapter, the interaction of plate
elements which make up cross-sections such as those of Fig. 4.1 are discussed along with the
three modes of cross-section instability: local, distortional, and global bucking inherent to most
thin-walled members. Significant additional material related to the topics examined in this
chapter are provided in Chapters 6, 7 and 13 of this Guide.
4.2 ELASTIC LOCAL BUCKLING OF FLAT PLATES
An examination of the buckling behavior of a single plate supported along its edges is an
essential preliminary step toward the understanding of local buckling behavior of plate
assemblies. The buckling stresses are obtained from the concept of bifurcation of an initially
perfect structure. In practice, the response of the structure is continuous, due to the inevitable
presence of initial imperfections. Thus the critical stress is best viewed as a useful index to the
behavior, as slender plates can continue to carry additional loads well after initial buckling. Post-
buckling and strength of plates is discussed in Section 4.3.

When the member cross section is composed of various connected elements (see Fig. 4.1)
a lower bound of the critical stress can be determined by assuming, for each plate element, a
simple support condition for each edge attached to another plate element or a free condition for
any edge not so attached. The smallest value of the critical stress found for any of the elements is
a lower bound of the critical stress for the cross section. This lower bound approximation may be
excessively conservative for many practical design situations, more thorough methods are
discussed in Section 4.3.8.

4.2.1 Uniform Compression

Long Rectangular Plates. In 1891, Bryan (1891) presented the analysis of the elastic critical
stress for a rectangular plate simply supported along all edges and subjected to a uniform
longitudinal compressive stress. The elastic critical stress of a long plate segment is determined
by the plate width-to-thickness ratio b/t, by the restraint conditions along the longitudinal
boundaries, and by the elastic material properties (elastic modulus, E, and Poisson’s ratio ν). The
elastic critical stress, σc, is expressed as
π 2E
σc = k (4.1)
12(1 − v 2 )(b / t ) 2
in which k is a plate buckling coefficient determined by a theoretical critical-load analysis; k is a
function of plate geometry and boundary conditions such as those shown in Fig. 4.2.
 Fig. 4.2 Local plate buckling coefficient, k of Eq. 4.1, for plates in compression
with varied boundary conditions

Short Plates. When a plate element is relatively short in the direction of the compressive stress
their may exist an influence in the elastic buckling stress due to the fact that the buckled half-
waves which take integer values are forced into a finite length plate. Fig. 4.3 demonstrates how k
varies as a function of normalized plate length; the variation is a function of the plate boundary
conditions and the loading. Full analytical solutions for k as a function of a/b and m may be
found in Timoshenko and Gere (1961), Allen and Bulson (1980) and others. When a plate
element is very short in the direction of the compressive stress (i.e. a/b « 1), the critical stress
may be conservatively estimated by assuming that a unit width of plate behaves like a column.
 9

8
fix
ss ss
7
m=1 2 3 4 5 6 7 8 9 10 fix
6

5 ss
k ss ss
4 ss
m=1 2 3 4 5 6 7
3
fix
ss ss
2
free
ss
1 m=1 2 3 4
ss ss
m=1 free
0
0 2 4 6 8 10
β
Fig. 4.3 Plate buckling coefficient, k, as a function of normalized plate length (β=a/b) for different boundary
conditions, m=number of buckled half-waves along the length of the plate (after Yu and Schafer (2007))

4.2.2 Compression and Bending

When compression plus bending loads are applied to a structural member, plate elements of the
member can be subjected to in-plane stresses which vary along the loaded edges of the plate,
from a maximum compressive stress, σ1, to a minimum stress, σ2, as shown in Fig. 4.4. For this
situation, elastic critical plate stresses are dependent on the edge-support conditions and the ratio
of bending stress to uniform compression stress. Long plate values of kc that can be substituted
for k in Eq. 4.1 are tabulated in Fig. 4.4 for several cases. For plates with a free edge the kc
values vary slightly with Poisson’s ratio.
 Fig. 4.4 Plate buckling coefficients for long plates under compression and bending
(Brockenbrough and Johnston, 1974; Bijlaard, 1957).

Beyond the k values provided in Fig. 4.4 closed-form expressions also exist for the compression
and bending case. For plates simply supported on all four sides k may be found based on the
work of Peköz (1987) to be:
k c = 4 + 2(1 −ψ ) + 2(1 −ψ )
3
(4.2)
where ψ = σ1/σ2. For plates with one longitudinal edge free Bambach and Rasmusen (2004a)
provide a series of solutions summarized in Fig. 4.5. These closed-form expressions for k are
employed in the AISI (2007) Specification. Diagrams for buckling coefficients for rectangular
plates under combined bending and compressive stresses in two perpendicular directions are
given by Yoshizuka and Naruoka (1971).
 (a) compression gradient, free edge f2 (b) compression gradient, free edge f1

(c) bending, free edge in compression (d) bending, free edge in tension
notes: ψ=|f2/f1|, f1 > f2
Fig. 4.5 Plate buckling coefficients for unstiffened elements in compression and bending,
after (Bambach and Rasmussen 2004a)

4.2.3 Shear
When a plate is subjected to edge shear stresses as shown in Fig. 4.6, it is said to be in a state of
pure shear. The critical shear buckling stress can be obtained by substituting τc and ks, for σc and
k in Eq. 4.1, in which ks is the buckling coefficient for shear buckling stress. Critical stress
coefficients, ks, for plates subjected to pure shear have been evaluated for three conditions of
edge support. In Fig. 4.9 these are plotted with the side b, as used in Eq. 4.1, always assumed to
be shorter than side a. Thus α is always greater than 1 and by plotting ks in terms of 1/α, the
complete range of ks can be shown and the magnitude of ks remains manageable for small values
of α. However, for application to plate-girder design it is convenient to define b (or h in plate-
girder applications) as the vertical dimension of the plate-girder web for a horizontal girder.
Then α may be greater or less than unity and empirical formulas for ks together with source data
are as follows:

Plate Simply Supported on Four Edges. Solutions developed by Timoshenko (1910), Bergmann
and Reissner (1932), and Seydel (1933) are approximated by Eqs. 4.3a and 4.3b, in which α =
a/b:
⎧ 5.34 (4.3a)
⎪⎪ 4.00 + α 2 for α ≤ 1
ks = ⎨
⎪5.34 + 4.00 (4.3b)
for α ≥1
⎪⎩ α 2
 Fig. 4.6 Plate buckling coefficients for plates in pure shear. (Side b is the short side.)

Plate Clamped on Four Edges. In 1924, Southwell and Skan obtained ks = 8.98 for the case of
the infinitely long rectangular plate with clamped edges. For the finite-length rectangular plate
with clamped edges, Moheit (1939) obtained
⎧ 8.98 (4.4a)
⎪⎪ 5.6 + for α ≤ 1
ks = ⎨
α2
⎪8.98 + 5.6 (4.4b)
for α ≥ 1
⎪⎩ α2
Plate Clamped on Two Opposite Edges and Simply Supported on the Other Two Edges. A
solution for this problem has been given by Iguchi (1938) for the general case, and by Leggett
(1941) for the case of the square plate. Cook and Rockey (1963) later obtained solutions
considering the antisymmetric buckling mode which was not considered by Iguchi. The
expressions below were obtained by fitting a polynomial equation to the Cook and Rockey
results as shown in Fig. 2.36 of the book by Bulson (1970).
For long edges clamped,
 ⎧ 8.98 (4.5a)
⎪⎪ α 2 + 5.61 − 1.99 α for α ≤ 1
ks = ⎨
⎪8.98 + 5.61 − 1.99 (4.5b)
for α ≥ 1
⎪⎩ α 2
α 3

and for short edges clamped,

⎧ 5.34 2.31 (4.6a)
⎪⎪ α 2 + − 3.44 + 8.39α for α ≤ 1
ks = ⎨
α
⎪5.34 + 2.31 − 3.44 + 8.39 (4.6b)
for α ≥ 1
⎪⎩ α α2 α3
Curves for α ≥ 1 are plotted in Fig. 4.6. Tension and compression stresses exist in the plate, equal
in magnitude to the shear stress and inclined at 45°. The destabilizing influence of compressive
stresses is resisted by tensile stresses in the perpendicular direction, often referred to as ‘tension
field action’ in the literature. Unlike the case of edge compression, the buckling mode is
composed of a combination of several waveforms and this is part of the difficulty in the buckling
analysis for shear.

4.2.4 Shear and Compression

The case of shear combined with longitudinal compression, with all sides simply supported, was
treated by Iguchi (1938). His results are approximated by the following interaction equation, also
shown graphically in Fig. 4.7:
2
σc ⎛τc ⎞
+⎜ ⎟ =1 (4.7)
σ c* ⎝ τ c* ⎠
where σ c* and τ c* denote critical stress, respectively, under compression or shear alone.
 Fig. 4.7 Interaction curve for buckling of flat plates under shear and uniform compression.

Equation 4.7, shown in Fig. 4.7, is for ratios of a/b greater than unity. Batdorf and
associates (Batdorf and Houbolt, 1946; Batdorf and Stein, 1947) have shown that when the
loaded side b is more than twice as long as a, Eq. 4.29 becomes overly conservative. This
situation is the exception in actual practice, and Eq. 4.7 may be accepted for engineering design
purposes.

4.2.5 Shear and Bending

For a plate simply supported on four sides, under combined bending and pure shear, Timoshenko
(1934) obtained a reduced kc as a function of τc/ τ c* for values of α = 0.5, 0.8, and 1.0, where τc is
the actual shearing stress and τ c* is the buckling stress for pure shear. This problem was also
solved by Stein (1936) and Way (1936), whose results for four values of α are plotted in Fig. 4.8.
Chwalla (1936a) suggested the following approximate interaction formula, which agrees well
with the graphs of Fig. 4.8.
2 2
⎛ σ cb ⎞ ⎛ τ c ⎞
⎜ * ⎟ +⎜ * ⎟ =1 (4.8)
⎝ σ cb ⎠ ⎝ τ c ⎠
For a plate simply supported on four sides, under combined bending and direct stress at
the ends (of dimension b), combined with shear, an approximate evaluation of the critical
combined load is obtained by use of a three-part interaction formula, Eq. 4.9 (Gerard and Becker,
1957/1958).
 2 2
σ c ⎛ σ cb ⎞ ⎛ τ c ⎞
+⎜ ⎟ +⎜ ⎟ =1 (4.9)
σ c* ⎝ σ cb* ⎠ ⎝ τ c* ⎠
The foregoing problem, with the further addition of vertical compressive force along the
top and bottom edges of length a, has been treated by McKenzie (1964) with results given in the
form of interaction graphs. The results are in good agreement with the special case of Eq. 4.9.
Interaction equation 4.9, valid when a/b is greater than unity, is shown graphically in Fig. 4.9, as
presented in Brockenbrough and Johnston (1974).

Fig. 4.8 Buckling coefficients for plates in combined bending and shear.
 Fig. 4.9 Interaction curve for buckling of flat plates under shear, compression, and bending.

4.2.6 Biaxially Compressed

Pavlovic and Baker (1989) presented an exact solution for the stability of a rectangular plate
under biaxial compression. The case when both longitudinal and transverse stresses were
uniform was used as a starting point to investigate the much more complex problem of partial
loading on two opposite edges. A parametric study was carried out covering different plate
geometrics, load ratios, and varying edge lengths over which the applied load acted. The limiting
cases of very long and very wide plates were considered in the study, as was the problem in
which two opposite edges were subjected to concentrated forces.

4.2.7 Longitudinally varying Compression

When members such as those of Fig. 4.1 are used as beams they are typically exposed to a
moment gradient. Under a moment gradient, the plate which compromises the compression
flange is subject to a longitudinal variation in the compression force, as shown in Fig. 4.10.
 τ
τ o x τ
σ min σ max
simply supported and b
elastically restrained
a

τ
y
Fig. 4.10 Simply supported plate subject to a longitudinal stress gradient
(note shear stresses are required for equilibrium, after (Yu and Schafer 2007))

This case was first treated by Libove, Ferdman et al. (1949), recently Yu and Schafer (2007)
provided a closed-form expression for k:
α 1r + α 2
k = k∞ + (4.10)
α 3r + α 4 + β α5
where, k∞ is the traditional plate buckling coefficient (i.e., k for pure compression as the plate
length tends to infinity), r is the stress gradient (σmin/σmax), β is the plate aspect ratio (a/b), and
α1 through α5 are empirical coefficients dependent on the plate boundary conditions along the
unloaded longitudinal edges. Table 4.1 summarizes those coefficients.

Table 4.1 Coefficients for plate buckling under longitudinal stress gradients
k∞ α1 α2 α3 α4 α5
ss-ss 4.000 -1.70 1.70 0.20 -0.20 0.75
fix-fix 6.970 -2.20 2.20 0.20 -0.20 0.65
ss-free 0.425 -0.80 1.00 0.00 -0.60 0.95
fix-free 1.277 -0.60 0.60 0.00 -0.65 0.60
Note: applicable for 1 < β < 30 and -1 < r < 1, ss = simply supported

4.2.8 Plates on one-way foundations

Often, one way considered to constrain local buckling in metal plates is to have the thin plate
adjacent to a foundation of some sort, for example thin steel plate may be adjacent to a concrete
beam, or the bottom flange of a beam may rest against a floor, foundation, or the ground. If the
foundation is two-way (i.e. works in tension and compression) then the buckling load essentially
follows the foundation stiffness, and traditional methods such as those of Timoshenko and Gere
(1961) may readily be used, the energy from the foundation simply needs to be included in the
formulation. The more common case is a foundation that is only engaged when the thin plate
buckles into the foundation, this is known as a tensionless or one-way foundation. The elastic
critical local buckling load for this case has seen some study, in particular the work of Smith,
Bradford et al. (1999) provide plate buckling coefficients, k, for one-way foundations in
compression, bending, shear, and combinations thereof. An important practical finding that may
be gleaned from Shahwan and Waas (1998) is that the increase in the elastic critical local
buckling load, even for a rigid one-way foundation is limited. For instance, in uniform
compression the increase in k due to a rigid one-way foundation is only to 5.33 (from 4.0). Thus
suggesting that one-way foundations only provide a limited relief from local buckling.
4.2.9 Singly curved plates
It is not uncommon that one of the members of Fig. 4.1 may be curved either in plan or profile.
In such a case the plates making up the cross-section are singly curved. Featherston and Ruiz
(1998) provide a summary of the fundamental work in this area as well as additional testing to
verify the available plate buckling coefficients. A common application for this situation is the
horizontally curved steel I-girder treated in Chapter 9 of this Guide. Recent work on elastic
buckling of plates in I-girders (Davidson et al. 1999) is detailed in Chapter 9.

4.2.10 Square Plate with a Central Hole

Designers frequently find it necessary to introduce openings in the webs of girders and other
large plate structures. The introduction of an opening changes the stress distribution within the
member and will, in many instances, also change the mode of failure. Buckling is a key aspect in
the behavior of thin perforated plates. This section covers the case of a square plate with a single
hole, while the following section covers the case of a rectangular plate with multiple holes.

Compression: Though limited in many respects the first problem to see significant study in this
area was that of a square plate with a central hole and either simply supported or clamped-edge
conditions, research includes: Levy et al. (1947), Kumai (1952), Schlack (1964), Kawai and
Ohtsubo (1968), and Fujita et al. (1970). The work indicates reductions in the plate buckling
coefficient of 25% from the unperforated k for holes on the order of 50% of the width of the
square plate. Square holes are shown to result in greater reductions than round holes of the same
diameter (Yang, 1969). Uniform stress loading is shown to be more critical than uniform
displacement loading on the member edges. Also, it has been demonstrated that by suitably
reinforcing the hole, it is possible to increase the critical stress beyond that of the unperforated
plate (Levy et al., 1947).

Shear: The buckling of a square plate with a central circular cutout has been examined by
Rockey et al. (1969) using the finite element method. The relationship between the buckling
stress of the plate and the relative size of the hole (d/b) was obtained for both simply supported
and clamped-edge conditions. Rockey’s work suggests a simple linear relationship between the
critical stress and the d/b ratio in the form
⎛ d⎞
τ cp = τ c ⎜1 − ⎟ (4.11)
⎝ b⎠
where τcp and τc are the critical stresses for the perforated plate and unperforated plate,
respectively. The relationship holds for both clamped and simply supported end conditions (Fig.
4.11). Shear buckling of square perforated plates was also investigated by Grosskurth et al.
(1976) using the finite element approach. They considered the case of uniform shear deformation
instead of uniform shear stress and obtained critical stresses that were in closer agreement with,
although higher than, the experimental values.
The behavior of plates with cutouts reinforced by a ring formed by a pressing process
was also studied by Rockey, both analytically and experimentally (Rockey, 1980). It was found
that the buckling stress increases with tr/t, the ratio of the depth of the lip to the plate thickness,
and the larger the hole, the greater must be the tr/t ratio to achieve a buckling strength equal to
that of the unperforated plate.
 Fig. 4.11 Simply supported and clamped plates with hole under shear loading (Rockey et al., 1969).

Combined loads: Elastic buckling of square plates with holes under combined loads:
compression, bending, and shear are considered by Brown and Yettram (1986).

