Resistance Factors for Reinforced Concrete

Qi Zhang

Written in Oct. 2002 at Ryerson University

Supervised by Dr. Reza Kianoush
Resistance Factor for Reinforced Concrete Qi Zhang

Introduction

The report reviews the concept of resistance factors in the limit state design. A brief

comparison of resistance factors among Canada, United States and China is presented.

Then the development of limit state design is introduced from an international academic

view. Finally, a procedure is illustrated to deduce resistance factors in the 1995 National

Building Code (NBC) of Canada.

The report provides the relative knowledge preparation for the recent project, Evaluation

and Retrofitting of Existing Reinforced Concrete Frame Structures Using Fiber

Reinforced Polymers, which has been being carried out at the Department of Civil

Engineering in Ryerson University. The objective of this project is to develop a viable

rehabilitation system using fiber reinforced polymer (FRP) composite materials to

improve the behavior of frame structures. It is proposed to develop a simple design

procedure for the practicing professionals and code type applications. Thus, how to

deduce and establish the resistance factors of FRP composite materials is the key point

of this project. Reviewing the derivation of the resistance factors of concrete and

reinforcement would benefit the future research on FRP composite materials.

-2-
Resistance Factor for Reinforced Concrete Qi Zhang

Limit State Design

The basic mission of structural design is to accurately deal with the contradiction

between the structural safety and economy, to choose a reasonable balance between

them, and then to meet the planned structural requirement during the prescriptive period

with the minimum price. In order to obtain the goals, for couples of decades the structural

design has adopted several methods, such as allowable stress design, working stress

design, and limit state design. Accepted as a probabilistic approach to structural safety,

the limit state design (LSD) has been used by structural engineers in Canada since the

mid 1970’s. It provides more consistent safety for various load combinations and various

combinations of materials than the past popular method, such as working stress design.

There are two basic functional requirements for all building structures: serviceability

during the useful life of the building and safety from collapse during the construction and

useful life of the building. LSD defines the various types of collapse and unserviceability

that are to be avoided. Those concerning safety are called the ultimate limit states,

others concerning unserviceability are called the serviceability limit states.

Compared with the past structural design methods, LSD uses partial safety factors

instead of the traditional single safety factors. The partial safety factors include the load

factors, , the load combination factor,  , the importance factor, , and the resistance

-3-
Resistance Factor for Reinforced Concrete Qi Zhang

factor,  , which will be reviewed primarily latter. Moreover, safety and serviceability are

also controlled by defining specified loads and material properties statistically, in terms of

the probability level (e.g., 5% maximum probability of underrun for material properties) or

the return period (10 to 100 years for snow, wind, and earthquake loads).

The LSD criteria can be expressed as follows:

[1] Factored resistance  effect of factored loads

 R effect of 




[ D D+ ( LL+ QQ+ TT)]

Where D, L, Q, and T refer to dead, live, wind (or earthquake) loads and imposed


deformation (temperature, etc.) respectively, and R is the factored resistance. Here

factor resistance,  R, is somehow different with the CSA A23.3 Standard.

Different Resistance Factors in Building Code of Canada, US and

China

The establishment of resistance factors shows the development of reliability analysis

through probability study. Resistance factors are derived for a number of factors causing

variability in strength:

-4-
Resistance Factor for Reinforced Concrete Qi Zhang

1. Variability in member strength due to variability of material properties in the structure.

2. Variability in member strength due to variability of dimensions.

3. Variability in member strength due to simplifying assumptions in the resistance

equations, such as the use of a rectangular stress block in concrete design.

4. Increased risk to building occupants if failure occurs without warning and the

post-failure strength is less than the original strength.

The resistance factors in NBC of Canada have been developed progressively from being

applied to the structural action to being applied to material strength directly. In the 1994

edition of the CSA A23.3 Standard, the resistance factor is applied to the material

strength,  c f'c or s fy , rather than to the nominal resistance,  Mn , Vn etc, which

are indicated in the latest American Concrete Institute (ACI) Building Code. However, the

application of resistance factors in Canada is similar to the application in China. Three

different resistance factors are summarized in Table1, Table2 and Table3.

