Lehigh University

Lehigh Preserve
Fritz Laboratory Reports Civil and Environmental Engineering

1980

Lrfd, a comparison with allowable stress design and

plastic design, "Visuals," Presented at American
Petroleum Inst. Conf. Lehigh University, September
1980, 34p.
L. S. Beedle

Follow this and additional works at: http://preserve.lehigh.edu/engr-civil-environmental-fritz-lab-

reports

Recommended Citation
Beedle, L. S., "Lrfd, a comparison with allowable stress design and plastic design, "Visuals," Presented at American Petroleum Inst.
Conf. Lehigh University, September 1980, 34p." (1980). Fritz Laboratory Reports. Paper 2032.
http://preserve.lehigh.edu/engr-civil-environmental-fritz-lab-reports/2032

This Technical Report is brought to you for free and open access by the Civil and Environmental Engineering at Lehigh Preserve. It has been accepted
for inclusion in Fritz Laboratory Reports by an authorized administrator of Lehigh Preserve. For more information, please contact
preserve@lehigh.edu.
 D
A Com pari .son with. ' ·~ §
I
Allowlible Stress Des1qn · li ,i
· and ~~ · ·;
Pla.st,·c De.si~r:> · ~~;
: !
. i .
I
i .

StZpt 1980 ftitl: EV)q IV) q€(fVI~ la..bDf't.dory i : Report 371. bv

. · Le~t:~ U~'ltreS i-ly i' .'
 • FUNCTIONAL RE.QUIREME.NTS

• ·suPPORT LOAD J PROVIDE STIFFNESS

• ECONOMICAL REQUIREMENTS

Fig. 1. Structurql Design Objectives. The main objectives of Structural

Design do not change with the design method.
 ""''· 3'71.1
n::::

Steps. in Destqn
1. Function
Z. . Structure and Loadinq
3. Loadinq Conditions
4. Preliminary Desiqn
5. Analysis
6. Selection of Section I ;
: ! . '

7. Secondary Deslqn Check ·

Fig. 2. The steps in design are also independent of the method that is
used •
. -~--- ---· ----- ·---~------------------------------------------------------ .. ·····-------- -- ---- -----------------------------·· ------·
~
 Allowable Stress Design
Working Stress Design
Elastic Design

Plastic Design
Ultimate Strength Design
Limit Design

Load and Resistance Factor Design

Load Factor Design
Limit States Design
- - - - - - - - - ' - - - - - - -----------··--- - - - - - · - - · - - - - - - - - - - - -

Fig. 3. There are three groups of design concepts: The "allowable stress"
group, the "plastic design" group, and the "load and resistance
factor" gr·oup.
 ·•'

,. ·.;.
:: ;: . -:

< frR
:::
.. ~ :.i :

'·-
· ...
. ::

,.'
.:: ....

·<~Q
. .

-
.,<.
. .

,l

!.
.. ·
. ' ;
::: ~ : : ) : ~ . : : . .. : : ; :.

:' ::
. [ :~ . :

Fig. 4. The LRFD formulation is simple: The load factor times the load
effect must be less than the resistance factor times the resistance
of the member.· (The format shown in the second line of the
formulation :l.s typical of that which appears in research papers.)
,
 DEAD LOAD PLASTIC LIMIT
LIVE LOAD P STABILITY
LIMIT
Long Term ELASTIC LIMIT
Short Term - FATIGUE.
FRACTURE
Extraordinary DEFLECTION
I
COMBINATIONS I
I
(VIBRATION)
I

--------------------------- -- --· --·------ ····-

Fig. 5. LRFD involves the examination of the loading function (left) and
the resistance function (right). Design is equating the two through
analytical'techniques and the use of the basic LRFD equation.
 /
/

,L-Ry
I
!!lllli!I!IIIIJ

I
I
I
I-~
a
p p

------------- ----------

Fig. 6. LRFD is compared with allowable stress design at the left. At the
right it is compared with a form of load factor design, first with
single load.factors and then with multiple load factors. The
comparison is, in fact, with a plastic design, except that the use
of multiple-load factors can lead to lighter members.
,
p

Fig. 7. The load deflection curves and load bars of Fig. 6 are simplistic.
Actually there is uncertainty. This figure shows an example of
uncertainty in the loading function fl.
 p

---·----· -------- ·--··· ····-- ·---------------

Fig. 8. The uncertainty in response is shown (fr). Comparing with Fig. 7,

there ~s less variation in response or resistance than there is in
load .
.. ·----------
 pp ~R T

I
I
I
p I
I
I

FD
L R fe L ~ ~R
-.

