Vous êtes sur la page 1sur 20

Reliability of Partially

Prestressed Beams at
Serviceability Limit States
Antoine E. Naaman
Professor of Structural Design
Department of Civil Engineering
Mechanics and Metallurgy
University of Illinois at Chicago
Chicago, Illinois

Amnuayporn Siriaksorn
Project Engineer
Engineering Company, Ltd.

the variables at hand. One of them re-

onventional design methods re-
quire that a structure be designed quires the knowledge of the entire dis-
so that its predicted capacity is larger tributions of resistance and loads (Level
than the specified design loads. In this III approach) and is, therefore, quite
approach (called Level I Design) it is complex and tedious to apply; the other
assumed that all variables and factors one (Level II), called the first order
are deterministic and their values ex- second moment (FOSM) method, has
actly known. It is generally accepted, been proven to apply with little diffi-
however, that the basic variables used culty to any type of structural mem-
in the prediction equations such as ber.s,s,io,11,
materials properties, section geometry, The Level II approach requires the
and the like are random variables and knowledge of only the mean and vari-
thus are characterized by a probability ance of the distributions of the resis-
distribution function. tance and load, and it has been shown
The concepts of reliability theory are to lead to results not significantly dif-
being increasingly applied in structural ferent from the more exact Level III
engineering not only to assess the approach. In the FOSM method a
probability of survival of a structure but safety index or reliability index, 6i, is
also to determine rational values of load needed as an input variable specified
and resistance factors prescribed by by the code authority.
various building codes. 119 Two design Although in a number of previous
methods have been so far proposed to studies the reliability index /3 has been
account for the probabilistic nature of obtained for concrete or steel structures

at the ultimate limit state, 7' 9' 13, 17,i 8, 2o lit-
tle has been done to evaluate /f3 at ser-
viceability limit states. In this investi-
gation the reliability index /3 of partially
prestressed concrete beams at six ser-
viceability limit states is evaluated. The
reliability index a
is directly related to
the probability of failure or unservice-
ability where failure does not necessar-
ily refer to collapse but to violation of
any code specified limit; hence it im-
plies abnormal levels of maintenance
and repair during service life.
The six states considered are: maxi-
mum crack width, short- and long-term
deflection and fatigue stress ranges in
the concrete, the reinforcing steel and
the prestressing steel. In order to pro-
vide a basis of comparison, 3 was also
partly evaluated at the ultimate flexural
strength limit state where it is related to
the probability of failure or collapse.
In order to determine the /3 values for
typical design situations, the method
called Code Calibration 12,21 is used. Re-
sistance prediction equations at each
serviceability limit state are developed
in terms of moments. A Monte Carlo
simulation model is used to compute
the statistical properties of the resis-
tance at each limit state. The statistical
properties of the load are collected from
Determination of
available literature22 -29 and used in the Reliability Index
evaluation of the reliability index. In classical reliability theory the total
The effects of many parameters on load (S) and the resistance (R) are con-
the observed values of the reliability sidered random variables; thus they can
index a are analyzed. They include the be characterized by their probability
effect of various types of beam cross distribution functions. Assuming S and
sections (representative of those used R are dimensionally consistent, the
in the American precast prestressed in- failure event is described by the rela-
dustry), span length, magnitude of live tion:
loads, amount of non-prestressed rein-
forcement and some important materi- R < S (1)
als properties such as concrete com- or
pressive strength. A systematic com-
parison with the boundary cases of fully RS<0 (2)
reinforced or fully prestressed beams is A new random variable m termed
also presented. A summary of the safety margin or reliability margin can
method used and results obtained is be defined [see Fig. 1 (top)] as:
given next. Detailed information can be
found in Ref. 30. m =RS (3)

PCI JOURNAL/November-December 1982 67

f(S) I f (R)

S x LOAD i i PR R s RES
rF w = R- g --^

ISM s /R- g

QM - R^Qg

0 16*

m= Margin of Safety
Fig. 1. Illustration of concept of reliability index.

Thus, failure is associated with m < 0 or

and survival is associated with m > 0.
The mean value of m and its standard ! - Rim = 0 (7)
deviation can be derived from those of
R and S as follows:The above equation at equality com-
bined with Eqs. (4) and (5) can be used
(4) to determine /3:
Mm = AR - s

An - As
Qm = QR + - QS (5) Qm QR + Qs

In order to maintain a prescribed The coefficient /3 is called the "relia-

value of the probability of survival (or bility index" or safety index. It is the
reliability), that is m _- 0, it may be re- distance from the mean of the prob-
quired that the mean value of m be ability distribution function of n to the
larger than or equal to a coefficient /6 origin in standard deviation units [see
times its standard deviation, that is: Fig. 1 (bottom)].
In the above treatment "failure" does
A. a0m (6) not necessarily mean collapse (or the

Table 1. Serviceability Limit States Considered and Their
Specified Limitations.