4.2.11 Rectangular Plate with Multiple Holes

Compression: Recent research on elastic buckling of rectangular plates with holes under
compression loading provide a more nuanced picture than the early results on square plates
detailed in the previous section. In a square plate with a hole the buckling mode shape is
constricted to the width of the plate, in rectangular plates longer buckling mode shapes are
possible and likely. Research employing thin shell finite element analysis shows that the
presence of holes can either increase or decrease the critical elastic buckling stress as well as
change the length and quantity of the buckled half-waves, depending upon the quantity of hole
material removed relative to the size of the plate, as well as the hole spacing (May and Ganaba
1988, Brown 1990, Brown and Yettram 2000; El-Sawy and Nazmy 2001; Moen and Schafer
2006).
However, the research on square plates with holes, in particular Kawai and Ohtsubo
(1968), did lead to a useful approximation of elastic buckling stress for rectangular plates with
holes, based on assuming the strips adjacent to the hole act as unstiffened elements (i.e., plates
with one free edge) in compression (see Fig. 4.12). For a plate in uniform compression the
“unstiffened strip” buckling stress uses Eq. 4.1 with hA or hB replacing b and k=0.425 per Fig.
4.2. case 4.
 L

S/2 S Detail A
h

Plate with holes

Unstiffened strip “A”
hA

hB
Unstiffened strip “B”

Detail A

Fig. 4.12 Illustration of unstiffened strips adjacent to a hole

The unstiffened strip approximation is useful, but one must also consider the possibility
of buckling away from the hole. In addition the length of the hole, as well as its width, can
influence the results. Moen and Schafer (2008) provide expressions for the plate buckling
coefficient, k, that account for these effects. Further, their work also covers the case of a
rectangular plate with holes in which one longitudinal edge is supported and the other free.

Bending: The problem of a rectangular plate with holes in bending has seen little attention, in
part due to the fact that shear is often more critical in this situation. Moen and Schafer (2008)
provide plate buckling coefficients for stiffened and unstiffened elements under bending with
holes present along the length of the plate.

Shear: The problem of a long shear web with holes has been examined by Michael (1960); he
suggested semiempirical expressions for the critical stress in terms of d/a and a/b (notation
indicated in Fig. 4.16). These are plotted graphically in Fig. 4.16 and are applicable for the web
fixed along the top and bottom edges.
 Fig. 4.13 Critical shear stress for webs with holes (Michael, 1960).

Combined loading: Redwood and Uenoya (1979) have investigated the problem of webs with
holes subjected to combined bending and shear. They studied the problem of shear webs with
aspect ratios from 1.5 to 2.5 with circular or rectangular holes. They suggested a classic circular
interaction formula for τc and σcb (critical values of the maximum shear and bending stresses,
respectively) in the form:
2 2
⎛ τ c ⎞ ⎛ σ cb ⎞
⎜⎜ ⎟⎟ + ⎜⎜ ⎟⎟ = 1.0 (4.12)
⎝ τ cp ⎠ ⎝ σ cbp ⎠
in which τcp and σcbp are the pure shear and pure bending critical stresses of the plate with the
hole. These, in turn, can be expressed in terms of the corresponding critical stresses of plates
without holes (τ c* , σ cb* ) and the relative sizes of the holes with respect to the plate dimensions.
With the notation indicated in Fig. 4.17, the expressions for plates with rectangular holes take the
form
⎡ ⎛ a ⎞⎤
σ cbp = ⎢1.02 − 0.04 ⎜ ⎟ ⎥ σ cb* but ≤ σ cb* (4.13a)
⎣ H
⎝ ⎠⎦
⎡ ⎛ 2H ⎞ ⎛ a ⎞⎤
τ cp = ⎢1.24 − 1.16 ⎜ ⎟ − 0.17 ⎜ ⎟ ⎥ τ c
*
but ≤ τ c* (4.13b)
⎣ ⎝ h ⎠ H
⎝ ⎠⎦
and for circular holes,
σcbp = σ cb* (4.14a)
⎡ ⎛ 2R ⎞⎤
τ cp = ⎢1.15 − 1.05 ⎜ ⎟ ⎥ τ c* but ≤ τ c* (4.14b)
⎣ ⎝ h ⎠⎦
where R is the radius of the hole.
The values of σ cb* and τ c* can be obtained from a knowledge of the aspect ratio and
boundary conditions of the plate. For example, for a simply supported plate with an aspect ratio
of 2, these stresses are given with sufficient accuracy by the following expressions
 σ*cb = 23.90 σ c* (4.15a)
τ* = 6.59 σ c* (4.15b)
in which
2
π 2E ⎛t⎞
σ c* = 2 ⎜ ⎟ (4.15c)
12(1 − v ) ⎝ h ⎠

Fig. 4.14 Perforated rectangular plate under combined action of shear and bending.
4.3 INELASTIC BUCKLING, POST-BUCKLING, AND STRENGTH OF
FLAT PLATES
The elastic critical plate buckling stresses, or corresponding plate buckling coefficients (k’s),
provided in the previous section represent an important benchmark for understanding the
behavior of thin plates. However, such elastic critical buckling stresses do not directly indicate
the actual behavior that may occur in such a thin plate. In thin plates loaded to failure material
and geometric effects complicate the response.
It is common, though artificial, to use the elastic critical buckling stress as a benchmark
for delineating different forms of plate buckling: if material yielding occurs prior to the elastic
critical buckling stress this is known as inelastic buckling; strength at magnitudes greater than
the elastic critical buckling stress, and the associated deformations that occur under such loading,
are referred to as post-buckling and may be either elastic or inelastic. Finally, ultimate strength
refers to the maximum load the plate may carry, typically independent of deformation, which
may indeed be quite large.
Actual plate response under load is more complicated than the simple notions of inelastic
buckling and post-buckling, this is due in part to unavoidable imperfections. In an imperfect
plate out-of-plane deformations begin immediately upon loading, such deformations lead to
second order (geometrically nonlinear) forces (and strains) that must be accounted for throughout
the loading/deformation, and thus the notions of buckling and post-buckling are not definitively
distinct. Under load the stress field response of a thin plate is complicated and varies along the
length, across the width, and through the thickness of the plate. Residual stresses that may exist
in the plate further complicate the response. A plate with an applied stress well below the elastic
critical plate buckling stress may still have portions of the plate yielding; thus determining a
definitive regime where a plate enters inelastic buckling is difficult. For gradual yielding metals
(e.g., aluminum, stainless steel) the distinction between elastic and inelastic buckling becomes
even more difficult.
Currently, inelastic buckling, post-buckling, and the strength of thin plates (and plate
assemblages such as Fig. 4.1) are most robustly examined through the use of numerical methods
such as finite element analysis. Finite element models for stability critical structures are
discussed further in Chapter 21, but key considerations for plates include: the manner in which
shear in the plate is handled (namely Kirchoff vs. Mindlin plate theory), the material stress-strain
relation including residual stresses and strains, the yield criterion (von Mises is by far the most
common in metals), the hardening law (isotropic hardening is the most common for static
loading, but is inadequate if large strain reversals are present), magnitude and distribution of
geometric imperfections, inclusion of higher-order strain terms in the development of the plate
stiffness, enforcement of equilibrium on the deformed geometry, details of the boundary
conditions, and for finite elements the order of the elements and the discretization of the plate
both in terms of element density and element aspect ratio. Finite element analysis is not the only
method able to provide post-buckling and collapse analysis of plates, finite strip (Bradford and
Azhari 1995; Kwon and Hancock 1991; Lau and Hancock 1986; Lau and Hancock 1989), and
recently generalized beam theory (Goncalves and Camotim 2007; Silvestre and Camotim 2002)
have proven to be able to provide reliable solutions.
For typical design, fully nonlinear numerical collapse analysis of thin plates remains too
involved of a task; in this situation one turns to classical and semi-empirical approaches. These
design approximations are the focus of this section. In particular, the effective width method, has
wide use as an approximate technique for determining ultimate strength of plates that accounts
for inelastic buckling and post-buckling and is discussed in detail.

4.3.1 Inelastic buckling

The notion of “inelastic buckling” is an attempt to extend the elastic critical buckling
approximations of Section 4.2 to situations where material yielding has already occurred. Bleich
(1952) generalized the expression for the critical stress of a flat plate under uniform compressive
stress in either the elastic or inelastic range in the following manner:
π 2E η
σc = k (4.16)
12(1 − v 2 )(b / t ) 2
in which η = Et/E. This modification of Eq. 4.1 to adapt it to a stress higher than the proportional
limit is a conservative approximation to the solution of a complex problem that involves a
continuous updating of the constitutive relations depending on the axial stress carried (Stowell,
1948; Bijlaard, 1949, 1950).
In combined loading the work of Stowell (1949) and Peters (1954) suggest that the
inelastic buckling interaction is not the same as the elastic buckling interaction. Under combined
compressive and shear stress for loads applied in constant ratio Peters found that a circular
stress-ratio interaction formula as expressed by Eq. 4.17 was conservative and agreed better with
test results than Eq. 4.7 which was provided for elastic buckling interaction.
2 2
⎛ σc ⎞ ⎛ τc ⎞
⎜ * ⎟ +⎜ * ⎟ =1 (4.17)
⎝ σc ⎠ ⎝τc ⎠

4.3.2 Post-buckling
Post-buckling of plates may readily be understood through an analogy to a simple grillage
model, as shown in Fig. 4.15. In the grillage model the continuous plate is replaced by vertical
columns and horizontal ties. Under edge loading the vertical columns will buckle, if they were
not connected to the ties they would buckle at the same load and no post-buckling reserve would
exist. However, the ties are stretched as the columns buckle outward, thus restraining the motion
and providing post-buckling reserve.
In an actual plate the tension in the transverse ties is represented by membrane tension
and shear. Note also that the columns nearer to the supported edge are restrained by the ties more
than those in the middle. This too occurs in a real plate, as more of the longitudinal membrane
compression is carried near the edges of the plate than in the center. Thus, the grillage model
provides a working analogy for both the source of the post-buckling reserve and its most
important result, re-distribution of longitudinal membrane stresses.
 Fig. 4.15 Simple model for post-buckling of a flat plate in uniform compression (after AISI 2007)

Elastic post-buckling stiffness may be measured in terms of the apparent modulus of

elasticity E* (the ratio of the average stress carried by the plate to average strain) (see Fig. 4.16).
The values of E* for long plates (a»b) for some typical longitudinal edge conditions are given by
Allen and Bulson (1980). The values given below are sufficiently accurate up to twice the critical
stress.
Simply supported longitudinal edges.
Sides straight but free to move laterally:
E* = 0.5E (4.18a)
Sides free to move:
E* = 0.408E (4.18b)
Clamped longitudinal edges.
Sides straight but free to move laterally:
E* = 0.497E (4.18c)
One longitudinal edge simply supported, the other free
E* = 0.444E (4.18d)
 Fig. 4.16 Post-buckling stiffness of plates having simply supported edges (Allen and Bulson, 1980)

Another means for accounting for the loss in stiffness in the post-buckling regime is to
assume a correlation between the strength loss of the plate and the stiffness loss. The effective
width of the plate (as defined in the following section) is used in place of the actual width for
determining cross-section moment of inertia. This method is employed in the cold-formed steel
specification (AISI 2007) and discussed further in Chapter 13.
Nemeth and Michael (1990) presented an experimental study of the buckling and post-
buckling behavior of square and rectangular compression-loaded aluminum plates with centrally
located circular, square, and elliptical cutouts. The results indicated that the plates exhibit overall
trends of increasing buckling strain and decreasing initial post-buckling stiffness with increasing
cutout width. Results showed that the reduction in initial post-buckling stiffness due to a cutout
generally decreases as the plate aspect ratio increases. Also, the square plates with elliptical
cutouts having large cutout width/plate width ratio generally lose pre-buckling and initial post-
buckling stiffness as the cutout height increases.

4.3.3 Strength and Effective Width

4.3.3.1 Uniform Compression
Local buckling causes a loss of stiffness and a redistribution of stresses (see the previous
section). Uniform edge compression in the longitudinal direction results in a nonuniform stress
distribution after buckling (Fig. 4.17) and the buckled plate derives almost all of its stiffness
from the longitudinal edge supports. It is worth noting that the longitudinal stress distribution of
Fig. 4.17 is an idealization of the membrane stress at the cross-section where the buckling wave
is at maximum deformation; the longitudinal membrane stress varies along the length of the plate
and the bending stress at the face of the plate provide a wholly different picture than Fig. 4.17.
 Fig. 4.17 Longitudinal stress after buckling and definition of effective width

An important semiempirical method of estimating the maximum strength of plates is by

the use of the effective width concept. The fact that much of the load is carried by the region of
the plate in the close vicinity of the edges suggests the simplifying assumption that the maximum
edge stress acts uniformly over two “strips” of plate and the central region is unstressed (Fig.
4.17). Thus, only a fraction of the width is considered effective in resisting applied compression.
The concept of effective width is, however, not confined to calculation of post-buckling strength
of uniformly compressed plates but has become the means of allowing for local buckling effects
in columns, panels, or flexural members that have the dual function of supporting loads and
acting as walls, partitions, bulkheads, floors, or roof decking. In a plate structure, use of the
effective width leads to an effective cross section consisting of portions of members meeting
along a junction. It is near these junctions that the plates will begin to yield preceding failure.
The effective-width concept has been used in design specifications for many years.
Specifications of the AISI (2007), the Aluminum Association (2005), and the AISC (2005) all
permit the use of an effective width in the design of members having plate elements with b/t
ratios greater than the limits for full effectiveness. Though implementations of the effective
width method vary greatly for the three design specifications.
The effective-width concept seems to have had its origin in the design of ship plating
(Murray, 1946). It had been found that longitudinal bending moments in ships caused greater
deflections than those calculated using section properties based on the gross area of the
longitudinal members. Deflections could be calculated more accurately by considering only a
strip of plate over each stiffener having a width of 40 or 50 plate thicknesses as effective in
acting with the stiffeners in resisting longitudinal bending.
The advent of all-metal aircraft construction provided another opportunity for the use of
the effective-width concept, since it was advantageous to consider some of the metal skin
adjacent to stiffeners as being part of the stiffener in calculating the strength of aircraft
components. Cold-formed members used in steel buildings also provide useful applications of
stiffened-sheet construction. A thorough discussion of the effective-width concept as applied to
cold-formed steel design has been prepared by Winter (1983).
Tests by Schuman and Back (1930) of plates supported in V-notches along their unloaded edges
demonstrated that, for plates of the same thickness, increasing the plate width beyond a certain
value did not increase the ultimate load that the plate could support. It was observed that wider
plates acted as though narrow side portions or effective load-carrying areas took most of the
load. Newell (1930) and others were prompted by these tests to develop expressions for the
ultimate strength of such plates. The first to use the effective-width concept in handling this
problem was von Kármán (1932). He derived an approximate formula for the effective width of
simply supported plates, and in an appendix to his paper, Sechler and Donnell derived another
formula based on slightly different assumptions. Subsequently, many other effective-width
formulas have been derived, some empirical, based on approximate analyses, and some based on
the large-deflection plate-bending theory, employing varying degrees of rigor.
For plates under uniform compression, stiffened along both edges parallel to the direction
of the applied compression stress, von Kármán (1932) developed the following approximate
formula for effective width, based on the assumption that two strips along the sides, each on the
verge of buckling, carry the entire load:
⎡ π E ⎤
be = ⎢ ⎥t (4.19)
⎢⎣ 3(1 − v 2 ) σ e ⎥⎦
Combining Eqs. 4.19 and 4.1, for k = 4 (simple edge supports), the formula suggested by
Ramberg et al. (1939) is obtained (see Fig. Fig. 4.17 for notation):
be σc
= (4.20)
b σe
From Fig. 4.17, the average stress is
b
σ av = e σ e (4.21)
b
Substituting Eq. 4.20 into Eq. 4.21 with the edge stress equal to the yield stress (σe = σy) gives us
σ av = σ cσ y (4.22)
As a result of many tests and studies of post-buckling strength, Winter (1947) and Winter et al.
(1950) suggested the formula for effective width that was adopted in the 1946 through 1962
editions of the AISI specifications for cold-formed steel members:
be E ⎛ E t⎞
= 1.9 ⎜⎜1 − 0.475 ⎟ (4.23)
t σe ⎝ σ e b ⎟⎠
or, alternatively, in the form of Eq. 4.5,
be σc ⎛ σc ⎞
= ⎜⎜ 1 − 0.25 ⎟ (4.24)
b σe ⎝ σ e ⎟⎠
Equations 4.23 and 4.24 are basically the same as Eqs. 4.19 and 4.20, respectively, but include a
correction coefficient determined from tests and reflecting the total effect of various
imperfections, including initial deviations from planeness. Equation 4.23 was found to be
satisfactory also for austenitic stainless steel in the annealed and flattened condition (Johnson
and Winter, 1966) and for quarter- and half-hard type 301 stainless steel (Wang, 1969; Wang et
al., 1975).
Introducing the coefficient B = b / t σ e / E , Winter’s formula, Eq. 4.23, can be written as
 be 1.90 0.90
= − 2 (4.25)
b B B
A formula proposed by Conley et al. (1963) is nearly the same as that proposed by Winter and
can be expressed as
be 1.82 0.82
= − 2 (4.26)
b B B
A useful form of Eq. 4.25 or 4.26 is obtained by introducing the material yield strength into the
dimensionless parameter B. If B is defined as
σy b σy
B=B = (4.27)
σe t E
and if σav/σe is substituted for be/b, and both sides of the equation are multiplied by σe/σy, Eq.
4.26 can be written as
σ av 1.82 σ e 0.82
= − 2 (4.28)
σy B σy B
By introducing discrete values of B into Eq. 4.28, the relationships shown in Fig. 4.18 between
σav/σy and σe/σy for B values greater than 1.0 were determined. It was assumed that there would
be no loss of plate effectiveness for values of B ≤ 1.0 and thus the straight line from (0, 0) to
(1.0, 1.0) was drawn for B = 1.0. Lines of constant B when plotted fully are tangent to the B =
1.0 line, and only their upper portions are shown. Thus for any given strength level of plate steel,
a relationship between average stress after buckling and the maximum or edge stress of the plate
panel is established as a function of the actual b/t ratio. This relationship is valid for stiffened
plates in which the ratio of stiffener cross-sectional area to plate-panel cross-sectional area is
small. If the cross section of a structural member includes a buckled plate, the effective-width
approach should be used in computing deflections, in determining the location of the neutral
axis, or in other calculations where the effective moment of inertia or radius of gyration of the
member is important.
 Fig. 4.18 Chart for determining σe/σy