(yQ~cpR)

--------------------------·-

Fig. 9. Failure is defined according to the following criterion: The

maximum load, fl is less than the minimum possible resistance
(f r x R)..
 .; ::

1.0
LOAD p
psf Per
I 00. . ~---:-·.. . ..... . . :---•.••.•.•.. . ··_.__...._____,
..
~ i :· ~ ; ~ ;. :
:~:::
.. ::: t----'--.-.-.-~------"'---..,

.~ !~ ..
.. ~----------------~
.. .
.. .
: •. ·~------,.-------....J
. . ~ :: ~ j :
::·· :

:1

0. j. NO.
I •• I

: . i

Fig. 10. Actual measured load and resistance data. To the left is shown
the variation in floor load. ro the right, the variation in
resistan~e of continuous beams. These observations illustrate
the greater scatter in load as compared to that of resistance.
=·,·
 ,-------
I
,'
I
I
I

. ,''
I

! .Po.
.!

Rs R·s Ps
I

Load

ASD PD LRFD

D.ef lection
Fig. 11. How safety is achieved in the three design methods. Allowable
Stress Design: Start with yield and come down to allowable.
Plastic Design: Start with the service load and factor up to
design ultimate load. LRFD: Factor up from service load and
factor down 'from nominal resistance of the structure. The arrows
show where design attention is focussed in each of the three
methods.
 .
APPROXIMATIONS IN ANALYSIS
APPROXIMATIONS IN DESIGN
,
STREss· CONCENTRATIONS AND RESIDUALS
VARIATION IN PROPERTIES
VARIATION IN DIMENSIONS
WORKMANSHIP
LOCATION
VARIATION IN LOAD-- TYPE.
c·oMB INATION OF LOADS
INTENDED USE
---- -- --------

Fig. 12. Tabulation of the approximations and uncertainties in design,

workmanship, and loading. These factors must be accommodated
in any design method.
-- ·------·-----·----------·. ·- . - -- - .. . --- - ·-
I :'
. :: ':: .I~·~ : :
!:· ::·
. 3 '7 I 3 }\ '

load Factor

Tension Ft : 0. 60fy 1.67

Bending Fb = 0. 66Fy 1.70

z
·r ,_ (Kh) c
. Compresslon Fa =
2 c: IY
a L67 -l.9Z
_2_ + 3(Kf;r) . (Kf!r)
3 oc, Bc;

Fig. 13. How safety is achieved in Allowable Stress Design for three types
of loading. 'To the right is shown the corresponding load factors.
···----------------·-------- .·- ..... ----------------· .
-------.
, ··-··
 1.3 GRAVITY PLUS LATERAL LOAD

TENSION MEMBERS·
SHORT COLUMNS
1.7 BEAMS
PLASTIC DESIGN

1.9 LONG COLUMNS

RIVETS
BOLTS
2.2 WELDS
SHEAR CONNECTORS
-----------·------

Fig. 14. Uncertainty in Plastic Design is accommodated by load factors.

Note the'rational progression of load factors depending on the
importarice-of member or uncertainty of loading or response.
 ALLOWABLE STRESS PLASTIC
DESIGN DESIGN

P.u - ·- P. -· ---
u
FY INHERENT
MARGIN
OF
SAFETY
,,,,,,,,,,,,,,,,,

CTmax =20 ksi

R'""tj'"·o~

Deflection Deflection

--------------·-------------

Fig. 15. The philosophy behind selection of load factor F in Plastic Design
is this: The same safety in continuous beams as inherent in the
past ASD of simple beams.

r
I : '~

,(!":
::: :::
· '1
_)
'7/ t/

Pp Mp ~l y
F= - ••
Pa Ma Q'"d s

a-., -- 0. 66 (J"'i
z- . f ..!! LIZ
S.

Fig. 16. The ~a1cu1ation of F based on concept of Fig. 15.

:::. !~n:
:··· = ? Il
I ~~;

Q.R

~-----·-------

Fig. 17. Examining safety in LRFD. Variations of load and resistance

indicated in Fig. 9 are shown. Resistance at top, load at
" bottom. What is failure? When Q is greater than R. Area under.
curves (see shaded at left) is related to robabilit of failure.
 Tj[: .) /I lr

Pn < Pfz
P.-Q.

Fig. 18. Safety depends upon two things: The difference in Q and R and
the variability·of Q and R. This figure illustrates the first.

.. .. ·::: ····=··:
 Pfl < Pfz
Variability

Fig. 19. An illustration of how an increase in the variability (in this

case variability of load) increases the probability of failure.

1m11m l.l\~mmm1
 klormal

Event Mean
En Nomina\
(hand book)

Uo~ ol Occurrences
Fig. 20. Some of the terms and functions associated with measuring and
evaluating variability: Normal and skew distributions. Standard
deviations a. · Mean and nominal (handbook) events (Em and En ,
respectively).

l.
Q R.

Rm

Fig. 21. Functions shown are the mean value of load, Q standard deviation
m
of load a , mean-value of resistance R , and standard deviation
of resist2nce aR. Coefficient of vari~tion is ratio of standard
deviat_ion to mean value.