Limit states Specified

No. Symbol Description

1 Wmax Maximum crack width 0.016 in.

2 Concrete fatigue fm{R

0'4 f. -
stress range 2

3 f,., Non-prestressed steel

fatigue stress range 20.0 ksi

4 fir Prestressing steel

fatigue stress range 0.1 fD

5 all. Immediate live load l

deflection 180
6 add Additional long-term l
deflection 240

*Note: 1 in. = 25.4 mm; 1 ksi = 6.9 MPa.

attainment of an ultimate limit state) Step 1 are collected. The statistical

but may also imply unserviceability or properties (p., v) of these variables are
violation of a serviceability limit state defined.
such as a maximum deflection limita- 3. The value of each limitation for
tion or a maximum crack width limita- each limit state is selected from current
tion. The serviceability limit states con- codes or recommended practices, e.g.,
sidered in this study are described in the allowable maximum crack width is
the next section. taken as 0.016 in. (0.41 mm) for maxi-
The following steps were followed to mum crack width limit state as per the
evaluate the reliability index /3 of rein- ACI Code (see Table 1).
forced, prestressed and partially pre- 4. Using Steps 1, 2 and 3, and apply-
stressed beams at each serviceability ing a Monte Carlo simulation method,
limit state: the statistical properties of the resis-
1. The resistance prediction equa- tance ( p
., o) of the beam at each ser-
tions of the partially prestressed con- viceability limit state are computed.
crete member at any serviceability limit 5. Statistical data on live loads are
state are derived so that the resistance collected from available literature.
has the same unit as the applied loads Their statistics (L , o) are determined.
(i.e., moment as a function of crack Similarly, the statistical properties of
width or deflection limit). The method dead load (D , 0 D ) are determined.
and equations developed in Ref. 31 6. From Steps 4 and 5, the reliability
were used. The resistance at each limit index, /3, of the beam at each limit state
state is expressed in terms of basic vari- is calculated using Eq. (8).
ables and allowable maximum limita- A detailed description of the above
tion at that limit state. procedure is given in Ref. 30. The last
2. Statistical data on the variables three steps are clarified further in the
used in the equations developed in following sections.

PCI JOURNAL/November-December 1982 69

Limit States Statistical Properties of
"When a structure or structural ele-
ment becomes unfit for its intended In general if a random variable can
purpose it is said to have reached a be expressed in function of other (basic)
limit state."16 Limit states are generally variables, its mean value and COV
divided into two categories, ultimate (coefficient of variation) can be esti-
limit states and serviceability limit mated by expanding the function in a
states. Taylor series about the mean values of
Ultimate limit states are related to the other (basic) variables, truncating
a structural collapse of part or all of higher order terms, and taking the ex-
the structure. They are points that lie pectations of that function. Thus, its
on the failure surface of the structure, mean and COV can be expressed in
such as ultimate moment, ultimate terms of the means and COV's of the
shear, ultimate ductility, etc. Service- basic variables which appear in the
ability limit states are related to dis- function.
ruption of the functional use of the This approximation is reasonably ac-
structure, and/or its damage or deteri- curate, provided that the COV's of the
oration during service. Examples in- basic variables are not too large and the
clude excessive cracking, leakage, and equation is not highly nonlinear.11,13
deflections which lead to abnormal However, the above approach could not
maintenance and/or repair levels during be applied to compute the mean and
service. the COV of the resistance of partially
In this study the reliability of par- prestressed concrete beams at each ser-
tially prestressed beams was evaluated viceability limit state since the predic-
at six serviceability limit states. They tion equation of the resistance cannot
are summarized in Table 1 where the be always expressed in closed form.
limitations used in the parametric anal- In this study, relationships between
ysis for each limit state are also given. the resistance of partially prestressed
These limitations are in accordance beams at serviceability limit states and
with the ACI Building Code, 32 or vari- corresponding loading conditions were
ous ACI recommended practices. Also, first developed and expressed in .terms
in a limited number of cases and for of moments. The statistical properties
comparative purposes the reliability of the basic variables present in these
index at ultimate flexural strength limit relationships were collected from avail-
state was evaluated. able technical literature. 7,11, 13,20, 33 37
A note is in order here and is related Selected properties used in the relia-
to the manner in which the limit state bility analysis are summarized in Table
fatigue was handled. Fatigue limit state 2. A Monte Carlo simulation where val-
can be classified either as: (1) an "ulti- ues of the basic variables are selected at
mate limit state" because failure by random was used to determine the
fatigue results in the collapse of the mean and the coefficient of variation of
member or (2) a "serviceability limit the resistance. Each simulation com-
state" because it occurs during service prised 200 iterations.
loads and may have secondary effects Since the coefficient of variation (SR)
on other serviceability limit states 'such of the resistance obtained from the
as increases in crack widths, deflections Monte Carlo simulation fails to include
and debonding. Here the second ap- sources (such as incomplete informa-
proach was used. The meaning of /3 in tion, insufficient data, inaccuracy of
that case is clarified below in the sec- prediction equations, etc.) that contrib-
tion under "RESULTS." ute to the total variability (E R ) of the re-