In the 1968 and later editions of the AISI specification for cold-formed steel members
(e.g., AISI 2007) the coefficients in Eqs. 4.23 and 4.24 were reduced slightly, giving the
following expressions for effective width:
be E ⎛ E t⎞
= 1.9 ⎜⎜1.0 − 0.415 ⎟ (4.29)
t σe ⎝ σ e b ⎟⎠
or
be σc ⎛ σc ⎞
= ⎜⎜1.0 − 0.22 ⎟ (4.30)
b σe ⎝ σ e ⎟⎠
The limiting value of b/t when all of the width is considered to be effective is obtained by setting
be equal to b. The AISI value thus obtained from Eq. 4.14 is (b/t)lim = 221/ σ . AISC values of
effective width (2005) remain slightly more liberal than those of AISI. (See Section 4.3.3.5 for
further information on width-to-thickness limits.)
In the calculation of the ultimate compression load for plates supported along the two
unloaded edges, σe is taken equal to the compressive yield stress for steel. For aluminum alloys
and magnesium alloys, σe is taken as 0.7 times the yield strength, as determined by the offset
method. However, if the buckling stress σc exceeds 0.7 times the yield strength, the load capacity
as determined by inelastic plate-buckling analysis may be taken as btσc in which σc is determined
by Eq. 4.16 (inelastic buckling) and the effective width need not be calculated. The use of the
ultimate compressive buckling load in specifications for aluminum structures is discussed later in
this section.
Equation 4.30 can be used to determine a nondimensional ultimate-strength curve for
steel plates in the post-buckling range. The average stress on the plate at ultimate load, σav, is the
ultimate load divided by the total area. From Eq. 4.30,
 Pult ⎛ σ ⎞
σ av = = σ cσ y
⎜1.0 − 0.22 c ⎟ (4.31)
bt ⎜ σ y ⎟⎠
⎝
In Fig. 4.19 the average stress at ultimate load, by Eq. 4.31, is compared with the uniform-edge
compression stresses to cause buckling. A method for predicting the strength of simply supported
plates, taking into account initial out-of-flatness, is given by Abdel Sayed (1969) and Dawson
and Walker (1972). Out-of-flatness, residual stress, and strain hardening are considered by
Dwight and Ratcliffe (1968).
(delete equation numbers in figure or replace with new numbers 4.7Æ4.22, 4.16Æ4.31,4.18Æ4.33 (4.1 remains the same))

Fig. 4.19 Nondimensional buckling curves for plates under uniform

edge compression (adapted from Brockenbrough and Johnston, 1974).
For a plate supported along only one longitudinal edge, the effective width has been
experimentally determined by Winter as
bc σ ⎛ σ ⎞
= 1.19 c ⎜⎜ 1 − 0.30 c ⎟⎟ (4.32)
b σe ⎝ σe ⎠
This equation has also been confirmed by an analysis carried out by Kalyanaraman
(Kalyanaraman et al., 1977; Kalyanaraman and Peköz, 1978). The average stress at ultimate load
can then be expressed as
⎛ σ ⎞
σ av = 1.19 σ cσ y ⎜1 − 0.30 c ⎟ (4.33)
⎜ σ ⎟
⎝ y ⎠
Equation 4.33 is also shown in Fig. 4.19. This curve actually falls above and to the right of Eq.
4.31, but it must be remembered that in the elastic range σc for a plate supported on both
longitudinal edges is about eight times as large as that for the same plate supported along only
one longitudinal edge.
Considering that Eq. 4.29 is an appropriate formula for determining the effective design
width of stiffened compression elements with a k value of 4.0, a generalized formula for different
stiffened compression elements with various rotational edge restraints can be written as follows:
be kE ⎛⎜ kE t ⎞⎟
= 0.95 1 − 0.209 (4.34)
t σ e ⎜⎝ σ e b ⎟⎠
In fact Eq. 4.34 has also been extended to unstiffened elements. For example to determine the
effective width of a plate simply supported on only one longitudinal edge, the engineer would
use k=0.425 of Fig. 4.2 in Eq. 4.15. This basic approach where an appropriate k value is used to
determine the strength regardless of the boundary conditions (or loading) is known as the unified
method and has been shown to provide adequate strength predictions for a variety of conditions
(Peköz 1987) and forms the basis for the cold-formed steel specification discussed further in
Chapter 13.
Jombock and Clark (1962) list 14 effective-width formulas, along with their sources, and
discuss the assumptions on which they are based. Since the effective-width concept is also well
developed in current specifications and commentaries, it is suggested that reference be made
thereto for further information on this topic.
Certain box-girder-bridge structural failures in the early 1970s have interjected a warning
due to designers of plate structures that many aspects of plate behavior need further
investigation. For example, the concepts of effective width and average stress at failure consider
plate strains only up to the maximum plate capacity—that is, the ultimate load as characterized
by reaching the yield stress at the edge. Research has shown that inelastic strains beyond the
point may lead to a sudden and substantial reduction of the plate’s load-carrying capacity
(Dwight and Moxham, 1969; Dwight, 1971).
4.3.3.2 Compression and Bending
One approach to determining the ultimate strength of a plate in bending, or compression
and bending, is to use the plate buckling coefficients from Section 4.2.2 applied in Winter’s
effective width formula, i.e. Eq. 4.34. One difficulty that remains after determination of the
effective width, be, is how to distribute be to the plate edges. For example, for a simply supported
plate in bending if be/b is designated as ρ, Fig. 4.20 provides a rational method for distributing
the effective width based on maintaining force and moment equilibrium in the effective plate
based on the work of Schafer and Peköz (1999). Other alternatives exists, for example LaBoube
and Yu (1982), Peköz (1987), and Usami (1982). For plates with one longitudinal edge free the
plate buckling coefficients of Fig. 4.5 may be used with Eq. 4.34 to provide effective width
predictions. Distribution of the effective width is provided by Bambach and Rasmussen (2004b)
and has been adopted in AISI (2007).
A direct solution for the ultimate strength of plates under compression and bending
without recourse to effective width was studied and proposed by Rhodes and Harvey (1971,
1976, 1977). In the case of simply supported plates under eccentric loading P, the failure loads
for plates with various loading eccentricities can be accurately predicted by a simple expression
of the form (Rhodes and Harvey, 1977).
pb σ y + 11.4
Pult = 2 = (4.35)
π D b(e / b) + 0.85
where D = Et3/12(1 – v2), σ y = σ y (b 2t / π 2 D) , and e is the distance from the point of load
application to the remote edge of the plate.
 f1 ρ f1 f1
SS
½be The resultant
forces F1 and F2
stiffened element in
SS pure compression SS
b F1 F2 must maintain a
force balance:
F1=F2
SS ½be ρf1bt=f1bet
f1 ρ f1 f1

f1 ρ f1 f1
SS h1
The resultant
F1 F2 forces F1 and F2
stiffened element under
h2 must maintain a
SS a stress gradient SS
force and a
moment balance.
SS
f2 f2 f2
applied stress nonlinear stress linear approx. effective width
distribution due of nonlinear approximation at
to buckling stress full stress

Fig. 4.20 Distributing effective width in plates under a stress gradient, after (Schafer and Peköz 1999)

4.3.3.3 Longitudinally varying Compression

It has been demonstrated that the use of the plate buckling coefficients of Section 4.2.7 in
Winter’s effective width formula lead to reliable and conservative predictions of the strength of
plates under longitudinally varying compression, as occurs in the compression flange of
members subjected to moment gradient (Yu and Schafer 2006).
4.3.3.4 Plates with holes
The simplest means for determining the ultimate strength and effective width of plates with holes
is to adopt the “unstiffened strip” approach described in Section 4.2.11, in particular Fig. 4.12.
The strips of plate material adjacent to the hole are treated as plates with one longitudinal edge
free, i.e., unstiffened strips; the plate buckling coefficient may then be determined like any other
plate with one edge free (e.g., Fig. 4.2), and the effective width calculated using Winter’s
formula. This approach, which was developed by Vann (1971), Yu and Davis (1973), and Miller
and Peköz (1994), is essentially that adopted in AISI (2007).
4.3.3.5 Width-to-thickness limits
In plastic design of steel structures, it is necessary that the moment capacity of the member not
be impaired by local buckling until the required rotation is achieved. This can be achieved by
limiting the width-to-thickness ratios of elements that are vulnerable to local buckling in the
inelastic range. Such limiting width-to-thickness ratios have been proposed for flanges of I-
beams by Lay (1965). Further studies of this topic can be found in the following references:
Dawe and Kulak (1984, 1986), Kuhlmann (1989) and Kemp, (1996). Such provisions are also
available in design specifications (AISC 2005). Additional width-to-thickness limits have also
been adopted for seismic applications where cyclic inelasticity must be considered (AISC
2005b). (Earls 1999) shows that care must be taken in the use of these limits particularly as new
high strength materials are introduced and the buckling limit states may not follow the simple
empirical criteria previously employed.
4.3.4 Strength in shear and combined loadings with shear
The initial mode of buckling in pure shear, which takes the form of a half wave in the tension
direction and at least one full wave in the compression direction (Fig. 4.10a), undergoes a change
in the advanced post-buckling range, eventually taking the form of a family of diagonal folds
(Fig. 4.10b). These folds carry significant tensile stresses developed in the post-buckling range
and the displacement pattern is called a tension field. The dominance of this tension field
behavior has lead to methods other than effective width for determining the strength of thin
plates in shear.
The maximum shear load that can be applied before failure occurs due to a breakdown of
the material in the tension field, and it is influenced by the rigidity of the edge members
supporting the plate. This problem is dealt with in greater detail in Chapter 6 for bridge
applications and is discussed further in Section 4.3.8 below within the context of steel plate shear
walls for buildings.
For a plate with infinitely stiff edge members, the maximum shear strength can be
estimated by the formula (Allen and Bulson, 1980)
Vu = τ c bt + 12 σ ty bt (4.36)
where
σ ty = σ y2 − 0.75τ 2c − 1.5τ c provided that τ c σ y τc<<σy (4.37)
Stein and Manue (1989) presented buckling and post-buckling results for plates loaded by in-
plane shear. The buckling results had been plotted to show the effects of thickness on the stress
coefficient for aluminum plates. Results were given for various length-to-width ratios. Post-
buckling results for thin plates with transverse shearing flexibility were compared to results from
classical theory. The plates were considered to be long with side edges simply supported, with
various in-plane edge conditions and the plates were subjected to a constant shearing
displacement along the side edges.
Elangovan and Prinsze (1992) carried out a numerical investigation, using a finite
element buckling analysis to determine the critical shear stress of flat rectangular plates with two
opposite edges free. Plate sizes and the boundary conditions at the two edges loaded in shear
were the parameters considered in their study. Results showed a considerable difference in the
buckling strength of plates if the in-plane displacement normal to the loaded edges were
restrained either at one or at both of those edges.
 Fig. 4.21 Shear and tension fieelds (square plate).

Information on the post-buckling strength of plate elements subjected to the combined

action of shear and compression is limited. A semiempirical method for the determination of
stress levels at which permanent buckles occur in a long plate with simply supported edges under
the combined action of uniform axial compression and shear has been suggested by Zender and
Hall (1960). Additional information on post-buckling strength of plates subjected to the
combined action of shear and bending can be found in Chapters 6 and 7.

4.3.5 Strength of biaxially compressed plates

Few simplified design methods exist for post-buckling and strength prediction of
biaxially compressed plates. For example, the effective width method for plates under biaxial
compression has seen little study. Isami and Hidenori (1989) employed mechanism analysis
methods (see Section 4.3.7) to provide predictions for biaxially compressed plates. Post-buckling
and strength prediction relies primarily on geometric and material nonlinear finite element
analysis. For example biaxial compression of ship plates with particular attention paid to the role
of initial imperfections in determining the ultimate strength is studied by (Paik et al. 2004). In
addition, they show that the collapse behavior of biaxially compressed plates can vary
significantly from those under only longitudinal compression.

4.3.6 Average stress and strength of aluminum plates

As an alternative to the effective width concept for wide, thin plates, another approach is to use
the average stress at failure and the actual (unreduced) plate width. This is the basis for allowable
stresses on thin sections in the Aluminum Association specifications (AA, 1994). In applying
these specifications the designer does not, in general, calculate an effective width but uses
instead an allowable stress that has been derived by applying a factor of safety to the average
stress at failure for plate elements. For plates that buckle in the inelastic stress range, the average
stress at failure is considered to be the same as the local-buckling stress, since plates of these
proportions have little post-buckling strength (Jombock and Clark, 1968). Inelastic local-
buckling strength for aluminum plates is represented in the specifications by the following
straight-line approximation to Eq. 4.1 (Clark and Rolf, 1966):
b
σ c = B p − D pκ1 (4.38)
t
where
⎡ (σ )1/ 3 ⎤
B p = σ y ⎢1 + y ⎥ (4.39)
⎢⎣ k2 ⎥⎦
( B p )3/ 2
Dp = (4.40)
k3 ( E )1/ 2
12(1 − v 2 )
k1 = (4.41)
k
For aluminum alloys that are artificially aged (temper designations beginning with T5,
T6, T7, T8, or T9), k2 = 11.4 ksi1/3 (78.6 MPa1/3) and k3 = 10. For other aluminum alloys (temper
designations beginning with 0, H, T1, T2, T3, or T4), k2 = 7.6 ksi1/3 (52.4 MPa1/3) and k3 =
10 2 / 3 . Equation 4.20 has been shown to agree well with the results of tests on aluminum plate
elements (Clark and Rolf, 1966; Jombock and Clark, 1968). For plates that buckle elastically, the
average stress at failure is represented for purposes of the aluminum specification as
σ av = σ eσ c (4.42)
Equation 4.42 corresponds to Eq. 4.22. Jombock and Clark (1968) demonstrated that the edge
stress at failure σe for aluminum plates could be represented by a function of the intercept Bp in
Eq. 4.38. This results in a simple relationship between the ultimate-strength curves
corresponding to elastic and inelastic buckling. Generally, this edge stress at failure for
aluminum alloys is about 0.7σy.
The formulas used in the Aluminum Association specifications (AA, 1994) are illustrated
in Fig. 4.7. Comparisons with test results were published by Jombock and Clark (1968). The use
in these specifications of the average stress at failure for thin sections results in some
simplification, since the designer does not have to calculate an effective width. However, it
sacrifices some of the flexibility of the effective-width approach. For example, the average
stress-at-failure method does not treat the change in moment of inertia of a member when its
compression elements are in the post-buckling range, and hence does not readily lend itself to
calculation of deflections. Therefore, the Aluminum Association specifications include an
effective-width formula to be used in calculating deflections in the post-buckling range. See
Chapter 13 for further discussion.

Fig. 4.22 Equations for buckling stress and ultimate strength of plates used in
Aluminum Association specifications (AA, 1994).

The Aluminum Association specifications (AA, 1994) treat post-buckling strength of

webs in bending by means of an average stress approach similar to that used for plates in
compression. This approach is compared with test results by Jombock and Clark (1968).

4.3.7 Yield-line analysis and plastic buckling of plates

Yield-line analysis represents the extension of concentrated plasticity such as plastic hinge
analysis in beams to two-dimensional structures. For out-of-plane loading (applied loads on floor
slabs, etc) yield-line analysis has a long history; however for in-plane loading such as that which
has been the focus of this chapter the use is less widespread. One of the primary difficulties with
the extension to in-plane loads is that yielding mechanisms for in-plane loaded plates typically
involve yielded zones (membrane extension), as well as yield lines (where bending and twist)
occur. Successful applications include connection design, particularly for tubular and HSS
sections (Kosteski et al. 2003), determination of the limiting rotation for beams (Gioncu and
Petcu 1997; Shi et al. 1996), and patch loading or crippling of webs. Use of yield-line analysis
for generalized collapse analysis of plates is not widespread, Bambach (2008) represents a recent
summary and the most complete treatment still remains Murray and Khoo (1981), Murray
(1984).
Inoue et al. (1993) studied an analytical evaluation of the effective plastic shear modulus
of fully yielded steel plates at the instant of buckling. It was assumed that yielding of steel was to
follow the Tresca yield criterion and that plastic deformation of a steel plate was to be caused by
slips. The Tresca yield criterion provides lower bending stiffnesses than those obtained from the
von Mises yield criterion, but it does not lower the plastic shear modulus of the material at any
point on the yield plateau. They proposed a new theory that assumes a nonuniform distribution of
slips depending on the orientation of an infinite number of possible slip planes at each point in
the plate. The twisting of the plate is then accompanied by distortion of its sectional shape, and
this mode of buckling is shown to provide a considerable reduction in the effective plastic shear
modulus. Applying these sectional stiffnesses and solving differential equilibrium equations
leads to a lower bifurcation strength, which provides much better correlations with experimental
results than those of previous predictions.

4.3.8 Energy dissipation (steel plate shear wall)

The collapse of metal plate structures generally involve either fracture or stability limit states.
Stability limit states, such as inelastic local plate buckling are typically preferred over fracture
limit states, because they provide some post-buckling resistance, and hence greater ductility and
the potential for energy dissipation. For seismic design energy dissipation is a fundamental need
for most structures, rather than rely on inherent or assumed energy dissipation, modern designs
often attempt to explicitly use specific element(s) for energy dissipation. Properly designed and
detailed a thin plate undergoing post-buckling may provide reliable energy dissipation.
Steel plate shear walls are an important example of the application of metal plate
structures for the purposes of energy dissipation. Utilizing the post-buckling response of plates in
the shear experimental tests under cyclic load have been used to verify the potential energy
dissipation of unstiffened steel plate shear walls, as shown in Fig. 4.23. (Caccese et al. 1993)
(Elgaaly et al. 1993) (Elgaaly 1998) (Lubell et al. 2000). Currently phenomenological models are
favored for characterizing the response of steel plate shear walls, while such models have
limitations, they are readily implemented in nonlinear dynamic analysis allowing the influence of
steel plate shear walls on full structural systems to be studied (Roberts and Sabouri-Ghomi 1992;
Sabouri-Ghomi and Roberts 1992). Recent and ongoing researcher indicates that perforated steel
plates may offer further advantages for energy dissipation as shown in Fig. 4.24 (Hitaka and
Matsui 2003).