::
 R
Rm I fTR
T
Rm·Gm

Qrn
. - - ·-

I (JQ. l
Q
Saf~ty Marqin: R.-&

0
Fig. 22. Failure can now be defined more specifically. The safety margin
is R - Q, so safety will depend on two things: R - Q and on a.
m m
Area under curve ~s probability of.failure.

f!
 . :3, ..-,, . II

Q RI R-Q
'

Fig. 23. If safety depends on R - Q it can be plotted that way. See on

right.
I ; "~ . ·:.
- -:-:::
.~.

'''T - 3/1 _I,?J

·:::·

R-Q

II"'J
+ UQ

Fig. 24. B is defined as the "safety index" or "reliability index". The

relationship shown is only true for normal distributions.
 ol:,

:: , ..

R
ln-
·Q
-

-
0

Fig. 25. The relationship of S to R , Q , and standard deviation can also

· m m
be express~d in terms of logar1thmic functions of R and Q.
.: ':.

Beams Connections
f3 = 3. 0 f: 4.5

3.0 cr
4.5 cr
0

0 -
,~~----·

Fig. 26. "Calibration" is achieved by comparing the 8-v<~lues with what

would be obtained in the design of a beam by Allowable Stress
Design. Two cases are shown.
 Rm
Rn
I

"Related"
to t5
0

Fig. 27. Although the specific comparison is not indicated, this illustrates
the fact that F x R and F x Q are related to 8 .
.r q

. :. : ::: : ... ·:~. .

 ••

Fig. 28. The relationship of load and resistance factors with B is shown
mathematically.

·· ·: ·· ,. ·. ,·· ,, ·: · ••• •· ···i.ri:!!!Hlln\l11lT!n :l:!~ill 1 -:i :::1: 1::::·: :r:m:::::! :!:! ::·: ::1::.:: ~::: ·::l 11::: 1: ·:: ·. ·: •1:1 ·1:! ·•::,.:: ·: •: :~,: :1, ••.:: T1:: •1·:: 1: •r ·:! 1:r::.: m.::: •·: ·•·····' ·: ·: •: •·••, •
-Yf-.·(t3
I\ ,.::::

F.L <
-
t

Fig. 29. The simplification of the LRFD format shown earlier, is expanded
here to show the multiple-load-factor aspect.
 Load Combinations -- Load Factors·
_D_e_,a_d___;_____l__iv_rz__S_noY!______ '~~-~~9 -~o'-!h~u~ak~-----
;1.4 ·- • - -
l.2 1.6 - - -
1.2. i
0.5 I.6 • ! -

I.Z - 1.6 0.6 -

I.Z 0.5 - I ,3 •

I I
I

'.2. 0.5· - - t5'

I.Z - O.Z - :t5
- 0.9 - (. 3 -
- 0.~ -
---------:----~---'-----------------·----
- - ;1.5 . !
: , ! I

- - - 'i~~- .. -~ ..1-~---·--
I

Fig. 30. A possible ~et of load combinations and load factors for use in
LRFD.
;·1 -rr-~ •r:-T

::: II :: :i. ,'; .-'~ _.-'~ .-'~ ~ lj ~

H1l
 Resistance Factor&
Element Limit State Fr (¢)

Tension Member . YiQid 0.9

Fracture .8

Beam Bendinq (Mp.M,r) .9

Shear (Vp.Vcr.) .9
Co{umn I· shape · Stability . ,8
Other Stability .8

· Weld 6roove Fracture .9

Fillet Shear Fracture .s
Bolt Shear Fracture .7

Fig. 31. Possible resistance factors for use in LRFD. These have been
simplified and rounded off.
.. .. :·- ...

, r '~@ mmrm:
•:; •;
. . . ; .·: .

mmml"· u1·1l
 P: ZIf( 0:9
l: ll

fo Q1> + FL QL <
- fr lln - zo'

-PL4
f0 : L2 fL:. (.6 . FR. = 0.~
Qo: Mo QL: ML Rn: t!y 'i

l.Z M0 + I. 6 ML - 0.9
(.. <fy'l:.
<CJlzol< , o~<1oKrz)
12 .
12.)
4 t 1' 4 -
.( 0.9(36) c

Fig. 32. Some of the. essentials of the LRFD method are shown in this
design example.
. . . ' ' ::· ' :··

I, H II II
 64.0
60 6Z. z 59 9
59. 6 .
46.7
40 Save cost
·,

zo

0
0 L
-
All Dead Load Ot\. All liv~ load
Fig. 33. The required plastic modulus according to three different designs:
3 3 3
ASD (59.9 in. ), PD (59.6 in. ). LRFD (46.7 to 62.2 in. ,
depending on ratib of dead load to live load). Evidently the
multiple load factor aspect of LRFD has a great deal to do with
~hether material will be saved or not.