Table 2. Summary of Statistical Properties of Variables Used in the Reliability analysis.
Variables used Mean, COV., fl Remarks

b,bb,, Normal 0 to + 5 + b l 0.0 0.045

L 32 J

h,h fh Normal hR g to h+ 8 4
6.4 Fi

Normal dp to d1 + 16 0.04 8

11 I1
l,a Normal ^",a" -
32 16
nominal = 33
CE Normal 33.6 0.1217 E, =CEO (y, ) 15 f^

nominal = 7.5
Cry Normal 9.374 0.0938 f, = CJ r f^

f^ Normal 0.67 fc to 1.17f 0.1 0.25

yc Normal yc 0.03
nominal = 0.8
Cf^ 1 Normal 0.6445 0.073 f , = C,^tf,
Beta 1.07fR-1.19fv 0.09-0.15
A, Normal 0.9A,1, 1.01 A,. 0.015 0.04
E, Normal Ee 0.024 0.033
A, Normal 1.01176Ap81, 0.0125 APS = 0.153 in?
fp Normal 1.0387 fp R 0.0142 fpn = 270 ksi
fp Normal 1.027f,,,,, 0.022 f,, = 240 ksi
EP$ Normal 1.011 EP,, 0.01 E,,,, = 29000 ksi

Cf1 =0.7
C,, , Normal Cffin 0.08 f., = C f scfn,,
C, , =0.83
C, Normal CrSen 0.08 f8, = CfBefaf

YIO Normal Y.O 0.03

d,. Normal d,,,, 0.07

sistance, a prediction error (ER ) was Prediction errors assumed in this

considered. The prediction error, ER , for study are given in Table 3. Two limits
each limit state can be estimated as the are shown for the limit states "crack
coefficient of variation of the ratio of width" and "additional long-term de-
experimentally observed to theoreti- flection" reflecting various experimen-
cally predicted resistance values. tal results analyzed. The total variabil-

PCI JOURNAL/November-December 1982 71

Table 3. Prediction Errors of the weight of the structure. A normal
Assumed for Each Limit distribution with a coefficient of varia-
State. tion of 0.1 was assumed for the super-
imposed dead load, if any. In a treat-
Limit State eR
ment similar to that of the resistance,
Wmax 0.1or0.2 the mean D and the coefficient of vari-
ation SD of dead load were obtained
fc r 0.05
from the simulation; a prediction error
}rr 0.05 ED = 0.1 was considered leading to a
total coefficient of variation of dead
ftr 0.05
load given by:
DLL 0.05
SZD = Sp + Ep (10)
Ladd 0.1 or 0.2
Live loads are generally associated
with moving or movable loads such as
occupants and furniture. The intensity
ity, R, of the resistance was then ob- of maximum live load depends on the
tained from: type of occupancy, the tributary area
and the projected service life of the
SZR = Ss + ER (9)
structure. Live load intensity is in gen-
where eral measured as the equivalent uni-
SR = value of coefficient of variation formly distributed load which is the
obtained from the Monte Carlo load that wiII produce the same load
simulation. It is affected by effect as the actual loads. Based on
inherent randomness and live load survey data, 23, 26 probabil-
variabilities of the basic istic models of live loads and/or
variables their statistical characteristics were de
Eg = prediction error veloped 20.21,22.24,27
Ravindra, et al,12 expressed the mean
t, and variance (or coefficient of varia-
tion (IL ) of lifetime maximum live load
Statistical Properties of as functions of tributary area and num-
ber of tenancies (Table 4). A prediction
Loading error E L = 0.1 was included in the
Loads can be considered random evaluation of the coefficient of variation
variables and are generally divided into
L. Their proposed values of L and 11L
two groups termed "dead load" and were adopted in this study and led to
"live load." the consideration of sixteen sets of val-
The "dead load" comprises the ues of (L, (I,) depending on tributary
weight of the structure itself and the areas and number of tenancies.
weight of non-structural elements
(often called superimposed dead load)
attached to the structure such as parti-
Parametric Analysis
tions, curtain walls and the like. A In order to determine the reliability
number of studies have suggested val- index /3 of partially prestressed concrete
ues of mean and coefficient of variation beams, the following input parameters
of "dead load. " 11 , 12, 14,15,21 These are were studied (see Table 5 and Fig. 2):
summarized in Ref. 30. Eight different beam cross sections that
In this study, however, a Monte Carlo are representative of sections widely
simulation was used to generate from used by the precast prestressed con-
the basic random variables the statistics crete industry in the United States; for

Table 4. Mean and COV. of Lifetime Maximum Office Live Load,
PL (PSf), flL'
Number of L Tributary area, sq ft
tenancies or
f' L 50 350 800 1200

L 13.5 12.7 12.5 12.3

OL 0.8360 0.5787 0.5295 0.4900

L 19.6 16.8 16.1 15.6

flL 0.4800 0.3736 0.3544 0.3350

L 27.6 21.5 20.5 19.6

S2L 0.3640 0.3160 - 0.2970 0.2880

L 33.4 25.4 24.1 22.6

SZL 0.3544 0.2970 0.2880 0.2880

*See Ref. 45. Note: 1 sq ft = 0.093m2.