Fig. 4.23 Hysteretic response of a steel plate shear wall under cyclic loading (Lubell et al. 2000)
Fig. 4.24 Hysteretic response of steel plate shear walls with cutouts (Hitaka and Matsui 2003)
4.4 BUCKLING, POST-BUCKLING AND STRENGTH OF STIFFENED
PLATES
This section deals with the buckling and ultimate strength of stiffened flat plates under various
combinations of loadings. The behavior of the stiffened plate as a unit is emphasized rather than
the stability of its individual elements. However, in design, the structural properties of all of the
components must be considered.
Stiffened plates can fail through instability in essentially two different ways. In one,
overall buckling (also referred to as stiffener buckling or distortional buckling in the literature),
the stiffeners buckle along with the plate, and in the other, local buckling, the stiffeners form
nodal lines and the plate panels buckle between the stiffeners. In either case the stiffness of the
combination may be such that initial buckling takes place at fairly low stress levels.
Nevertheless, a significant amount of post-buckling strength may remain in the stiffened plate,
provided that proper attention is given to the design and fabrication of the structural details. A
great deal of information on this subject can be found in the Handbook of Structural Stability
(CRCJ, 1971) and in a book by Troitsky (1976).
Finally, as discussed in the introduction to Section 4.3, today it is becoming increasingly
common to study buckling, post-buckling, and strength through geometric and material nonlinear
finite element analysis using plate or shell elements. Such models provide the capability to
accurately include boundary conditions, imperfections, residual stresses, etc. and are a powerful
too in the study of stiffened plates. For the most part, the discussion provide herein focuses on
analytical methods and tools that can be used with far greater expediency than a complete
nonlinear finite element analysis. For further discussion on stiffened plates see Chapters 6,7 and
13, and for further discussion on finite element modeling for stability problems see Chapter 21.

4.4.1 Compression: Buckling

It is often economical to increase the compressive strength of a plate element by introducing
longitudinal and/or transverse stiffeners (Fig. 4.25). In the following paragraphs methods are
presented for determining the compressive strength of stiffened plate panels. The edges of the
plate are assumed to be simply supported in all cases, and it is also assumed that individual
elements of the panel and stiffeners are not subject to instability.

Fig. 4.25 Stiffened plate panels; (a) panel with longitudinal stiffeners; (b) panel with transverse stiffeners.
4.4.1.1 Longitudinal Stiffeners
Seide and Stein (1949), Bleich and Ramsey (1951), and Timoshenko and Gere (1961) have
presented charts and tables for determining the critical stress of plates simply supported on all
edges and having one, two, or three equally spaced longitudinal stiffeners parallel to the direction
of the applied compressive load. The solutions of Seide and Stein are also useful for other
numbers of equally spaced stiffeners. In all of these solutions the stiffeners are assumed to have
zero torsional rigidity. If we consider only overall buckling with m=1 then this classical solution
may be provided in a compact form (Schafer and Peköz 1998) where:
π2D
f cr = k cr 2 (4.43)
bo t

(4.44)

(4.45)

EI i ( As )i ci Et 3
γi = δi = αi = D= (4.46)
bo D bo t bo (
12 1 −ν 2 )
and I is the stiffener moment of inertia, bo the plate width, As the stiffener area, and ci the
distance the stiffener is from the edge, this form of the expressions is adopted in AISI (2007).
A conservative method of analysis proposed by Sharp (1966) divides the analysis of the
stiffened plate into two parts: one applying to short panels in which the buckled configuration
takes the form of a single half-wave in both the longitudinal and transverse directions and
another applying to long panels in which several longitudinal waves may occur along with a
single half-wave in the transverse direction. In very short panels, the stiffener and a width of
plate equal to the stiffener spacing, d, are analyzed as a column of length, a, with a slenderness
ratio
⎛L⎞ α
⎜ ⎟ = (4.47)
⎝ r ⎠eq re
where re is the radius of gyration of the section consisting of a stiffener plus a width of plate
equal to d.
In long panels the critical stress is larger than that calculated by the use of Eq. 4.47. In
this case an equivalent slenderness ratio is defined for use in column strength formulas,
L b 1 + ( As / bt )
= 6(1 − v 2 ) (4.48)
req t 1 + ( EI e / bD) + 1
where
b = Nd = overall width of longitudinally stiffened panel
N = number of panels into which the longitudinal stiffeners divide the plate
Ie = moment of inertia of section consisting of the stiffener plus a width of plate equal to d
As = cross-sectional area of stiffener
 Et 3
D=
12(1 − v 2 )
The smaller of the values from Eqs. 4.47 and 4.48 is used in the analysis, and it is assumed that
the plate is fully effective over the panel width d. For greater values of d, buckling of the
stiffeners and of the plate between the stiffeners would need consideration.
A flat aluminum sheet with multiple longitudinal stiffeners, or a formed stiffened sheet,
subjected to a uniform longitudinal compression (Sherbourne et al., 1971) will buckle into waves
of length Ψb, in which b is the plate width and Ψ = 1.8(Ix/t3)1/4, where Ix is the moment of inertia
in the strong direction and t is the plate thickness. For a formed aluminum stiffened sheet this
becomes Ψ = 1.8(ρrx/t)1/2, in which ρ is the ratio of the developed sheet width to the net width,
and rx is the radius of gyration in the strong direction. If the spacing, a, of transverse supports is
less than Ψb, the elastic critical stress is given approximately by
σ c = π 2 E[1 + (a /ψ b) 4 ]
(4.49a)
(a / rx ) 2
If the spacing exceeds Ψb, then
σ c = 2π 2 E
(4.49b)
(ψ b / rx ) 2
The Canadian standard (CSA, 1983), using this method, reduces the design procedure to
determining the equivalent slenderness ratios, which are, respectively, for the two cases,
a / rx
λ= (4.50a)
[1 + (a /ψ b) 4 ]1/ 2
0.7ψ b
= (4.50b)
rx
4.4.1.2 Transverse Stiffeners
The required size of transverse stiffeners for plates loaded in uniaxial compression has been
defined by Timoshenko and Gere (1961) for one, two, or three equally spaced stiffeners, and by
Klitchieff (1949) for any number of stiffeners. The stiffeners as sized provide a nodal line for the
buckled plate and thus prohibit overall buckling of the stiffened panel. The strength of the
stiffened panel would be limited to the buckling strength of the plate between stiffeners. These
authors also give formulas for calculating the buckling strength for smaller stiffeners. The
required minimum value of γ given by Klitchieff is
(4m 2 − 1)[(m 2 − 1)2 − 2(m 2 + 1) β 2 + β 4 ]
γ= (4.51)
2m5m 2 + 1 − β 2α 3
where
α2 a EI
β= α= γ= s (4.52)
m b bD
and m is the number of panels, m – 1 the number of stiffeners, and EIs flexural rigidity of one
transverse stiffener. These stiffener sizes should be understood as ideal stiffeners, in actuality
unavoidable imperfections and the desire to avoid coupled instabilities will require designers to
use stiffeners greater than those of Eq. 4.51.
An approximate analysis that errs on the conservative side but gives estimates of the
required stiffness of transverse stiffeners for plates, either with or without longitudinal stiffeners,
may be developed from a consideration of the buckling of columns with elastic supports.
Timoshenko and Gere (1961) show that the required spring constant, K, of the elastic supports
for a column, for the supports to behave as if absolutely rigid, is given by
mP
K= (4.53)
Ca
where
m 2π 2 ( EI )c
P= (4.54)
α 2

and C is the constant depending on m, which decreases from 0.5 for m = 2 to 0.25 for infinitely
large m, (EI)c the flexural stiffness of column, m the number of spans, and a the total length of
column. In the case of a transversely stiffened plate (Fig. 4.25b), a longitudinal strip is assumed
to act as a column which is elastically restrained by the transverse stiffeners. Assuming also that
the loading from the strip to the stiffener is proportional to the deflection of the stiffener, the
spring constant for each column support can be estimated. For a deflection shape of a half sine
wave the spring constant is
π 4 ( EI ) s
K= (4.55)
b4
Equating Eqs. 4.53 and 4.55 and inserting the value given for P results in the following:
( EI ) s m3
= 2 (4.56)
b( EI )c π C (a / b)3
In the case of panels without longitudinal stiffeners (EI)c = D and the left side of Eq. 4.56 is γ.
Values of C are tabulated by Timoshenko and Gere (1961) for m ≤ 11. As shown by Fig. 4.26,
these values are given approximately by
2
C = 0.25 + 3 (4.57)
m
 Fig. 4.26 Buckling of columns with elastic supports.

In Table 4.2 values calculated using Eq. 4.56 are compared to corresponding values
tabulated by Timoshenko and Gere (1961) for one, two, and three stiffeners and to those
calculated by Eq. 4.41 for 10 stiffeners. Equation 4.56 is always conservative and is highly
accurate for cases in which several stiffeners subdivide the panel.
Table 4.2 Limiting Values of gamma for Transverse Stiffeners

4.4.1.3 Longitudinal and Transverse Stiffeners.

For a combination of longitudinal and transverse stiffeners, Gerard and Becker (1957/1958) give
figures showing the minimum value of γ as a function of α for various combinations of equally
stiff longitudinal and transverse stiffeners. The same procedure used to establish Eq. 4.56 can
also be applied to this case. The difference in the development is that the spring constant, K, of
the support is dependent on the number of longitudinal stiffeners as illustrated below. The
flexural stiffness (EI)c in these cases is equal to the average stiffness per unit width of the plate-
longitudinal stiffener combination.
Number and Spacing of ( EI )
Longitudinal Stiffeners Spring Constant, K b( EI )c = γ
One centrally located 48( EI ) s 0.206m3
b3 C ( a / b )3
Two equally spaced 162 ( EI ) s 0.152m3
5 b3 C ( a / b )3
Four equally spaced 18.6( EI ) s 0.133m3
b3 C ( a / b )3
Infinite number equally spaced π 4 ( EI ) s m3 0.1013m3
=
b3 π 2C ( a / b)3 C ( a / b)3

The required size of transverse stiffeners in a panel also containing longitudinal stiffeners
is thus approximately
( EI ) s m3 ⎛ 1 ⎞
= 2 3 ⎜
1+ ⎟ (4.58)
b( EI )c π C (a / b) ⎝ N − 1 ⎠
This formula yields essentially the same values as does the formula presented for
aluminum panels in the Alcoa Structural Handbook (Aluminum Company of America, 1960).
With this size of transverse stiffener the strength of the panel is limited to the buckling strength
of the longitudinally stiffened panel between transverse stiffeners.
4.4.1.4 Stiffener Type
The methods of analysis described above are directly applicable to open-section stiffeners having
negligible torsional stiffness and are conservative when applied to stiffeners with appreciable
torsional stiffness. The influence of torsional stiffness on overall panel buckling has been studied
by Kusuda (1959) for the case of one longitudinal or one transverse stiffener. Stiffeners with
large torsional rigidity also provide partial or complete fixity of the edges of subpanel plating,
thereby increasing their critical stresses.
It has been shown by Lind (1973) and Fukumoto et al. (1977) that the stiffener type
affects the buckling mode as well as the ultimate carrying capacity of the stiffened plate. It has
also been shown by Tvergaard (1973), and Fok et al. (1977) that local imperfections of stiffeners
influence significantly the overall buckling behavior of stiffened plate panels.

4.4.2 Compression: Post-buckling and Strength

Buckling of a stiffened-plate panel may occur by primary instability with a half wavelength
which is on the order of the panel length, or by local instability with a half-wavelength which is
on the order of the width of plate elements of the plate and stiffeners. As plate panel length
increases, the ultimate stress decreases, until at large slenderness ratio the panel fails in flexure
as a long column. The ultimate strength in the long-column range can be predicted by the Euler
column formula, where the radius of gyration is computed for the combined section of the
stiffener and the effective width of the plating. At an intermediate slenderness ratio there is a
transition in the mode of failure from the purely local mode to one dominated by overall panel
failure. In the transition zone the panel fails through a combination of the primary buckling and
flexural modes and may involve stiffener twisting.
 It is well known that plates supported at their edges are often able to sustain compressive
load far in excess of their buckling loads. The margin between the buckling load and the ultimate
load in plates, known as the post-buckling strength, depends on whether the critical stress is
reached below or above the proportional limit of the material. If the buckling stress is well below
the proportional limit, the ultimate load may be many times greater than the buckling strength,
depending on the aspect ratio, a/b, of the plate element. As the buckling stress approaches the
yield strength of the material, the reserve strength in the post-buckling range approaches zero.
See Section 4.3 for further discussion of the distinctions between elastic buckling, inelastic
buckling, post-buckling and strength of plates.
Initial buckling modes of stiffened flat panels vary with the slenderness ratio of the panel
and with the type of construction: monolithic or built-up. In the following, as a basis for
evaluating post-buckling strength, a brief discussion is given regarding predictions of initial
buckling stresses and ultimate strength.
4.4.2.1 Local Instability Mode
The initial buckled form has one transverse sinusoidal half-wave, with perhaps a number of
longitudinal half-waves. As the compressive load is increased, the central portion of the
transverse half-wave becomes flattened, and, as shown in Fig. 4.27 the transverse deformation is
no longer a simple sinusoidal curve. The well-known effective-width approach to the analysis of
the post-buckling strength of a flat plate is based on the stress distribution associated with this
buckled form, as discussed in detail in Section 4.3.3.

Fig. 4.27 Initial and final buckled shapes

Another change of buckled form is possible when a rectangular flat plate, simply
supported on all sides, is subjected to uniform edge compression and is free to expand laterally.
A dynamic snap from one buckled form to another may occur by a sudden change of wavelength
of buckles along the direction of compression. The behavior of the flat plate in this sense is
analogous to the elastic post-buckling of a column supported laterally by a nonlinear elastic
medium (Tsien, 1942; Stein, 1959; Koiter, 1963). A column supported laterally by a finite
number of nonlinear elastic restraints buckles initially into m sinusoidal half-waves over its
length; then subsequently the buckled form may become unstable, and the column may snap into
n half-waves, with n being greater than m. The exact analysis of transition between the two
modes of buckling is not known at present. Sherbourne et al. (1971) determined the terminal
wavelength for flat plates at the ultimate capacity.
4.4.2.2 Failure Strength of Very Short Stiffened Panels
For a short stiffened plate with the slenderness ratio smaller than 20, Gerard and Becker
(1957/1958) note that the failure stress is independent of the panel length. The average stress at
failure in this slenderness range is known as the crippling, crushing, or local-failure stress, and
will be represented by σ f . Gerard and Becker (1957/1958) present a method for determining
the crippling stress of short longitudinally stiffened panels. However, for most cases, the yield
stress of the material can be considered the failure stress for short stiffened panels.
4.4.2.3 Ultimate Strength of Intermediate and Long Stiffened Panels
Gerard and Becker (1957/1958) describe a method to predict the buckling failure of stiffened
panels based on a curve analogous to the Johnson parabola shown in Fig. 4.28. At stresses lower
than the local buckling stress σc and the proportional limit σpl, the Euler column equation is used.
In the transition range between L/r = 20 and the long column, a parabola of the following form is
used:
⎡ σ ⎛ σ ⎞ ⎛ σ 1/ 2 − σ 1/ 2 ⎞ 2 ⎤
σ c = σ f ⎢1 − c ⎜⎜1 − c ⎟⎟ ⎜ 20 e
1/ 2 ⎟
⎥ (4.59)
⎢⎣ σ e ⎝ σ f ⎠ ⎝ σ 20 − σ c ⎠ ⎥⎦
1/ 2

where σ c is the failure stress, σe = π2 E/(L/r)2 the Euler stress for the panel, σ f the failure stress
for a short stiffened panel, and σ20 the Euler stress evaluated at L/r = 20. Many direct-reading
column charts have been prepared for panel ultimate strength. The type of plot is shown in Fig.
4.29. Gerard and Becker (1957/1958) have provided references and examples of this work.

Fig. 4.28 Column curve for stiffened panels.

Fig. 4.29 Column chart by Gerard and Becker (1957/1958).