each beam three values of partial pre- highest values of the mean, and the
stressing ratio (PPR) were explored, lowest and highest value of the COV.
namely, PPR = 0 (corresponding to These led for each beam and each
fully reinforced concrete), PPR = 1 limit state to six sets of values of mean
(corresponding to fully prestressed con- (,) and coefficient of variation (a,,) of
crete) and PPR = optimum (corre- the resistance. In evaluating the relia-
sponding to partially prestressed con- bility index /3, only the lowest and
crete with one limit state binding?9 highest values of both R and SIR were
various span lengths and representative used leading essentially to four combi-
live loads were also considered. In all, nations of R and d2R for each applied
64 different beam designs were ana- loading (s and fl.).
lyzed each for six different serviceabil- The mean value (s) and coefficient
ity limit states. In some cases the relia- of variation ((l. ) of the load S were de-
bility index at ultimate flexural strength termined from corresponding statistics
limit state was also determined. of dead load and live load as:
A note is in order to understand the
N-s = MAD + AL (11)
way in which the results of the para-
metric analysis were gathered.
Ors = QD + o (12)
In collecting from a multitude of
source s,7.11,13,20,33.37 information on the
mean and COV (coefficient of variation
defined as v/) of the basic variables, it
was generally found that a range be- where v stands for standard deviation.
tween two extreme values could be The statistics of dead load (D , SID)
identified for each variable. In this were obtained from a Monte Carlo
study a Monte Carlo simulation run of simulation in a treatment similar to that
the resistance was used for each of the used for the resistance R (see preceding
following six combinations of statistics section). In evaluating s , fl, and (3
of the variables: the lowest, mean and four sets of (p.s , f1D ) were considered

PC[ JOURNAL/November-December 1982 73






Fig. 2. Typical cross sections of beams analyzed.

Table 5. Design Cases Studied in the Reliability Analysis.

Beam Type of Live load* Span length, 1 (ft)*
No. beam lb per ft PPR Remarks
30 40 50 60 70
0 X X X
Ex. Single-T 400 Optimum X X X X
1 X X X
0 X
1 4 DT 14 200 Optimum X L= 35 ft
1 X
0 X X
2 8 DT 20 400 Optimum X X X X
1 X X
0 X
3 8 DT 24 400 Optimum X
1 X
0 X
4 8 ST 36 400 Optimum X
1 X
0 X
5 10 ST 48 500 Optimum X L = 100 ft
1 X
0 X X X X X
6 Rectangular 50 Optimum X X X X
slab 1 X X X X X
0 X X X X
7 Hollow- 200 Optimum X X X X
core slab 1 X X X X

Note: 1 ft = 0.305 m, 1 lb per ft = 14.59 N/rn.

Table 6. Average values of Reliabili ty Index fi for All Beams.

Limit states ER Average Average Average

maximum minimum me an

0.1 5.6754 1.1114 4.1097

Maximum crack
width, Wm a z 0.2 3.4204 0.7729 2.5265

Concrete fatigue
stress range, 0.05 3.8512 0.9890 2.6466

steel fatigue 0.05 2.3917 1.2137 1.5480
stress range,

steel fatigue 0.05 3.3612 1.9779 2.4116
stress range,
i t,.

Immediate live
load deflection, 0.05 5.9008 3.4526 4.7303

Additional long- 0.1 3.7395 0.4549 2.7722

term deflection,
0.2 2.3348 0.3953 1.8322

corresponding to the combination of of 13 were recorded. These were then

lowest and highest D with lowest and averaged for all beams leading to an av-
highest SZo. erage minimum, average maximum, and
The intensity of maximum live load average mean oft (Table 6).
depends on the type of occupancy, the
tributary area and the projected service
life of the structure. To account for
these parameters, the 16 different sets The results reported in Table 6 rep-
of values of (PL , SZL ) recommended in resent in a way, at each serviceability
Ref. 12 were considered in evaluating limit state, average ranges of reliability
s and U. (Table 4). Hence, in com- indexes /3 for the precast prestressed
puting the mean and coefficient of vari- concrete industry as a whole, assuming
ation of load (s, 11, ) from Eqs. (11-13), the current ACI Code is used. They are
64 combinations of dead and live load plotted in Fig. 3 and compared to simi-
statistics (4 x 16 = 64) were considered lar data generated for one beam (the ex-
for each beam and each limit state. ample beam) at ultimate flexural
To evaluate the reliability index 13, strength limit state.
the 64 combinations of statistics of load It can be observed that (1) for each
S were associated with the four combi- serviceability limit state the reliability
nations of statistics of resistance R. For index spans a relatively wide range of
each beam and each limit state, the values reflecting uncertainties in the
minimum, maximum, and mean value data and/or the prediction equations, (2)

PCI JOURNAL/November-December 1982 75


2 3 4 5



_ (1)4
4 c1- J

U a
W ( j




ZCl' W I I I

JD W I (

0.5 Id' 102 I03 I04 IOO 10 6 -7 -8


Fig. 3. Reliability index values representative of precast prestressed and partially

prestressed concrete beams.

serviceability limit states are on the av- the following causes: (1) fatigue is a
erage more critical than ultimate critical design condition and should be
flexural strength limit states, and (3) on controlled better, (2) the design limita-
the average, among various service- tions are too stringent, and (3) the qual-
ability limit states, fatigue in the steel ity control and workmanship associated
(reinforcing on prestressing steel) is with the influencing variables are poor.
most critical. To better focus on the value of /3 for a
While there is little ambiguity about fatigue limit state, the COV of in-
the meaning of /3 at ultimate flexural fluencing variables can be reduced by
stength limit state which implies failure improving their quality control and
or collapse, the meaning of /3 at service- workmanship (such as achieved in the
ability limit states is more subtle. This precast prestressed concrete industry)
is particularly true for fatigue since and the fatigue stress range limit can be
fatigue can be classified either as a ser- relaxed to reflect actual experimental
viceability limit state or as an ultimate data instead of a code limitation.
limit state. Reaching a serviceability limit state
Let us consider, for instance, the av- does not imply collapse of the struc-
erage mean value of p = 1.548 obtained ture; however, it may lead to "extensive
in Table 6 at the non-prestressed steel damage" and for all practical purposes
fatigue stress range limit state for which "failure." For instance, exceeding the
a code stress range limit of 20 ksi (138 maximum allowable crack width in a
MPa) was used. Assuming a normal liquid retaining structure may lead to
distribution for the safety margin leads extensive leakage and practically fail-
to a probability of 6 percent of exceed- ure of the structure to perform the ser-
ing the code limit each time the random vice it is intended for. Similarly, ex-
variable live load is applied. As the ceeding an allowable deflection does
code limit of 20 ksi (138 MPa) does not not lead to the collapse of a beam but
represent the real fatigue resistance of may lead to the failure of non-structural
the steel material for 2 million cycles of elements attached to it. Hence, the
load repetition (assumed representative meaning of reliability at any service-
of service life), the above probability ability limit state should be examined
does not necessarily imply failure dur- with the proper perspective.
ing service life or 2 million cycles. Typical results of the variation of the
Let us assume that the non-pre- reliability index a with each of the pa-
stressed steel fatigue stress range limit rameters studied (PPR, span length,
of 20 ksi (138 MPa) is changed to 30 ksi live load) are shown in Figs. 4 to 7 and
(207 MPa) to reflect the actual resis- are explained in the conclusions. More
tance of the material up to 2 million cy- detailed information can be obtained
cles. The corresponding mean value of from Ref. 30.
/3 becomes 2.32 (see Ref. 30) and the
probability of exceeding the stress CONCLUSIONS
range limit for each load application is
about 1 percent; it also implies that The following conclusions were
there is a 1 percent chance of failure by made (see Ref 30) in relation to the re-
fatigue of the steel during service -life or liability index /3 of partially prestressed
prior to 2 million cycles of load appli- beams; note that not all these conclu-
cation. sions are supported in the figures of
Note that the values of /3 obtained at this paper.
the three fatigue limit states (Table 6 1. The serviceability limit state
and Fig. 3) are relatively low. This may which controls the design in the deter-
be due to either one or a combination of ministic procedure is not necessarily

PCI JOURNAUNovember-December 1982 77

the one that controls in the reliability 2. The /3 values for rectangular
approach. This is because in the prob- beams are much lower than those for
abilistic approach, the uncertainties in typical T-beams in most of the service-
the values of the basic variables are ability limit states considered. Thus,
taken into account. the probability of failure or un'service-

Limit State For:
1. Maximum crack width (Wm.)
2. Concrete fatigue stress range (fe,)
3. Nonprestressed steel fatigue stress range (f)
4. Prestressing steel fatigue stress range (ftr)
5. Immediate live load deflection (15 LL)
6. Additional long-term deflection (ate)

W 5
} 6

J_ 2
. 6
m 4
Q 2
W 3^ X04


0.0 0.2 0.4 0 .6 08 1.0



Fig. 4. Typical variation of the reliability index with the partial prestressing ratio.

ability of rectangular beams (or one- at both "additional long-term deflec-
way slabs) is generally greater than the tion" limit state and "maximum crack
probability of failure or unserviceability width" limit state decrease substan-
of typical T-beams (Fig. 4). tially when PPR decreases.
3. For typical T-beams and for rec- 5. Everything else being equal, it
tangular beams, the p values at "addi- appears that when the span length in-
tional long-term deflection" limit state creases, the /3 values increase at the
decreases substantially when the partial "maximum crack width," "non-pre-
prestressing ratio (PPR) decreases. stressed steel fatigue stress range" and
Smaller variations in the f3 values are "prestressing steel fatigue stress
observed at other limit states (Fig. 4). range," limit states and decrease at the
4. For hollow-core slabs, the /3 values "concrete fatigue stress range," "im-

Limit State For:
1. Maximum crack width (Wma,,)
2. Concrete fatigue stress range (fcr)
3. Nonprestressed steel fatigue stress range (fi)
4. Prestressing steel fatigue stress range (f)
5. Immediate live load deflection (A)
6. Additional long-term deflection (A)

W 8
6 5
J 6 .
W 4 6

4 a

0 50 60 70
Fig. 5. Typical variation of the reliability index with span length.