In the determination of panel strength it is necessary to estimate the effective column

length of the plate-stiffener combinations as well as the effective width of plating that acts in
conjunction with the stiffener. When the critical stress for the individual panel of plating between
stiffeners is greater than the critical stress for the stiffened panel, the plating may be assumed
completely effective, and the effective column length of the panel is determined by the end
conditions. When the critical stress for the individual plating panel is significantly less than that
for the stiffened panel as a unit, the ultimate strength of the stiffened plate is considered to be the
lesser of (1) the load that causes the stress at the juncture of plate and stiffener to reach the yield
strength of the material, or (2) the column strength of the stiffener in conjunction with an
effective width of plating that is less than the actual width of plating between stiffeners.
The application of the first criterion for ultimate strength assumes that the stiffener is stiff
enough to allow the plate-edge stress to reach the yield stress before the stiffener buckles as a
column and that the stress distribution across the buckled plate can be determined, while the
second criterion assumes that the effective column length of the stiffener and the effective width
of the associated plating can be properly determined and that residual stresses and distortions due
to fabrication are properly accounted for with regard to column buckling. The effective-width
concept is treated in Section 4.2.
To provide a better understanding of the behavior of stiffened plates, a theoretical study
was carried out by Wittrick (1968). The collapse of box-girder bridges in the early 1970s
precipitated a great interest in research on various aspects of box girders, especially the
interactive buckling of an assembly of plates and the ultimate strength of a stiffened plate. The
research efforts in many ways were centered around the Merrison Committee (1973). Murray
(1975) reported on analysis and design procedure for the collapse load for the stiffened plates.
Crisfield (1975) presented a finite element formulation for the large-deflection elastic-plastic
full-range analysis of stiffened plating. A simple approach for the design of stiffened steel
compression flanges was proposed by Dwight and Little (1976).
A theoretical and experimental study on the inelastic buckling strength of stiffened plates
was reported by Fukumoto et al. (1977). Residual stresses were considered and the stiffened
plates treated had relatively low width-to-thickness ratios and with relatively rigid stiffeners. It
was found that partial yielding in the flat stiffeners considerably reduced the buckling strength of
a stiffened plate. In the case of plates stiffened by T-type stiffeners, the strength reduction due to
partial yielding in the stiffeners was much less pronounced. This is probably due to the fact that
the T-stiffeners were more rigid and less susceptible to initial imperfections than were the flat
stiffeners. Murray (1973) has indicated that based on experimental results there is often little
margin of strength above the load at which yielding first occurs in a stiffener. This phenomenon
can lead to a triggering effect which results in a sudden failure of the stiffened plate. Such a
viewpoint is also shared by Horne and Narayanan (1977).
Horne and Narayanan (1975, 1977) have proposed a design method for the prediction of
collapse loads of stiffened plates subjected to axial compression. They have compared the results
obtained from their method, methods proposed by other researchers, and experiments. The
results are all close to the observed strength. Horne and Narayanan’s method is based on the
British Perry–Robertson column formula and consists of analyzing the stiffened plates as a series
of isolated columns comprising a stiffener and the associated effective width of plating. The
criterion for plate failure is the attainment of yielding stress at the plate stiffener boundary. The
criterion for the stiffener-initiated failure is by yielding or by instability of the stiffener.
Little (1976) and Elsharkawi and Walker (1980) have studied the effects of continuity of
longitudinal stiffeners on the failure mode and strength of stiffened plates consisting of several
bays between cross frames. Little has indicated that there is a tendency for longitudinal
continuity to be strengthened where failure occurs in the plate, and weakened where failure
occurs in the stiffener. A simplified design method to account for the effects of continuity has
been proposed by Elsharkawi and Walker (1980).
Nonlinear finite element analysis has been compared to tests and shown to provide
reliable predictions of the ultimate strength of stiffened panels in a variety of limit states
(Grondin et al. 1998). The same researchers went on to perform a wide ranging parametric study
that resulted in design guidance for stiffened plates based on relevant plate and stiffener
dimensions. (Grondin et al. 1999). The specific role of the stiffener geometry, particularly
stiffener torsional stiffness has been the focus of tests (Ghavami 1994) as well as further
nonlinear finite element analysis (Sheikh et al. 2003).
Desmond et al. (1981a,b) have presented an experimental study of edge-stiffened
compression flanges and intermediately stiffened compression flanges. This work has been
extended for edge stiffeners (Schafer and Peköz 1999) and intermediate stiffeners (Schafer and
Peköz 1998). Effective width procedures for local and overall (distortional) buckling of the
stiffened panels were proposed and have been adopted by AISI (2007). Nguyen and Yu (1982)
have reported a study on longitudinally reinforced cold-formed steel beam webs. Based on the
experimental results, an effective-width procedure is proposed to predict the ultimate strength of
such members subjected to bending. Additional information on this subject matter is presented in
Chapters 6, 7 and 13.

4.4.3 Compression and Shear: Buckling

4.4.3.1 Initial Buckling
Analytical and experimental results on the buckling behavior of stiffened plates under combined
compression and shear are relatively scarce. Recourse is therefore usually made to data for
unstiffened plates supplemented with whatever data are available for the type of longitudinally
stiffened construction considered to be most important in practice, that is, that shown in Fig.
4.30. The case of unstiffened rectangular plates under combined compressive and shear stresses
has been presented in Section 4.2.4, which showed that a simple parabolic relationship of the
form of Eq. 4.60 was satisfactory for engineering purposes for all ranges of elastic restraint from
free rotation to complete fixity. The relationship takes the form
Rc + Rs2 = 1 (4.60)
where Rc is the ratio of compressive stress when buckling occurs in combined shear and direct
stress to compressive stress when buckling occurs in pure compression and Rs is the ratio of
shear stress when buckling occurs in combined shear and direct stress to shear stress when
buckling occurs in pure shear.
Johnson (1957) treated the problem of long plates with one and two stiffeners acted on by
axial compression and shearing stresses. The stiffeners were assumed to have bending stiffness
only and the resulting interaction curves show discontinuities which reflect the mode into which
the plate buckles. When the bending stiffness ratio of the stiffener to that of the plate (EI/bD) is
low, the buckle goes through the stiffener. However, when the stiffness ratio is increased so that
nodal lines occur along the stiffener lengths, buckling takes place without deflection along the
stiffeners. In this case another interaction relation exists between compressive and shearing
stresses. Another somewhat more limited study of the interaction relationship of infinitely wide
stiffened plates under compression and shearing stresses was reported by Harris and Pifko
(1969). In this study the finite-element method was used. The stiffeners were assumed to have
both bending and torsional stiffness, and the grid refinement used was judged to be adequate to
ensure accurate results. The results shown in Fig. 4.31 were compared to the parabolic
expression (Eq. 4.49). Except for the case of assumed large torsional stiffness ratio (GJ/bD =
106) the analytical points follow the parabolic relationship very well.

Fig. 4.30 Longitudinally stiffened plate under combined compression and shear stresses

Fig. 4.31 Analytical interaction relations for infinitely wide stiffened plate under
combined compression and shearing stresses.
4.4.3.2 Buckling in the Inelastic Range
Analytic prediction of the interaction relationship for stiffened panels which buckle in the
inelastic range under combined compression and shear are practically nonexistent in the
literature. One such computation reported by Harris and Pifko (1969) was made for an integrally
stiffened panel made of aluminum 2024-T351 and having the dimensions shown in Fig. 4.26.
The predicted interaction curves are shown in this figure for both the elastic case, which agrees
very well with the parabolic relationship (Eq. 4.60), and the inelastic-buckling case. Because of
the limited nature of the data, no general relationship for the inelastic-buckling case can be
derived. However, it should be noted from Fig. 4.32 that the circular relationship lies above the
analytical curve of the inelastic-buckling case.

Fig. 4.32 Analytically predicted elastic and inelastic interaction curves for an integrally stiffened panel.

4.4.4 Compresion and Shear: Post-buckling and Strength

Post-buckling behavior is complex, and exact treatment of the problem has not yet been
achieved. Because of the importance of predicting the strength of stiffened panels loaded beyond
initial buckling in the aircraft industry semiempirical methods have been in existence for many
years. In the case of light stiffened plates under combined compression and shear, the structure
will usually take a considerable load in excess of the initial buckling load of the plate. Chapter 6
covers the design aspects of using these post-buckling or tension-field theories; hence no further
discussion of these methods is made here.

4.4.5 Laterally Loaded plates in compression

Plates with stiffeners in the direction of the axial load that are also subjected to a distributed
lateral load are commonly encountered as the bottom plates in ships. A study based on a large-
deflection post-buckling theory (Supple, 1980) shows that lateral pressure produces effects
similar to initial geometric imperfections with a long buckling wavelength. It is shown that a
sufficiently high pressure induces stable post-buckling in the long-wave mode.
 The stiffened plating is supported on heavy transverse structural members which may be
assumed to be rigid. The panels have width-to-length ratios of about 2.5 to 4.0, and there is
essentially no interaction in the transverse direction and thus the panel undergoes cylindrical
bending. It behaves as if it were a wide beam-column under axial and lateral loads. It is
convenient to isolate one longitudinal stiffener together with a plate of the width equal to the
spacing of the stiffeners, b, and to consider that all other stiffeners behave in a similar manner.
Such a beam-column is shown in Fig. 4.33.

Fig. 4.33 Idealized beam-column.

Ultimate-strength analysis, that is, the determination of the maximum values of the lateral
and axial loads that can be sustained by the member, requires a consideration of: 1. the yield
strength of plate and stiffener materials, 2. Large-deflection theory, 3. Partial plastification of the
section, and 4. Pos-tbuckling and post-ultimate-strength behavior of the plate. The complexity of
the problem, in particular, consideration of items 2, 3, and 4 suggests that numerical methods be
used. Some solutions have been obtained for standard wide-flange shapes by Ketter (1962), and
Lu and Kamalvand (1968) in the form of interaction curves between the axial and lateral loads.
Unfortunately, these results cannot be used for plating since items 1 and 4 were not considered.
Another complication is that the wide-flange sections are doubly symmetrical, whereas the
section shown in Fig. 4.33 is singly symmetrical with a very large top flange.
This means, for example, that although it was possible to nondimensionalize the results
for wide-flange sections in terms of L/r and make them applicable with a very small error to any
wide-flange section, the unsymmetrical section of the stiffened panel must be treated as a special
case for every combination of relative proportions. Nomographs for the ultimate strengths of
stiffened plates with a yield point of 47 ksi and the ranges of geometrical and loading parameters
commonly encountered in ship structures are given by Vojta and Ostapenko (1967).
4.5 BUCKLING OF ORTHOTROPIC PLATES
Problems related to rectangular plates with stiffeners parallel to one or both pairs of sides can be
solved approximately by methods applicable to orthotropic plate theory. An orthotropic plate is
one whose material properties are orthogonally aniostropic; a uniformly stiffened plate is
reduced to this case by effectively “smearing’ the stiffness characteristics of its stiffeners over
the domain of the plate. Clearly, the theory is best applicable when the spacing of the stiffeners is
small.
The calculation of buckling strength of orthotropic plates is based on the solution of the
following differential equation governing the small deflection w(x, y) of the buckled plate:
∂4w ∂4w ∂4w ∂2w ∂2w ∂2w
D1 4 + 2 D3 2 2 + D2 4 + N x 2 + N y 2 + 2 N xy =0 (4.61)
∂x ∂x ∂y ∂y ∂x ∂y ∂x∂y
where
( EI ) x
D1 =
1 − vx v y
( EI ) y
D2 =
1 − vx v y
D3 = 12 (v y D1 + vx D2 ) + 2(GI ) xy
in which Nx, Ny, and Nxy are in-plane forces per unit width (Fig. 4.34), (EI)x and (EI)y are flexural
stiffnesses, per unit width, of beam strips in the x and y directions, respectively; vx and vy are
flexural Poisson ratios; and 2(GI)xy is a measure of torsional stiffness.

Fig. 4.34 Plate subjected to axial and shear stress.

Theoretical buckling data for several cases of rectangular plates with supported edges (w
= 0), under uniform in-plane loadings, Nx, Ny, and Nxy, applied singly or in certain combinations,
are presented. The following are some of the additional notations that will be employed: a and b
are the lengths of plate in x and y directions, respectively (see Fig. 4.34); m(integer) is the
number of buckles or half-waves in the x-direction when the buckle pattern is sinusoidal in that
direction; and n(integer) is the number of buckles or half-waves in the y-direction when the
buckle pattern is sinusoidal in the y-direction.
4.5.1 Compression
Uniaxial Compression in x-Direction, Loaded Edges Simply Supported, Unloaded Edges
Elastically Restrained Against Rotation: For these conditions, referring to Fig. 4.34, Nx is the
only loading, the edges x = 0, a are simply supported, and the edges y = 0, b are each elastically
restrained against rotation by a restraining medium whose stiffness (moment per unit length per
radian of rotation) is K. The quantity Kb/D2 to be denoted by ε, will be used as a dimensionless
measure of this stiffness. An exact solution leads to the following formula defining the value of
Nx that can sustain a buckle pattern containing m sinusoidal half-waves in the x-direction:
N x b 2 ⎛ b ⎞ D1
2
⎛ ⎛ a / m ⎞2 D2 ⎞
= + 2 + f1 ⎜ ε , ⎜
⎜ ⎝ b ⎟⎠ D3 ⎟⎟
(4.62)
π 2 D3 ⎜⎝ a / m ⎟⎠ D3 ⎝ ⎠
where f1 is the function plotted in Fig. 4.35. If (a/mb)2(D2/D3) > 0.4, Eq. 4.62 can be very closely
approximated by the formula
2 2
N x b 2 ⎛ b ⎞ D1 ⎛ a / m ⎞ D2
=⎜ ⎟ + 2 + f 2 (ε ) + ⎜ ⎟ f 3 (ε ) (4.63)
π D3 ⎝ a / m ⎠ D3
2
⎝ b ⎠ D3
where f2 and f3 are the functions plotted in Fig. 4.36. The buckling load is the smallest Nx
obtained by substituting different integer values of m (m = 1, 2, 3,...) into Eq. 4.62 or, if it
applies, Eq. 4.63. In performing this minimization, one should take into account the fact that for
most practical restraining media the stiffness K is not fixed but is a function of the half-
wavelength a/m of the edge rotation. K may also depend on the axial load in the restraining
medium, therefore on Nx, necessitating a trial-and-error calculation to determine Nx for any
selected m. Negative values of K are physically possible but are excluded from consideration in
Fig. 4.35 and Fig. 4.36. Unequal restraints along the edges y = 0 and b can be handled
approximately by first assuming the y = 0 constraint to be present at both edges, then the y = b
constraint, and averaging the two values of Nx thus obtained.

Fig. 4.35 Function f1 in Eq. 4.5

 Fig. 4.36 Functions f2 and f3 in Eq. 4.52.

Uniaxial Compression, Loaded Edges Simply Supported, Unloaded Edges Clamped: Here Nx is
the only loading, the edges x = 0 and a are simply supported, and the other two edges are
clamped. The exact solution is contained in the condition ε = ∞ of the previous example. An
approximate solution is given by Wittrick (1952) in the form
N xb2 ⎛ D3 ⎞
= k − c ⎜1 − ⎟⎟ (4.64)
π D1 D2
2 ⎜ D D
⎝ 1 2 ⎠
where c = 2.4 and k is taken from curve (c) of Fig. 4.37. Equation 4.64 is virtually equivalent to
Eq. 4.63 and must therefore be subject to the same restriction, [i.e., (a/mb)2(D2/D3) > 0.4].

Uniaxial Compression, All Edges Clamped: Wittrick (1952) gives an approximate solution for
these conditions in the form of Eq. 4.64 with c = 2.46 and k taken from curve (d) of Fig. 4.37.

4.5.2 Biaxial compression

Biaxial Compression, All Edges Simply Supported: When all four edges are simply supported,
combinations of Nx and Ny that can sustain a buckle pattern of m sinusoidal half-waves in the x-
direction and n sinusoidal half-waves in the y-direction are defined exactly by the interaction
equation
2 2 2
1 N xb2 1 N ya ⎛ b / n ⎞ D1 ⎛ a / m ⎞ D2
+ =⎜ ⎟ +2+⎜ ⎟ (4.65)
n 2 π 2 D3 m 2 π 2 D3 ⎝ a / m ⎠ D3 ⎝ b / n ⎠ D3
In using Eq. 4.53, one must substitute different combinations of m and n until that combination is
found which minimizes Nx for a given Ny, Ny for a given Nx, or Nx and Ny simultaneously for a
given ratio between them. It can be shown (Libove, 1983) that these minima will occur with at
least one of the two integers equal to unity. Therefore, only the combinations m = 1; n = 1, 2, 3,...
and n = 1; m = 2, 3,... need to be tried. The condition
N yb2
<1 (4.66)
π 2 D2
is sufficient to ensure that n = 1 will govern. In that case, with Ny regarded as given, Wittrick
(1952) has shown that the Nx required for buckling is defined by
1/ 2
N xb2 ⎡ D1 D2 ⎛ N y b 2 ⎞ ⎤
= 2 + (k − 2) ⎢ 2 ⎜ 1 − 2 ⎟ ⎥ (4.67)
π 2 D3 ⎜ ⎟
⎢⎣ D3 ⎝ π D2 ⎠ ⎥⎦
where k is the function plotted as curve (a) in Fig. 4.37. The case of the infinitely long plate (a/b
= ∞) is of interest as the limiting case of a very long plate. For that case it can be deduced from
Eq. 4.65 that if Nxb2/π2D3 ≤ 2, the longitudinal compression Nx is too small to sustain a
multilobed buckle pattern, and the plate will buckle in a cylindrical mode (i.e., as a wide plate-
column of length b) when Nyb2/π2D2 = 1. On the other hand, if Nxb2D3 > 2, the buckle pattern
will be sinusoidally lobed in the x-direction with a half-wave length-to-width ratio of
−1/ 4
a / m ⎡ D2 ⎛ N y b ⎞ ⎤
2

= ⎢ ⎜1 − 2 ⎟ ⎥ (4.68)
⎜ ⎟
b ⎢⎣ D1 ⎝ π D2 ⎠ ⎦⎥
and the buckling will occur when Nx and Ny satisfy Eq. 4.67 with k set equal to 4. The interaction
curve of Fig. 4.38 summarizes the results just given for the case a/b = ∞.

Fig. 4.37 Buckling coefficients for orthotropic plates.

 Fig. 4.38 Interaction curve for orthotropic plates.

Compression in x-Direction Applied to Clamped Edges, Limited Compression in y-Direction

Applied to Simply Supported Edges: Referring to Fig. 4.34, and considering the case in which the
edges x = 0, a are clamped, the edges y = 0, b are simply supported, Nxy is absent, and Ny satisfies
the inequality (Eq. 4.66). Then, with Ny regarded as given, the value of Nx required to cause
buckling is defined by Eq. 4.67 with k taken from curve (b) of Fig. 4.37 (Wittrick, 1952).

4.5.3 Shear
Shear, Various Boundary Conditions: Theoretical data for the shear flow Nxy required to cause
buckling of rectangular orthotropic plates have been collected by Johns (1971). Three of his
graphs are reproduced in Fig. 4.39. They apply, respectively, to the boundary conditions of (a)
all edges simply supported; (b) edges y = 0 and y = b simply supported, the other two edges
clamped; and (c) all edges clamped. In Fig. 4.39, ks stands for N xy b 2 / π 2 D11/ 4 D23/ 4 .
 Fig. 4.39 Shear buckling coefficients for orthotropic plates. [Adapted from Johns (1971), printed with the
permission of Her Majesty's Stationery Office.]