PCI JOURNAL/November-December 1982 79


PPR = Partial Prestressing Ratio


} 4
Q 2

30 40 31) 60 70


Fig. 6. Reliability index at maximum crack width limit state for various PPR (example

mediate live load deflection" and "ad- PPR) and the corresponding probability
ditional long-term deflection" limit of failure assuming a normal distribu-
states (Figs. 5 and 6). tion for the safety margin are (see Table
6. Everything else being equal, it 6), respectively, 4.11 and 2 x 10- g at
appears that when the nominal live "maximum crack width" limit state;
load increases, the /6 values increase at 2.65 and 400 x 10- 5 at "concrete fatigue
"maximum crack width" and "pre- stress range"; 1.55 and 6100 x 10- 5 at
stressing steel fatigue stress range" "non-prestressed steel fatigue stress
limit states and decrease at "concrete range" limit state; 2.41 and 800 x 10- 5 at
fatigue stress range" and "additional "prestressing steel fatigue stress range"
long-term deflection" limit states (Fig. limit state; 4.73 and 0.1 x 10- 5 at "im-
7). mediate live load deflection" limit
7. The average /3 values for all par- state; and 2.77 and 280 x 10-5 at "addi-
tially prestressed beams (at optimum tional long-term deflection" limit state.

Limit State For:
1. Maximum crack width (W)
2. Concrete fatigue stress range (fa)
3. Nonprestressed steel fatigue stress range (ff)
4. Prestressing steel fatigue stress range (ftr)
5. Immediate live load deflection (ALL)
6. Additional long -term deflection (a)

S 5



2- 6
-J 3 I. 4
W 4.

3 3

50 100 ISO 200


Fig. 7. Typical variation of the reliability index with nominal live load.

PCI JOURNAL/November-December 1982 81

8. Serviceability limit states in rein- ticularly true for crack width and
forced, prestressed and partially pre- long-term deflection.
stressed concrete beams are more criti- In the analysis of stresses static
cal than their ultimate flexural strength short-term loading was assumed. The
limit state. Violating serviceability limit effect of time on redistribution of
states generally leads to abnormal stresses was not considered. Substantial
maintenance and repair levels during changes in stresses and crack widths
service and may lead, for all practical occur when creep and shrinkage of
purposes, to failure. concrete are considered in the analysis
It is finally observed that reliability- of the section.
based design (Level II approach) can There is a need to develop a proce-
be successfully applied to prestressed dure where the cumulative damage due
and partially prestressed concrete to fatigue under the repetitive applica-
beams provided an appropriate relia- tion of random levels of live loads can
bility index, /3, is recommended by the be accounted for in the reliability anal-
code for each limit state and provided ysis. Such cumulative damage should
the mean and the COV values of resis- affect not only the fatigue life of the
tance and loads are known. materials involved but also other ser-
viceability limit states such as crack
widths, debonding and deflection.

The limitations at various service- This study is based on the PhD thesis
ability limit states used in this investi- of A. Siriaksorn prepared under the di-
gation were taken from the 1977 ACI rection of the senior author. It was sup-
Code. In general, these limitations are ported in part by a Research Fellow-
lower bound values to actual data. It ship Award from the Prestressed Con-
appears that a more accurate estimate of crete Institute and by the University of
reliability at these limit states would be Illinois at Chicago. Their support is
achieved if actual data were used in- gratefully acknowledged. The authors
stead of code limitations. are also indebted to Dr. James G. Mac-
There is a need to develop a method Gregor of the University of Alberta for
to incorporate in the reliability analysis his careful review of and constructive
the often great uncertainty associated comments on the first draft of this
with a prediction equation. This is par- paper.

NOTE: Discussion of this paper is invited. Please submit

your discussion to PCI Headquarters by July 1, 1983.