To make use of the buckling data presented above, one must of course, know the values
of the elastic constants appearing in Eq. 4.61. These constants are best determined
experimentally, by tests such as those described by Libove and Batdorf (1948) and Becker and
Gerard (1963). If the plates are of a simple enough construction, the constants can also be
evaluated theoretically. For example, for a sheet of thickness t and shear modulus G orthogonally
stiffened by x-wise stiffeners of torsional stiffness C1 spaced a distance b1 apart, and y-wise
stiffeners of torsional stiffness C2 spaced a distance a1 apart, (EI)x may be taken as the flexural
stiffness of the composite beam consisting of one x-wise stiffener and its associated width b1 of
sheet, divided by b1, while (EI)y is computed in an analogous manner using a y-wise stiffener and
its associated width a1 of sheet. For such plates, vx and vy may usually be taken as zero with little
error, and then
Gt 3 1 ⎛ C1 C2 ⎞
D3 = 2(GI ) xy = + ⎜ + ⎟ (4.69)
6 2 ⎝ b1 a1 ⎠
(The factor 12 in this formula is sometimes erroneously omitted.) Plates with integral wafflelike
stiffening may also be modified as orthotropic plates, provided that the ribbings are so oriented
as to create axes of elastic symmetry parallel to the plate edges; formulas for estimating the
elastic constants of such plates are derived by Dow et al. (1954). Corrugated-core sandwich
plates with the corrugations parallel to the x or y axis can similarly be treated as orthotropic
plates, and formulas for their elastic constants are developed by Libove and Hubka (1951);
however, for such sandwich plates, deflections due to transverse shear, neglected in the present
discussion, may sometimes be important. It is common practice to treat a corrugated plate also as
an orthotropic plate, and the appropriate elastic constants when the profile of the plate is
sinusoidal are discussed by Lau (1981). However, there are indications (Perel and Libove, 1978)
that modelling the corrugated plate as an orthotropic plate may lead to an underestimate of its
shear buckling strength.
The orthotropic plate model has an additional shortcoming when applied to stiffened
plates, namely its neglect of any coupling between in-plane forces and out-of-plane deflections.
That is, underlying Eq. 4.61 is the tacit assumption that there exists a reference plane in which
the forces Nx, Ny, and Nxy can be applied without producing any curvatures or twist. In the case of
a sheet with identical stiffening on both sides, there does of course exist such a plane—it is the
middle surface of the sheet. If the stiffening is one-sided, however, it is usually not possible to
find a reference plane that will eliminate completely the coupling between in-plane forces and
out-of-plane deflections. It has been shown (Whitney and Leissa, 1969; Jones, 1975) that such
coupling, occurring in the context of composite laminated plates, can have a marked effect on the
buckling loads. Therefore, it is very likely that in metal plates with one-sided stiffening it can
also have a marked effect on the buckling loads. A thorough investigation of this effect, using an
appropriately generalized orthotropic plate theory, would be a worthwhile subject for future
research and some work has preceded in this direction (Mikami and Niwa 1996).
Finally, orthotropic plate theory is incapable of modeling local buckling, that is, buckling
in which the buckle wavelengths are of the same order as the stiffener spacings or the widths of
the plate elements of which the stiffeners are composed. Wittrick and Horsington (1984) have
developed more refined approaches that can account for local buckling and modes of buckling in
which local and overall deformations appear in conjunction. Their methods are applicable to
plates with unidirectional stiffening possessing certain boundary conditions and subjected to
combinations of shear and biaxial compression.
4.6 INTERACTION BETWEEN PLATE ELEMENTS
In the preceding sections, attention has been confined to the behavior of a single plate element
supported along one or both of its longitudinal edges with or without stiffeners. The structural
sections employed in practice (Fig. 4.1) are composed of plate elements arranged in a variety of
configurations. It is clear that the behavior of an assembly of plates would be governed by an
interaction between the plate components. In this section the mechanics of such an interaction
and its implication in design are discussed briefly.