1. Freudenthal, T. V., Garrelts, J., and Design," Journal of the Structural Divi-
Shinozuka, M., "The Analysis of Struc- sion, ASCE, V. 100, No. ST9, Sept. 1974,
tural Safety," Journal of the Structural pp. 1789-1811.
Division, ASCE, V.92, No. ST1, Feb. 13. Ellingwood, B., "Statistical Analysis of
1966, pp. 267-325. RC Beam Column Interaction," Journal
2. Warner, R. F., and Rabaila, A. P., of the Structural Division, ASCE,
"Monte Carlo Study of Structural V. 103, No. ST7, Proc. Paper 13061, July
Safety," Journal of the Structural Divi- 1977, pp. 1377-1388.
sion, ASCE, V. 94, No. ST12, Proc. 14. Ellingwood, B., "Reliability-Based Cri-
Paper 6275, Dec. 1968, pp. 2847-2859. teria for Reinforced Concrete Design,"
3. Shah, H. C., "The Rational Probabilistic Journal of the Structural Division,
Code Format," ACI Journal, Proceed- ASCE, V. 105, No. ST4, Proc. Paper
ings, V. 66, No. 9, Sept. 1969, pp. 689- 14480, April 1979, pp. 713-727.
697. 15. Ellingwood, B., "Reliability of Current
4. Sexsmith, R. B., and Nelson, M. F., Reinforced Concrete Designs," Journal
"Limitations in Application of Prob- of the Structural Division, ASCE,
abilistic Concepts," ACI Journal, Pro- V. 105, No. ST4, Proc. Paper 14479,
ceedings, V. 66, No. 10, Oct. 1969, pp. April 1979, pp. 699-712.
823-828. 16. Ellingwood, B., Galambos, T. V., Mac-
5. Benjamin, J. R., and Lind, N. C., "A Gregor, J. G., and Cornell, C. A., "De-
Probabilistic Basis for a Deterministic velopment of a Probability-Based Load
Code," ACI Journal, Proceedings, V. 66, Criterion for American National Stan-
No. 11, Nov. 1969, pp. 857-865. dard A58," National Bureau of Standards
6. Cornell, C. A., "A Probability-Based Special Publication 577, U.S. Depart-
Structural Code," ACI Journal, Pro- ment of Commerce, Washington, D.C.,
ceedings, V. 66, No. 12, Dec. 1969, pp. June 1980, 228 pp.
974-985. 17. Shinozuka, M., and Yao, J. T. P. (Edi-
7. Allen, D. C., "Probabilistic Study of tors), "Probabilistic Methods in Struc-
Reinforced Concrete in Bending," ACI tural Engineering," Proceedings of the
Journal, V. 67, No. 12, Dec. 1970, pp. ASCE Symposium, St. Louis, Oct. 1981,
989-993. American Society of Civil Engineers,
8. Ang, A. H.-S., "Structural Risk Analysis 1981, 409 pp.
and Reliability-Based Design," Journal 18. Galambos, T. V., Ellingwood, B., Mac-
of the Structural Division, ASCE, V. 99, Gregor, J. G., and Cornell, C. A., "Prob-
No. ST9, Proc. Paper 10011, Sept. 1973, ability-Based Load Criteria: Assessment
pp. 1891-1910. of Current Design Practice," Journal of
9. Galambos, T. V., and Ravindra, M. K., the Structural Division, ASCE, V. 108,
"Tentative Load and Resistance Factor No. ST5, May 1982, pp. 959-977.
Design Criteria for Steel Buildings," 19. Ellingwood, B., MacGregor, J. G.,
Structural Division Research Report No. Galambos, T. V., and Cornell, C. A.,
18, Washington University, St. Louis, "Probability-Based Load Criteria: Load
Missouri, Sept. 1973. Factors and Load Combinations," Jour-
10. Ang, A. H.-S. and Cornell, C. A., "Relia- nal of the Structural Division, ASCE, V.
bility Bases of Structural Safety and De- 108, No. ST5, May 1982, pp. 978-997.
sign," Journal of the Structural Divi- 20. Grant, L. H., Mirza, S. A., and Mac-
sion, ASCE, V. 100, No. ST9, Proc. Gregor, J. G., "Monte Carlo Study of
Paper 10777, Sept. 1974, pp. 1755-1769. Strength of Concrete Columns," ACI
11. Ellingwood, B., and Ang, A. H.-S., "Risk Journal, V. 75, No. 8, Aug. 1978, pp.
Based Evaluation of Design Criteria," 348-358.
Journal of the Structural Division, 21. Siu, W. W. C., Parimi, S. R. and Lind,
ASCE, V. 100, No. ST9, Proc. Paper N. C., "Practical Approach to Code
10778, Sept. 1974, pp. 1771-1788. Calibration," Journal of the Structural
12. Ravindra, M. K., Lind, N. C., and Siu, Division, ASCE, V. 101, No. ST7, Proc.
W., "Illustrations of Reliability-Based Paper 11404, July 1975, pp. 1469-1480.