4.6.1 Buckling Modes of a Plate Assembly

Unlike a single plate element supported along the unloaded edges, a plate assembly can buckle in
one of several possible modes. For the case of axial compression, the buckling mode can take
one of the following forms:
● Mode I / local. This is the purely local buckling mode discussed earlier. The mode
involves out-of-plane deformation of the component plates with the junctions
remaining essentially straight, and it has a half-wavelength of the same order of
magnitude as the widths of the plate elements.
● Mode II / distortional. The buckling process may involve in-plane bending of one
or more of the constituent plates as well as out-of-plane bending of all the
elements, as in a purely local mode. Such a buckling mode is referred to as a
distortional buckling mode, stiffener buckling mode, local torsional mode, or
orthotropic mode, depending on the context. The associated wavelengths are
considerably greater than those of mode I (local buckling) but there is a half-
wavelength at which the critical stress is a minimum.
● Mode III / global. The plate structure may buckle as a column in flexural or
flexural-torsional mode with or without interaction of local buckling.
The three modes are illustrated for an I-beam in major axis bending in Fig. 4.40, based on
the finite strip analysis results of Hancock (1978). Finite strip analysis has become a popular
alternative to shell finite element analysis for this problem with several programs available for
this analysis (Papangelis and Hancock 1995; Schafer and Adany 2006). Another alternative
seeing significant recent attention is generalized beam theory (Silvestre and Camotim 2002).
Attention in this section is primarily on the mode I / local type of buckling. Global buckling
(mode III) and interaction between local and global modes of buckling, as well as mode II /
distortional type of buckling and its interaction with the other modes are treated in detail in
Chapter 13 and elsewhere in this guide.
 Fig. 4.40 Stability response of an I-beam in major axis bending using the finite strip method (a) half-
wavelength vs. buckling stress, (b) buckling mode shapes, (c) plate buckling coefficient (Hancock 1978)
4.6.2 Local Buckling of a Plate Assembly
A prismatic plate structure is often viewed simply as consisting of stiffened and unstiffened plate
elements. The former are plate elements supported on both of their longitudinal edges by virtue
of their connection to adjacent elements, while the latter are those supported only along one of
their longitudinal edges. Thus, the critical local buckling stress of a plate assembly may be taken
as the smallest of the critical stresses of the plate elements, each treated as simply supported
along its junctions with other plates. This stress will be conservative because the element
providing the lower bound will be restrained by the more stable adjoining plate elements if
longitudinal edge joints provide effective continuity. More complete information on k-factors as
influenced by the interaction between plate components can be found in a number of classic
references (Bleich, 1952); (Stowell et al., 1952; Gerard and Becker, 1957/1958; Timoshenko and
Gere, 1961). However, such a calculation must be used with caution for the following reasons:
1. The results can be unduly conservative when the plate structure consists of
elements with widely varying slenderness. This is the result of neglecting the
rotational restraints at the functions.
2. The results are inapplicable unless it is ensured that all the plate elements buckle
locally (i.e., the junctions remain essentially straight). If, on the other hand, mode
II or III type of buckling is critical, the result of such a simplified calculation
would be on the unsafe side.
The intervention of distortional, or stiffener buckling (mode II) may be averted in
practice by designing “out” the stiffener buckling mode by the provision of stiffeners (edge or
intermediate) of adequate rigidity. This may be advantageous because of the limited post-
buckling resistance associated with the mode II type of buckling. However, it is not always
practical or economical, designs where distortional buckling must be considered are discussed
further in Chapter 13.
Figure 4.18 gives the variations of the local buckling coefficients kw for a wide-flange I-
section, a box section and a Z- or channel section, respectively, with respect to the geometrical
properties of the member. Each of these charts is divided into two portions by a dashed line
running across it (Kroll et al., 1943). In each portion, buckling of one of the elements dominates
over the other, but the proportions exactly on the dashed line represent the case where the
buckling stress of the elements are equal. Additional charts and related information may be
found in the references such as the Japanese Handbook of Structural Stability (CRCJ, 1971). In
addition, free or low-cost computational tools (Papangelis and Hancock 1995; Schafer and
Adany 2006; Silvestre and Camotim 2002); are now readily available and able to provide
similar results.
Fig. 4.41 Plate buckling coefficients for plate assemblies (Kroll et al., 1943)
4.6.3 Post-buckling of a Plate Assembly
Interaction between the elements of a plate assembly is inescapable because of the equilibrium
and compatibility conditions that must be satisfied at the junction. In the case of local buckling it
is possible to simplify these conditions considerably, as has been shown by Benthem (1959). The
smallness of the in-plane displacements in comparison to the out-of-plane displacements makes
it possible to assume that normal displacements are zero for each plate element meeting at a
corner. Also, because of the smallness of the bending rigidity in comparison to the extensional
rigidity, it is possible to assume that the in-plane membrane meets another plate element at an
angle (Walker, 1964; Graves-Smith, 1968; Rhodes and Harvey, 1971; Tien and Wang, 1979).
In the post-buckling range in-plane displacements and membrane stresses dominate the
behavior of the buckled plates. The interactions between plate elements along the junctions
become very complex. The problem is compounded when interaction between global and local
buckling is considered, or when interaction between distortional and the other modes is
considered. Suffice to say post-buckling and strength of plate assemblies considering all possible
instabilities (modes I – III) remains a topic of considerable interest in research.
The earlier references on interaction of local and overall buckling in the post-buckling
range include those by Bijlaard and Fisher (1952), Cherry (1960), Graves-Smith (1969), Klöppel
et al. (1969), Sharp (1970), and Škaloud and Zornerova (1970). Research in this field has also
been carried out by Rhodes and Harvey (1976), Graves-Smith and Sridharan (1978, 1980),
Kalyanaraman (1978), Thomasson (1978), Little (1979), Rhodes and Loughlan (1980), Hancock
(1980, 1981), Sridharan and Graves-Smith (1981), Fukumoto and Kubo (1982), Sridharan (1982)
and Bradford and Hancock (1984). Using the concept of effective width, the problem has been
investigated by DeWolf et al. (1974), Wang and Pao, (1981), and Desmond et al. (1981a and b).
As design methods for local-overall buckling (modes I and III) mature, recent research
attentions has been focused on mode II: distortional buckling. Lau and Hancock (1987) provided
analytical formula for predicting elastic distortional buckling and later Hancock, Kwon et al.
(1994) provided empirical formulas for predicting the ultimate strength in distortional buckling
for a variety of different cold-formed steel cross-sections. Bradford (1992) has examined the
distortional mode in the context of hot-rolled steel I-beams, while Schafer and Peköz (2002;
1999) examined local, distortional, and global buckling of cold-formed steel beams and columns
in developing the methods currently used in the AISI (2007) Specification. The application
aspect of this subject is discussed further for thin-walled members in Chapter 13.
4.7 REFERENCES
AA (1994), Aluminum Design Manual, Aluminum Association, Washington, D.C.
Abdel-Sayed, G. (1969), “Effective Width of Thin Plates in Compression,” ASCE J. Struct. Div., Vol. 95,
No. ST10, pp. 2183–2204.
AISC (1993), Load and Resistance Factor Design for Structural Steel Buildings, American Institute of
Steel Construction, Chicago.
AISI (1996), Specification for the Design of Cold-Formed Steel Structural Members, American Iron and
Steel Institute, Washington, D.C.
Allen, H. G., and Bulson, P. S. (1980), Background to Buckling, McGraw-Hill, Maidenhead, Berkshire,
England.
Aluminum Company of America (1960), Alcoa Structural Handbook (also Handbook of Design Stresses
for Aluminum), Alcoa, Pittsburgh, Pa.
Azhari, M., and Bradford, M. A. (1993), “Inelastic Initial Load Buckling of Plates with and Without
Residual Stresses,” J. Eng. Struct., Vol. 15, No. 1, 1993, pp. 31–39.
Batdorf, S. B., and Houbolt, J. C. (1946), “Critical Combinations of Shear and Transverse Direct Stress
for an Infinitely Long Flat Plate with Edges Elastically Restrained Against Rotation,” NACA Rep.
No. 847.
Batforf, S. B., and Stein, M. (1947), “Critical Combinations of Shear and Direct Stress for Simply
Supported Rectangular Flat Plates,” NACA Tech. Note No. 1223.
Becker, H., and Gerard, G. (1963), “Measurement of Torsional Rigidity of Stiffened Plates,” NASA Tech.
Note No. D-2007.
Benthem, J. P. (1959), “The Reduction in Stiffness of Combinations of Rectangular Plates in
Compression After Exceeding the Buckling Load,” NLL-TR S-539.
Bergman, S., and Reissner, H. (1932), “Über die Knickung von rechteckigen Platten bei
Schubbeanspruchung,” Z. Flugtech Motorluftschiffahrt, Vol. 23, p. 6.
Bijlaard, P. P. (1949), “Theory and Tests on Plastic Stability of Plates and Shells,” J. Aeronaut. Sci., Vol.
16, pp. 529–541.
Bijlaard, P. P. (1950), ‘“On the Plastic Buckling of Plates,” J. Aeronaut. Sci., Vol. 17, p. 493.
Bijlaard, P. P. (1957), “Buckling of Plates Under Nonhomogeneous Stress,” ASCE J. Eng. Mech. Div.,
Vol. 83, No. EM3, Proc. Paper 1293.
Bijlaard, P. P., and Fisher, G. P. (1952), “Column Strength of H Sections and Square Tubes, in Post
Buckling Range of Component Plates,” NACA Tech. Note No. 2994.
Bleich, F. (1952), Buckling Strength of Metal Structures, McGraw-Hill, New York.
Bleich, F., and Ramsey, L. B. (1951), A Design Manual on the Buckling Strength of Metal Structures,
Society of Naval Architects, Washington, D.C.
Brockenbrough, R. L., and Johnston, B. G. (1974), U.S. Steel Design Manual, 2nd ed., U.S. Steel
Corporation, Pittsburgh, Pa.
Brown, C. J. (1990), “Elastic Buckling of Perforated Plates Subjected to Concentrated Loads,” J. Comput.
Struct., Vol. 36, No. 6, pp. 1103–1109.
Bryan, G. H. (1891), “On the Stability of a Plane Plate Under Thrusts in Its Own Plane, with Applications
to the ‘Buckling’ of the Sides of a Ship,” Proc. London Math. Soc., Vol. 22.
Bulson, P. S. (1970), Stability of Flat Plates, American Elsevier, New York.
Cherry, S. (1960), “The Stability of Beams with Buckled Compression Flanges,” Struct. Eng., Vol. 38,
No. 9.
Chwalla, E. (1936a), “Beitrag zur Stabilitãtstheorie des Stegbleches vollwandiger Träger,” Stahlbau, Vol.
9, p. 81.
Chwalla, E. (1936b), “Die Bemessung der waagerecht ausgesteiften Stegbleche vollwandiger Träger,”
Prelim Publ. IABSE, 2nd Cong., Berlin-Munich, p. 957.
Clark, J. W., and Rolf, R. L. (1966), “Buckling of Aluminum Columns, Plates and Beams,” ASCE J.
Struct. Div., Vol. 92, No. ST3, pp. 17–38.
Conley, W. F., Becker, L. A., and Allnutt, R. B. (1963), “Buckling and Ultimate Strength of Plating
Loaded in Edge Compression: Progress Report 2. Unstiffened Panels,” David Taylor Model Basin
Rep. No. 1682.
Cook, I. T., and Rockey, K. C. (1963), “Shear Buckling of Rectangular Plates with Mixed Boundary
Conditions,” Aeronaut. Q., Vol. 14.
CRCJ (Column Research Committee of Japan) (1971), Handbook of Structural Stability, Corona
Publishing Co., Tokyo.
Crisfield, M. A. (1975), “Full Range Analysis of Steel Plates and Stiffened Plating Under Uniaxial
Compression,” Proc. Inst. Civ. Eng., Part 2, pp. 1249–1255.
CSA (1974, 1984), Cold Formed Steel Structural Members, CSA Standard S136, Canadian Standards
Association, Rexdale, Ontario, Canada.
CSA (1983), Strength Design in Aluminum, CAN3-S157-M83, Canadian Standards Association, Rexdale,
Ontario, Canada.
Dawe, J. L., and Kulak, G. L. (1984), “Local Buckling of W-Shaped Columns and Beams,” ASCE J.
Struct. Eng., Vol. 110, No. 6, pp. 1292–1304.
Dawe, J. L., and Kulak, G. L. (1986), “Local Buckling of Beam-Columns,” ASCE J. Struct. Eng., Vol.
112, No. 11, pp. 2447–2461.
Dawson, R. G., and Walker, A. C. (1972), “Post-buckling of Geometrically Imperfect Plates,” ASCE J.
Struct. Div., Vol. 98, No. ST1, pp. 75–94.
Desmond, T. P., Peköz, T., and Winter, G. (1981a), “Edge Stiffeners for Thin-Walled Members,” ASCE
J. Struct. Div., Vol. 107, No. ST2, Proc. Pap. 16056, pp. 329–354.
Desmond, T. P., Peköz, T., and Winter, G. (1981b), “Intermediate Stiffeners for Thin Walled Members,”
ASCE J. Struct. Div., Vol. 107, No. ST4, Proc. Pap. 16194, pp. 627–648.
DeWolfe, J. T., Peköz, T., and Winter, G. (1974), “Local and Overall Buckling of Cold Formed
Members,” ASCE J. Struct. Div., Vol. 100, No. ST10, pp. 2017–2306.
Dow, N. F., Libove, C., and Hubka, R. E. (1954), “Formulas for the Elastic Constants of Plates with
Integral Waffle-like Stiffening,” NACA Rep. No. 1195.
Dwight, J. B. (1971), “Collapse of Steel Compression Panels,” Proc. Conf. Dev. Bridge Des. Constr.,
Crosby Lockwood, London.
Dwight, J. B., and Little, G. H. (1976), “Stiffened Steel Compression Flanges, a Simpler Approach,”
Struct. Eng., Vol. 54, No. 12, pp. 501–509.
Dwight, J. B., and Moxham, K. E. (1969), “Welded Steel Plates in Compression,” Struct. Eng., Vol. 47,
No. 2.
Dwight, J. B., and Ratcliffe, A. T. (1968), “The Strength of Thin Plates in Compression,” in Thin-Walled
Steel Structures (ed. K. C. Rockey and H. V. Hill), Crosby Lockwood, London.
Elangovan, P. T., and Prinsze, M. (1992), “Critical Shear Stress of Flat Rectangular Plates with Free
Edges,” J. Thin-Walled Struct., Vol. 13, No. 5, pp. 409–425.
Elsharkawi, K., and Walker, A. C. (1980), “Buckling of Multibay Stiffened Plate Panels,” ASCE J. Struct.
Div., Vol. 106, No. ST8, pp. 1695–1716.
Fok, W. C., Walker, A. C., and Rhodes, J. (1977), “Buckling of Locally Imperfect Stiffeners in Plates,”
ASCE J. Eng. Mech. Div., Vol. 103, No. EM5, pp. 895–911.
Fujita, Y., Yoshida, K., and Arai, H. (1970), “Instability of Plates with Holes,” 3rd Rep., J. Soc. Nav.
Archit. Jpn., Vol. 127, pp. 161–169.
Fukomoto, Y., and Kubo, M. (1982), “Buckling in Steel U Shaped Beams,” ASCE J. Struc. Div., Vol.
108, No. ST5, pp. 1174–1190.
Fukomoto, Y., Usami, T., and Yamaguchi, K. (1977), “Inelastic Buckling Strength of Stiffened Plates in
Compression,” IABSE Period. 3/1977, IABSE Proc. p-8/77, pp. 1–15.
Gerard, G., and Becker, H. (1957/1958), “Handbook of Structural Stability,” six parts, NACA Tech. Notes
Nos. 3781–3786.
Graves-Smith, T. R. (1968), “The Postbuckled Behavior of Thin Walled Columns,” 8th Cong. IABSE,
New York, pp. 311–320.
Graves-Smith, T. R. (1969), “The Ultimate Strength of Locally Buckled Columns of Arbitrary Length,”
in Thin-Walled Steel Structures (ed. K. C. Rockey and H. V. Hill), Crosby Lockwood, London.
Graves-Smith, T. R., and Sridharan, S. (1978), “A Finite Strip Method for the Buckling of Plate
Structures Under Arbitrary Loading,” Int. J. Mech. Sci., Vol. 20, pp. 685–693.
Graves-Smith, T. R., and Sridharan, S. (1980), “The Local Collapse of Elastic Thin-Walled Columns,” J.
Struct. Mech., Vol. 13, No. 4, pp. 471–489.
Grosskurth, J. F., White, R. N., and Gallagher, R. H. (1976), “Shear Buckling of Square Perforated
Plates,” ASCE J. Eng. Mech. Div., Vol. 102, No. EM6, Proc. Pap. 12641, pp. 1025–1040.
Hancock, G. J. (1980), “Design Methods for Thin-Walled Box Columns,” Rep. No. 359, University of
Sydney, Sydney, Australia.
Hancock, G. J. (1981), “Interaction Buckling of I-Section Columns,” ASCE J. Struct. Div., Vol. 107, No.
ST1, pp. 165–181.
Harris, H. G., and Pifko, A. B. (1969), “Elastic-Plastic Buckling of Stiffened Rectangular Plates,” Proc.
Symp. Appl. Finite Elem. Methods Div. Eng., Vanderbilt University, Nashville, Tenn., Nov.
Horne, M. R., and Narayanan, R. (1975), “An Approximate Method for the Design of Stiffened Steel
Compression Panels,” Proc. Inst. Civ. Eng., Part 2, Vol. 59, pp. 501–514.
Horne, M. R., and Narayanan, R. (1977), “Design of Axially Loaded Stiffened Plates,” ASCE J. Struct.
Div., Vol. 103, No. ST11, pp. 2243–2257.
Iguchi, S. (1938), “Die Kninckung der rechteckigen Platte durch Schubkräfte,” Ing. Arch., Vol. 9, p. 1.
Inoue, T., and Kato, B. (1993), “Analysis of Plastic Buckling of Steel Plates,” Int. J. Solids Struct., Vol.
30, No. 6, pp. 835–856.
Isami, T., and Hidenori, K. (1990), “New Strength Prediction Method for Biaxially Compressed Plates,”
Proc. 39th Jpn. Nat. Congr. Appl. Mech. Theor. Appl. Mech., Vol. 39, University of Tokyo Press,
Tokyo, pp. 109–118.
Johns, D. J. (1971), “Shear Buckling of Isotropic and Orthotropic Plates: A Review,” Tech. Rep. R&M
No. 3677, British Aeronautical Research Council, London.
Johnson, A. E., Jr. (1957), “Charts Relating the Compressive and Shear Buckling Stress of Longitudinally
Supported Plates to the Effective Deflectional Stiffness,” NACA Tech. Note No. 4188.
Johnson, A. L., and Winter, G. (1966), “Behavior of Stainless Steel Columns and Beams,” ASCE J.
Struct. Div., Vol. 92, No. ST5, p. 97.
Jombock, J. R., and Clark, J. W. (1962), “Post-buckling Behavior of Flat Plates,” Trans. Am. Soc. Civ.
Eng., Vol. 127, Part I, pp. 227–240.
Jombock, J. R., and Clark, J. W. (1968), “Bending Strength of Aluminum Formed Sheet Members,”
ASCE J. Struct. Div., Vol. 94, No. ST2, pp. 511–528.
Jones, R. M. (1975), Mechanics of Composite Materials, Scripta, Silver Spring, Md., pp. 264–267.
Kalyanaraman, V., and Peköz, T. (1978), “Analytical Study of Unstiffened Elements,” ASCE J. Struct.
Div., Vol. 104, No. ST9, pp. 1507–1524.
Kalyanaraman, V., Peköz, T., and Winter, G. (1977), “Unstiffened Compression Elements,” ASCE J.
Struct. Div., Vol. 103, No. ST9, pp. 1833–1848.
Kawai, T., and Ohtsubo, H. (1968), “A Method of Solution for the Complicated Buckling Problems of
Elastic Plates with Combined Use of Raleigh–Ritz Procedure in the Finite Element Method,”
Proc. 2nd Air Force Conf. Matrix Methods Struct. Mech., Oct.
Kemp, A. R. (1996), “Inelastic Local and Lateral Buckling in Design Codes,” ASCE J. Struct. Eng., Vol.
122, No. 4, pp. 374–382.
Ketter, R. L. (1962), “Further Studies on the Strength of Beam-Columns,” Trans. ASCE, Vol. 127, No. 2,
p. 224.
Klitchieff, J. M. (1949), “On the Stability of Plates Reinforced by Ribs,” J. Appl. Mech., Vol. 16, No. 1.
Klöppel, K., Schmied, R., and Schubert, J. (1969), “Ultimate Strength of Thin Walled Box Shaped
Columns Under Concentric and Eccentric Loads: Post Buckling
Behavior of Plate Elements and Large Deformations Are Considered,” Stahlbau, Vol. 38, pp. 9–19, 73–
83.
Koiter, W. T. (1963), “Introduction to the Post-buckling Behavior of Flat Plates,” Int. Colloq.
Comportement Postcritiques Plaques Utilisée Constr. Metall., Institute of Civil Engineering,
Liege, Belgium.
Kroll, W. D., Fisher, G. P., and Heimerl, G. J. (1943), “Charts for the Calculation of the Critical Stress for
Local Instability of Columns with I, Z, Channel and Rectangular Tube Sections,” NACA Wartime
Rep. No. L-429.
Kuhlmann, U. (1989), “Definition of Flange Slenderness Limits on the Basis of Rotation Capacity
Values,” J. Constr. Steel Res., Vol. 14, No. 1, pp. 21–40.
Kumai, T. (1952), “Elastic Stability of a Square Plate with a Central Circular Hole Under Edge Thrust,”
Rep. Inst. Appl. Mech. Kyusyn Univ., Vol. 1, No. 2, p. 1.
Kusuda, T. (1959), “Buckling of Stiffened Panels in Elastic and Strain-Hardening Range,” Rep. No. 39,
Transportation Technical Research Institute (Unyn-Gijutsu Kenkyrijo Mejiro, Toshima-Ku),
Tokyo.
LaBoube, R. A., and Yu, W. W. (1982), “Bending Strength of Webs of Cold-Formed Steel Beams,”
ASCE J. Struct. Div., Vol. 108, No. ST7, pp. 1589–1604.
Lau, J. H. (1981), “Stiffness of Corrugated Plate,” ASCE J. Eng. Mech. Div., Vol. 17, No. EM1, pp. 271–
275.
Lau, S. C. W., and Hancock, G. J. (1989), “Inelastic Buckling Analysis of Beams, Columns and Plates
Using the Spline Finite Strip Method,” J. Thin-Walled Struct., Vol. 7, No. 3–4, pp. 213–238.
Lay, M. G. (1965), “Flange Local Buckling in Wide-Flange Shapes,” ASCE J. Struct. Div., Vol. 91, No.
ST6, pp. 49–66.
Legget, D. M. A. (1941), “The Buckling of a Square Panel in Shear When One Pair of Opposite Edges Is
Clamped and the Other Pair Simply Supported,” Tech. Rep. R&M No. 1991, British Aeronautical
Research Council, London.
Levy, S., Wolley, R. M., and Knoll, W. D. (1947), “Instability of Simply Supported Square Plate with
Reinforced Circular Hole in Edge Compression,’ J. Res. Natl. Bur. Stand., Vol. 39, pp. 571–577.
Libove, C. (1983), “Buckle Pattern of Biaxially Compressed Simply Supported Orthotropic Rectangular
Plates,” J. Compos. Mater., Vol. 17, No. 1, pp. 45–48.
Libove, C., and Batdorf, S. B. (1948), “A General Small-Deflection Theory for Flat Sandwich Plates,”
NACA Rep. No. 899.
Libove, C., and Hubka, R. E. (1951), “Elastic Constants for Corrugated-Core Sandwich Plates,” NACA
Tech. Note No. 2289.
Lind, N. C. (1973), “Buckling of Longitudinally Stiffened Sheets,” ASCE J. Struct. Div., Vol. 99, No.
ST7, pp. 1686–1691.
Lind, N. C., Ravindra, M. K., and Schorn, G. (1976), “Empirical Effective Width Formula,” ASCE J.
Struct. Div., Vol. 102, No. ST9, pp. 1741–1957.
Little, G. H. (1976), “Stiffened Steel Compression Panels: Theoretical Failure Analysis,” Struct. Eng.,
Vol. 54, No. 12, pp. 489–500.
Little, G. H. (1979), “The Strength of Square Steel Box-Columns: Design Curves and Their Theoretical
Basis,” Struct. Eng., Vol. 57.
Lu, L. W., and Kamalvand, H. (1968), “Ultimate Strength of Laterally Loaded Columns,” ASCE J. Struct.
Div., Vol. 94, No. ST6, pp. 1505–1524.
May, I. M., and Ganaba, T. H. (1988), “Elastic Stability of Plates with and Without Openings,” J. Eng.
Comput., Vol. 5, No. 1, pp. 50–52.
McKenzie, K. I. (1964), “The Buckling of Rectangular Plate Under Combined Biaxial Compression,
Bending and Shear,” Aeronaut. Q., Vol. 15, No. 3, pp. 239–246.
Merrison Committee (1973), “Inquiry into the Basis of Design and Method of Erection of Steel Box
Girder Bridges (Interior Design and Workmanship Rules),” Merrison Committee, London.
Michael, M. E. (1960), “Critical Shear Stress for Webs with Holes,” J. R. Aeronaut. Soc., Vol. 64, p. 268.
Moheit, W. (1939), “Schubbeulung rechteckiger Platten mit eingespannten Rändern,” Thesis, Technische
Hochschule Darmstadt, Leipzig, Germany.
Murray, J. M. (1946), “Pietzker’s Effective Breadth of Flange Reexamined,” Engineering, Vol. 161, p.
364.
Murray, N. W. (1973), “Buckling of Stiffened Panel Loaded Axially and in Bending,” Struct. Eng., Vol.
51, No. 8, pp. 285–301.
Murray, N. W. (1975), “Analysis and Design of Stiffened Plates for Collapse Load,” Struct. Eng., Vol.
53, No. 3, pp. 153–158.
Nemeth, J. and Michael, P. (1990), “Buckling and Post-buckling Behavior of Compression-Loaded
Isotropic Plates with Cutouts,” Proc. 31st Struct. Dyn. Mater. Conf., April 2–4 Part 2, American
Institute of Aeronautics and Astronautics, New York, pp. 862–876.
Newell, J. S. (1930), “The Strength of Aluminum Alloy Sheets,” Airway Age, Vol. 11, p. 1420; Vol. 12,
p. 1548.
Nguyen, R. P., and Yu, W. W. (1982), “Longitudinally Reinforced Cold-Formed Steel Beam Webs,”
ASCE J. Struct. Div., Vol. 108, No. ST11, pp. 2423–2442.
Pavlovic, M. N., and Baker, G. (1989), “Effect of Non-uniform Stress Distribution on the Stability of
Biaxially Compressed Plates,” Int. J. Struct. Mech. Mater. Sci., Vol. 28, No. 1–4, pp. 69–89.
Perel, D., and Libove, C. (1978), “Elastic Buckling of Infinitely Long Trapezoidally Corrugated Plates in
Shear,” Trans. ASME J. Appl. Mech., Vol. 45, pp. 579–582.
Peters, R. W. (1954), “Buckling of Long Square Tubes in Combined Compression and Torsion and
Comparison with Flat-Plate Buckling Theories,” NACA Tech. Note No. 3184.
Ramberg, W., McPherson, A. E., and Levy, S. (1939), “Experiments on Study of Deformation and of
Effective Width in Axially Loaded Sheet-Stringer Panels,” NACA Tech. Note No. 684.
Redwood, R. G., and Uenoya, M. (1979), “Critical Loads for Webs with Holes,” ASCE J. Struct. Div.,
Vol. 105, No. ST10, pp. 2053–2068.
Rhodes, J., and Harvey, J. M. (1971), “Effects of Eccentricity of Load or Compression on the Buckling
and Post-buckling Behavior of Flat Plates,” Int. J. Mech. Sci., Vol. 13, pp. 867–879.
Rhodes, J., and Harvey, J. M. (1976), “Plain Channel Section Struts in Compression and Bending Beyond
the Local Buckling Load,” Int. J. Mech. Sci., Vol. 8, pp. 511–519.
Rhodes, J., and Harvey, J. M. (1977), “Examination of Plate Post Buckling Behavior,” ASCE J. Eng.
Mech. Div., Vol. 103, No. EM3, pp. 461–480.
Rhodes, J., and Loughlan, J. (1980), “Simple Design Analysis of Lipped Simple Columns,” Proc. 5th Int.
Spec. Conf. Cold-Formed Steel Struct., St. Louis, Mo.
Rockey, K. C. (1980), “The Buckling and Post-buckling Behavior of Shear Panels Which Have a Central
Circular Cut Out,” in Thin Walled Structures (ed. J. Rhodes and A. C. Walker), Granada, London.
Rockey, K. C., Anderson, R. G., and Cheung, Y. K. (1969), “The Behavior of Square Shear Webs Having
a Circular Hole,” in Thin-Walled Steel Structures (ed. K. C. Rockey and H. V. Hill), Crosby
Lockwood, London.
Roorda, J., and Venkataramaiah, K. R. (1979), “Effective Width of Stiffened Cold-Formed Steel Plates,”
Can J. Civ. Eng., Vol. 6, No. 3.
Schlack, A. L. (1964), “Elastic Stability of Pierced Square Plates,” Exp. Mech., p. 167.
Schuman, L., and Back, G. (1930), “Strength of Rectangular Flat Plates Under Edge Compression,”
NACA Tech. Rep. No. 356.
Seide, P., and Stein, M. (1949), “Compressive Buckling of Simply Supported Plates with Longitudinal
Stiffeners,” NACA Tech. Note No. 1825.
Seydel, E. (1933), Über das Ausbeulen von rechteckigen isotropen oder orthogonalanisotropen Platten bei
Schubbeanspruchung,” Ing. Arch., Vol. 4, p. 169.
Sharp, M. L. (1966), “Longitudinal Stiffeners for Compression Members,” ASCE J. Struct. Div., Vol. 92,
No. ST5, pp. 187–212.
Sharp, M. L. (1970), “Strength of Beams or Columns with Buckled Elements,” ASCE J. Struct. Div., Vol.
96, No. ST5, pp. 1011–1015.
Sherbourne, A. N., Marsh, C., and Lian, C. Y. (1971), “Stiffened Plates in Uniaxial Compression,” IABSE
Publ., Vol. 31-I, Zurich, pp. 145–178.
Škaloud, M., and Zornerova, M. (1970), “Experimental Investigation into the Interaction of the Buckling
of Compressed Thin-Walled Columns with the Buckling of Their Plate Elements,” Acta Tecnica
CSAV No. 4.
Southwell, R. V., and Skan, S. (1924), “On the Stability Under Shearing Force of a Flat Elastic Strip,”
Proc. R. Soc., Vol. A105.
Sridharan, S. (1982), “A Semi-analytical Method for the Post-Local-Torsional Buckling Analysis of
Prismatic Plate Structures,” Int. J. Numer. Methods Eng., Vol. 18, pp. 1685–1607.
Sridharan, S., and Graves-Smith, T. R. (1981), “Post-buckling Analysis with Finite Strips,” ASCE J. Eng.
Mech. Div., Vol. 107, No, EM5, pp. 869–888.
Stein, O. (1936), “Stabilität ebener Rechteckbleche unter Biegung and Schub,” Bauingenieur, Vol. 17, p.
308.
Stein, M. (1959), “The Phenomenon of Change in Buckle Pattern in Elastic Structures,” NACA Tech.
Note No. R-39.
Stein, M. (1989), “Effect of Transverse Shearing Flexibility on Post-buckling on Plates in Shear,” AIAA
J., Vol. 27, No. 5, pp. 652–655.
Stowell, E. Z. (1948), “A Unified Theory of Plastic Buckling of Columns and Plates,” NACA Tech. Note
No. 1556.
Stowell, E. Z. (1949), “Plastic Buckling of a Long Flat Plate Under Combined Shear and Longitudinal
Compression,” NACA Tech. Note No. 1990.
Stowell, E. Z., Heimerl, G. J., Libove, C., and Lundquist, E. E. (1952), “Buckling Stresses for Flat Plates
and Sections,” Trans. Am. Soc. Civ. Eng., Vol. 117, pp. 545–578.
Supple, W. J. (1980), “Buckling of Plates Under Axial Load and Lateral Pressures,” in Thin Walled
Structures (ed. J. Rhodes and A. C. Walker), Granada, London.
Thomasson, P.-O. (1978), “Thin-Walled C-Shaped Panels in Axial Compression,” Document D.I. 1978,
Swedish Council for Building Research, Stockholm.
Tien, Y. L., and Wang, S. T. (1979), “Local Buckling of Beams Under Stress Gradient,” ASCE J. Struct.
Div., Vol. 105, No. ST8, pp. 1571–1588.
Timoshenko, S. (1910), “Einge Stabilitätsprobleme der Elastizitätstheorie,” Z. Math. Phys., Vol. 58, pp.
337.
Timoshenko, S. (1934), “Stability of the Webs of Plate Girders,” Engineering, Vol. 138, p. 207.
Timoshenko, S. P., and Gere, J. M. (1961), Theory of Elastic Stability, 2nd ed., McGraw-Hill, New York.
Troitsky, D. S. C. (1976), Stiffened Plates, Bending, Stability and Vibrations, Elsevier, New York.
Tsien, H. S. (1942), “Buckling of a Column with Nonlinear Lateral Supports,” J. Aeronaut. Sci., Vol. 9,
No. 4.
Tvergaard, V. (1973), “Imperfection-Sensitivity of a Wide Integrally Stiffened Panel Under
Compression,” Int. J. Solids Struct., Vol. 9, pp. 177–192.
Usami, T. (1982), “Post Buckling of Plates in Compression and Bending,” ASCE J. Struct. Div., Vol. 108,
No. ST3, pp. 591–609.
Usami, T., and Fukumoto, Y. (1982), “Local and Overall Buckling of Welded Box Columns,” ASCE J.
Struct. Div., Vol. 108, No. ST3, pp. 525–542.
Vojta, J. F., and Ostapenko, A. (1967), “Ultimate Strength Design Curves for Longitudinally Stiffened
Plate Panels with Large b/t,” Fritz Eng. Lab. Rep. No. 248.19, Lehigh University, Bethlehem, Pa.
von Kármán, T., Sechler, E. E., and Donnell, L. H. (1932), “Strength of Thin Plates in Compression,”
Trans. A.S.M.E., Vol. 54, No. APM-54-5, p. 53.
Walker, A. C. (1964), “Thin Walled Structural Forms Under Eccentric Compressive Load Actions,”
Doctoral thesis, University of Glasgow, Glasgow, Scotland.
Wang, S. T. (1969), “Cold-Rolled Austenitic Stainless Steel: Material Properties and Structural
Performance,” Cornell Univ. Dept. Struct. Eng. Rep. No. 334.
Wang, S. T., and Pao, H. Y. (1981), “Torsional-Flexural Buckling of Locally Buckled Columns,” Int. J.
Comput. Struct., Vol. 11.
Wang, S. T., Errera, S. J., and Winter, G. (1975), “Behavior of Cold Rolled Stainless Steel Members,”
ASCE J. Struct. Div., Vol. 101, No. ST11, pp. 2337–2357.
Wang, S. T., Yost, M. I., and Tien, Y. L. (1977), “Lateral Buckling of Locally Buckled Beams Using
Finite Element Techniques,” Int. J. Comput. Struct., Vol. 7.
Way, S. (1936), “Stability of Rectangular Plates Under Shear and Bending Forces,” J. Appl. Mech.
ASME. Dec. Vol. 3, No. 4, pp. A131–A135.
Webb, S. E., and Dowling, P. J. (1980), “Large Deflection Elasto-plastic Behavior of Discretely Stiffened
Plates,” Proc. Inst. Civ. Eng., Part 2, Vol. 69, pp. 375–401.
Whitney, J. M., and Leissa, A. W. (1969), “Analysis of Heterogeneous Anisotropik Plates,” Trans. ASME
J. Appl. Mech., Vol., 36, pp. 261–266.
Winter, G. (1947), “Strength of Thin Steel Compression Flanges,” Trans. ASCE, Vol. 112, p. 527.
Winter, G. (1983), “Commentary on the 1968 Edition of the Specification for the Design of Steel
Structural Members,” in Cold-Formed Steel Design Manual, American Iron and Steel Institute,
Washington, D.C.
Winter, G., Lansing, W., and McCalley, R. B. (1950), “‘Four Papers on the Performance of Thin Walled
Steel Structures,” Eng. Exp. Strn. Rep. No. 33, Cornell University, Ithaca, N.Y., pp. 27–32, 51–
57.
Wittrick, W. H. (1952), “‘Correlation Between Some Stability Problems for Orthotropic and Isotropic
Plates Under Bi-axial and Uniaxial Direct Stress,” Aeronaut. Q., Vol. 4, pp. 83–99.
Wittrick, W. H. (1968), “A Unified Approach to the Initial Buckling of Stiffened Panels in Compression,”
Aeronaut. Q., Vol. 19, p. 265.
Wittrick, W. H., and Horsington, R. W. (1984), “Buckling and Vibration of Composite Folded Plate
Structures of Finite Length in Combined Shear and Compression,” Proc. R. Soc. London, Vol.
A392, pp. 107–144.
Yamaki, N. (1959/1960), “Post-buckling Behavior of Rectangular Plates with Small Initial Curvature
Loaded in Edge Compression,” J. Appl. Mech. ASME, Vol. 26, pp. 407–414; Vol. 27, pp. 335–
342.
Yang, H. T. Y. (1969), “A Finite Element Formulation for Stability Analysis of Doubly Curved Thin
Shell Structures,” Doctoral thesis, Cornell University, Ithaca, N.Y.
Yoshizuka, J., and Naruoka, M. (1971), “Buckling Coefficient of Simply Supported Rectangular Plates
Under Combined Bending and Compressive Stresses in Two Perpendicular Directions,” Stahlbau,
Vol. 40, p. 217.
Yu, W. W., and Davis, C. S. (1973), “Cold Formed Steel Members with Perforated Elements,” ASCE J.
Struct. Div., Vol. 99, No. ST10, p. 2061.
Zender, W., and Hall, J. B., Jr. (1960), “Combinations of Shear Compressive Thermal and Compressive
Load Stresses for the Onset of Permanent Buckles in Plates,” NASA Tech. Note No. D-384.

New references
AA (2005), Aluminum Design Manual, Aluminum Association, Washington, D.C.
AISC (2005) Specification for Structural Steel Buildings, American Institute of Steel Construction,
Chicago, IL, AISC 360-05.
AISC (2005b) Seismic Provisions for Structural Steel Buildings, American Institute of Steel Construction,
Chicago, IL. AISC 341-05.
AISI (2007) Specification for the Design of Cold-Formed Steel Structural Members, American Iron and
Steel Institute, Washington, D.C., AISI-S100-07.
Brown, C.J., and Yettram, A.L. (2000). "Factors influencing the elastic stability of orthotropic plates
containing a rectangular cut-out." Journal of Strain Analysis for Engineering Design, 35(6), 445-
458.
El-Sawy, K.M., and Nazmy, A.S. (2001). "Effect of aspect ratio on the elastic buckling of uniaxially
loaded plates with eccentric holes." Thin-Walled Structures, 39(12), 983-998.
Kesti, J. (2000). "Local and Distortional Buckling of Perforated Steel Wall Studs," Dissertation/Thesis,
Helsinki University of Technology, Espoo, Finland.
Miller, T.H., and Peköz, T. (1994). "Unstiffened strip approach for perforated wall studs." ASCE Journal
of Structural Engineering, 120(2), 410-421.
Moen, C., Schafer, B.W. (2008). “Simplified Methods for Predicting Elastic Buckling of Cold-Formed
Steel Structural Members with Holes” Proceedings of the 19th International Specialty Conference
on Cold-Formed Steel, St. Louis, MO. pp 17-32.
Moen, C.D., and Schafer, B.W. (2006). " Impact of holes on the elastic buckling of cold-formed steel
columns with applications to the Direct Strength Method." Eighteenth International Specialty
Conference on Cold-Formed Steel Structures, Orlando, FL, 269-283.
Tovar, J., and Sputo, T. (2005). "Application of direct strength method to axially loaded perforated cold-
formed steel studs: Distortional and local buckling." Thin-Walled Structures, 43(12), 1882-1912.
Vann, P.W. (1971). "Compressive buckling of perforated plate elements." First Specialty Conference on
Cold-formed Structures, Rolla, Missouri, 58-64.

More new references

Bambach, M. R. (2008). "Inclined yield lines in flange outstands." Structural Engineering and
Mechanics, 29(6), 623-642.
Bambach, M. R., and Rasmussen, K. J. R. (2004a). "Design provisions for sections containing unstiffened
elements with stress gradient." Journal of Structural Engineering, 130(10), 1620-1628.
Bambach, M. R., and Rasmussen, K. J. R. (2004b). "Effective widths of unstiffened elements with stress
gradient." Journal of Structural Engineering, 130(10), 1611-1619.
Bradford, M. A. (1992). "Lateral-Distortional buckling of steel I-Section members." Journal of
Constructional Steel Research, 23(1-3), 97-116.
Bradford, M. A., and Azhari, M. (1995). "Inelastic local buckling of plates and plate assemblies using
bubble functions." Engineering Structures, 17(2), 95-103.
Bradford, M. A., and Hancock, G. J. (1984). "Elastic interaction of local and lateral buckling in beams."
Thin-Walled Structures, 2(1), 1-25.
Brown, C. J., and Yettram, A. L. (1986). "The elastic stability of square perforated plates under
combinations of bending, shear and direct load." Thin-Walled Structures, 4(3), 239-246.
Caccese, V., Elgaaly, M., and Chen, R. (1993). "Experimental study of thin steel-plate shear walls under
cyclic load." Journal of structural engineering New York, N.Y., 119(2), 573-587.
Davidson, J. S., Ballance, S. R., and Yoo, C. H. (1999). "Analytical model of curved I-girder web panels
subjected to bending." Journal of Bridge Engineering, 4(3), 204-212.
Earls, C. J. (1999). "On the inelastic failure of high strength steel I-shaped beams." Journal of
Constructional Steel Research, 49(1), 1-24.
Elgaaly, M. (1998). "Thin steel plate shear walls behavior and analysis." Thin-Walled Structures, 32(1-3),
151-180.
Elgaaly, M., Caccese, V., and Du, C. (1993). "Postbuckling behavior of steel-plate shear walls under
cyclic loads." Journal of structural engineering New York, N.Y., 119(2), 588-605.
Featherston, C. A., and Ruiz, C. (1998). "Buckling of curved panels under combined shear and
compression." Proceedings of the Institution of Mechanical Engineers, Part C: Journal of
Mechanical Engineering Science, 212(3), 183-196.
Ghavami, K. (1994). "Experimental study of stiffened plates in compression up to collapse." Journal of
Constructional Steel Research, 28(2), 197-221.
Gioncu, V., and Petcu, D. (1997). "Available Rotation Capacity of Wide-Flange Beams and Beam-
Columns: Part 1. Theoretical Approaches." Journal of Constructional Steel Research, 43(1-3),
161-217.
Goncalves, R., and Camotim, D. (2007). "Thin-walled member plastic bifurcation analysis using
generalised beam theory." Advances in Engineering Software, 38(8-9), 637-646.
Grondin, G. Y., Chen, Q., Elwi, A. E., and Cheng, J. J. (1998). "Stiffened Steel Plates under Compression
and Bending." Journal of Constructional Steel Research, 45(2), 125-148.
Grondin, G. Y., Elwi, A. E., and Cheng, J. J. R. (1999). "Buckling of stiffened steel plates - A parametric
study." Journal of Constructional Steel Research, 50(2), 151-175.
Hancock, G. J. (1978). "LOCAL, DISTORTIONAL, AND LATERAL BUCKLING OF I-BEAMS."
ASCE J Struct Div, 104(11), 1787-1798.
Hancock, G. J., Kwon, Y. B., and Stefan Bernard, E. (1994). "Strength design curves for thin-walled
sections undergoing distortional buckling." Journal of Constructional Steel Research, 31(2-3),
169-186.
Hitaka, T., and Matsui, C. (2003). "Experimental study on steel shear wall with slits." Journal of
Structural Engineering, 129(5), 586-595.
Kosteski, N., Packer, J. A., and Puthli, R. S. (2003). "A finite element method based yield load
determination procedure for hollow sturctural section connections." Journal of Constructional
Steel Research, 59(4), 453-471.
Kwon, Y. B., and Hancock, G. J. (1991). "A nonlinear elastic spline finite strip analysis for thin-walled
sections." Thin-Walled Structures, 12(4), 295-319.
Lau, S. C. W., and Hancock, G. J. (1986). "Buckling of thin flat-walled structures by a spline finite strip
method." Thin-Walled Structures, 4(4), 269-294.
Lau, S. C. W., and Hancock, G. J. (1987). "DISTORTIONAL BUCKLING FORMULAS FOR
CHANNEL COLUMNS." Journal of structural engineering New York, N.Y., 113(5), 1063-1078.
Lau, S. C. W., and Hancock, G. J. (1989). "Inelastic buckling analyses of beams, columns and plates
using the spline finite strip method." Thin-Walled Structures, 7(3-4), 213-238.
Libove, C., Ferdman, S., and Reusch, J. J. (1949). "Elastic buckling of a simply supported plate under a
compressive stress that varies linearly in the direction of loading." NACA Technical Note No.
1891, 1891.
Lubell, A. S., Prion, H. G. L., Ventura, C. E., and Rezai, M. (2000). "Unstiffened steel plate shear wall
performance under cyclic loading." Journal of structural engineering New York, N.Y., 126(4),
453-460.
Mikami, I., and Niwa, K. (1996). "Ultimate compressive strength of orthogonally stiffened steel plates."
Journal of Structural Engineering, 122(6), 674-681.
Paik, J. K., Thayamballi, A. K., and Lee, J. M. (2004). "Effect of Initial Deflection Shape on the Ultimate
Strength Behavior of Welded Steel Plates Under Biaxial Compressive Loads." Journal of Ship
Research, 48(1), 45-60.
Papangelis, J. P., and Hancock, G. J. (1995). "Computer analysis of thin-walled structural members."
Computers and Structures, 56(1), 157-176.
Peköz, T. (1987). "DEVELOPMENT OF A UNIFIED APPROACH TO THE DESIGN OF COLD-
FORMED STEEL MEMBERS." American Iron and Steel Institute.
Roberts, T. M., and Sabouri-Ghomi, S. (1992). "Hysteretic characteristics of unstiffened perforated steel
plate shear panels." Thin-Walled Structures, 14(2), 139-151.
Sabouri-Ghomi, S., and Roberts, T. M. (1992). "Nonlinear dynamic analysis of steel plate shear walls
including shear and bending deformations." Engineering Structures, 14(5), 309-317.
Schafer, B. W. (2002). "Local, distortional, and Euler buckling of thin-walled columns." Journal of
Structural Engineering, 128(3), 289-299.
Schafer, B. W., and Adany, S. "Buckling analysis of cold-formed steel members using CUFSM:
Conventional and constrained finite strip methods." Eighteenth International Specialty
Conference on Cold-Formed Steel Structures, Orlando, FL, United States, 39-54.
Schafer, B. W., and Peköz, T. (1998). "Cold-formed Steel Members with Multiple Longitudinal
Intermediate Stiffeners." Journal of Structural Engineering, 124(10), 1175-1181.
Schafer, B. W., and Peköz, T. (1999). "Laterally braced cold-formed steel flexural members with edge
stiffened flanges." Journal of Structural Engineering, 125(2), 118-126.
Shahwan, K. W., and Waas, A. M. (1998). "Buckling of unilaterally constrained infinite plates." Journal
of Engineering Mechanics, 124(2), 127-136.
Sheikh, I. A., Elwi, A. E., and Grondin, G. Y. (2003). "Stiffened steel plates under combined compression
and bending." Journal of Constructional Steel Research, 59(7), 911-930.
Shi, Y. J., Chan, S. L., and Wong, Y. L. (1996). "Modeling for moment-rotation characteristics for end-
plate connections." Journal of Structural Engineering, 122(11), 1300-1306.
Silvestre, N., and Camotim, D. (2002). "Second-order generalised beam theory for arbitrary orthotropic
materials." Thin-Walled Structures, 40(9), 791-820.
Smith, S. T., Bradford, M. A., and Oehlers, D. J. (1999). "Local buckling of side-plated reinforced-
concrete beams. I: Theoretical study." Journal of Structural Engineering, 125(6), 622-633.
Yu, C., and Schafer, B. W. (2006). "Effect of longitudinal stress gradient on the ultimate strength of thin
plates." Thin-Walled Structures, 44(7), 787-799.
Yu, C., and Schafer, B. W. (2007). "Effect of longitudinal stress gradients on elastic buckling of thin
plates." Journal of Engineering Mechanics, 133(4), 452-463.