PCI JOURNAL/November-December 1982 83

22. Peir, J.-C. and Cornell, A. C., "Spatial ments for the degree of Doctor of Phi-
and Temporal Variability of Live losophy, Chicago, Illinois, 1980. See
Loads," Journal of the Structural Divi- also Report No. 80-1 by A. Siriaksorn
sion, ASCE, V. 99, No. ST5, Proc. Paper and A. E. Naaman, Department of Mate-
9747, May 1973, pp. 903-922. rials Engineering, June 1980, 126 pp.
23. Siu, W. W. C., and Lind, N. C., "Some 31. Naaman, A. E., and Siriaksorn, A.,
Aspects of Office Design Live. Loads," "Serviceability-Based Design of Par-
Journal of the Structural Division, tially Prestressed Beams," Part 1: PCI
ASCE, V. 99, No. ST11, Nov. 1973, pp. JOURNAL, V. 24, No. 2, March-April
2245-2258. 1979, pp. 64-89; Part 2: PCI JOURNAL,
24. McGuire, R. K., and Cornell, C. A., V. 24, No. 3, May-June 1979, pp. 40-60.
"Live Load Effects in Office Buildings," 32. ACI Committee 318, "Building Code
Journal of the Structural Division, Requirements for Reinforced Concrete
ASCE, V. 100, No. ST7, Proc. Paper (ACI 318-77)," American Concrete In-
10660, July 1974, pp. 1351-1366. stitute, Detroit, Michigan, 1977.
25. Culver, C. G., "Live Load Survey Re- 33. Ellingwood, B., "Reliability Bases of
sults for Office Buildings," Journal of Load and Resistance Factors for Rein-
the Structural Division, ASCE, V. 102, forced Concrete Design," National Bu-
No. ST12, Dec. 1976, pp. 2269-2284. reau of Standards Building Science Se-
26. Ellingwood, B., and Culver, C., "Analy- ries 110, Washington, D.C., 1978.
sis of Live Loads in Office Buildings,"
34. Mirza, S. A., and MacGregor, J. G.,
Journal of the Structural Division,
"Variations in Dimensions of Rein-
ASCE, V. 103, No. ST8, Aug. 1977, pp.
forced Concrete Members," Journal of
Structural Division, ASCE, V. 105, No.
27. Corotis, R. B., and Jaria, A. V., "Sto-
ST4, Proc. Paper 14495, April 1979, pp.
chastic Nature of Building Live Loads,"
Journal of the Structural Division,
ASCE, V. 105, No. ST3, Proc. Paper 35. Mirza, S. A., and MacGregor, J. G.,
14441, March 1979, pp. 493-510. "Variability of Mechanical Properties of
Reinforcing Bars," Journal of the
28. Dunham, J. W., "Design Live Load in
Buildings," Transactions, ASCE, V. 112, Structural Division, ASCE, V. 105, No.
Paper No. 2311, 1947, pp. 725-744. ST5, Proc. Paper 14590, May 1979, pp.
29. "Minimum Design Loads in Buildings
and Other Structures," American Na- 36. Pauw, A., "Static Modulus of Elasticity
tional Standards Institute, A58.1, 1972. of Concrete as Affected by Density,"
30. Siriaksorn, A., "Serviceability and Re- ACI Journal, V. 32, No. 6, Dec. 1960, pp.
liability Analysis of Partially Prestressed 679-687.
Concrete Beams," Thesis presented to 37. Price, W. H., "Factors Influencing Con-
the University of Illinois at Chicago Cir- crete Strength," ACI Journal, V. 22,
cle in partial fulfillment of the require- No. 6, Feb. 1951, pp. 417-432.


= distance from draping point of = specified compressive strength

steel profile to support of concrete
Ap3 area of prestressing steel in f(x) = probability distribution

tension zone function of x
A. = area of non-prestressed tension f allowable stress range under
reinforcement repeated service load in
b = width of upper flange of a concrete
flanged member f m dn = minimum stress in concrete
b.. = web width of flanged member feu = ultimate strength of
bt = lower flange width of a flanged
member f prestressing steel
py = specified yield strength of
C." = coefficient associated with prestressing steel
modulus of elasticity of fr = modulus of rupture of concrete
concrete (see Table 2) f r = stress range under repeated
Cfci = coefficient associated with service load in non-prestressed

compressive strength of tension steel
concrete at transfer (Table 2) = effective stress in prestressing
Crr = coefficient associated with steel, after losses
modulus of rupture of concrete f8i = initial stress in the prestressing
(see Table 2) steel immediately after transfer
Cfoe = coefficient associated with ftr = stress range under repeated
effective prestress in steel service load in prestressing
(see Table 2) steel
Cf. = coefficient associated with L = specified yield strength of non-
initial prestress prestressed steel
Coy = coefficient of variation h = overall thickness or depth of
dr = distance from extreme member
compression fiber to centroid hrl = thickness of upper flange of
of prestressing steel flanged member
20 = distance from extreme h f2 = thickness of lower flange of
compression fiber to centroid flanged member
of non-prestressed tension l = clear span length of member
reinforcement L = live load
= depth of area of concrete tensile m = margin of safety
zone associated with crack n = subscript for nominal value
width prediction equation PPR = partial prestressing ratio
D = dead load R = resistance
eo = eccentricity of prestressing S = total load
force with respect to centroid W,nax = maximum crack width
of section at midspan (3 = reliability index
el = eccentricity of prestressing Ye = unit weight of concrete
force with respect to centroid S = inherent coefficient of
of section at supports variation
E,, = modulus of elasticity of & o4d = additional long-term deflection
concrete A,, = immediate live load deflection
Ep3 = modulus of elasticity of e = prediction error
prestressing steel = mean value
E, = modulus of elasticity of non- (1 = total coefficient of variation
prestressed tension steel o- = standard deviation

PCI JOURNAL/November-December 1982 85