Table1. Resistance factors in the CSA Standard A23.3

Kind of Material
Concrete, regular, c  1
Resistance factor
0.60

Concrete, precast elements manufactured and erected,
Reinforcing bars, s
2
  c 0.65
0.85
Prestressing tendons, p 0.90

Note:
Structural steel, s  0.90

1. The factored concrete strengths used in checking ultimate limit states shall be taken as

c* f'c and
c f'c1/2, where
f'c is the compressive strength of concrete.
2. The factored force in reinforcing bars, tendons, and structural shapes shall be taken as the product of the

-5-
Resistance Factor for Reinforced Concrete Qi Zhang

resistance factor,
sfy, where fy is the yield strength of reinforcement

Table2. Strength reduction factors in the ACI Code

Strength reduction
Kind of Strength

Flexure, without axial load

factor
0.90

Axial load, and axial load with flexure
Axial tension, and axial tension with flexure 0.90
Axial compression, and axial compression with flexure
Members with spiral reinforcement 0.75
Other members 0.70
Shear and torsion 0.85
Bearing on concrete 0.70
Note: The factored resistance used in checking ultimate limit states shall be taken as

Mn , Vn, etc, where Mn is
the nominal moment and Vn is the nominal shear.

Table3. Material subentry factors in the GBJ10-89 1

Material subentry
Kind of material

Concrete, compressed
factor 1/
0.74

Concrete, tensioned 0.74
Hot-roll steel,
Hot-roll steel,
,   
, ; Cold drawing steel, 0.91
0.87
Cold drawing steel,
Others 2
, ,  0.83

Note:
1. In building code of China, the strength design value is equal to the strength standard value divided by . The

strength standard value is a characteristic value based on probability reliability.
2. The subentry factors of other materials are detailed in GBJ10-89.

These resistance factors are computed and deduced through analysis of the reliable

target and engineering experience. It is difficult to say which one is better than others,

because every country has its own concern about materials, design familiarity,

experiment methods, statistical data, etc. The resistance factor of ACI slightly increases

simplicity in application; the others can account for the effect of varying the portion of the

-6-
Resistance Factor for Reinforced Concrete Qi Zhang

load assigned to the steel and concrete. But the research and application for new

materials would be able to take advantage of the independent resistance factors in NBC

of Canada and China. A simple design procedure for the new materials can be

developed by establish the new resistance factors, which are used conveniently for

practicing professionals and code type applications

Derivation of Resistance Factors in NBC of Canada

The ongoing project concerning the FRP composite materials has been carried out by

the Ryerson University for a few years. It is important to deduce and establish the

resistance factors of FRP materials through probability study based on experiments to

support the analytical data. Therefore the reviewing the resistance factors of traditional

material or structural strength is necessary and helpful to establish the new one.

In the 1994 edition of the CSA A23.3 Standard, the resistance factor is applied to the

material strength,  
c f'c or s fy. The resistance factor for steel,  s, accounts for

variability of the strength of reinforcement and to a lesser extent the variability in d, etc,

resulting from placement tolerances for reinforcement. The variability of the strength of

concrete and dimensions of concrete are reflected in the resistance factor for concrete,

 c. The brittle nature of failures initiated by failure of the concrete and the higher

-7-
Resistance Factor for Reinforced Concrete Qi Zhang

variability of design equations for failure initiated by failure of the concrete ( the shear Vc,

for example) were accounted for by using a higher safety index when evaluating the

value of  
c than s.

Referring to the papers written by J. G. MacGregor, the resistance factors,  c and s

are established through following procedure. The derivation of the resistance factors is a

huge job considered in many aspects: material variability, strength format, loading

combination, etc. Then taking flexure analysis of reinforced concrete as the example

introduces how to establish the resistance factors of reinforcement and concrete.

Consider a large family of similar floor beams each subjected to the same type of

occupancy, each for a 50 years lifetime. Each of the beams has been designed to

support an unfactored or service load moment, Q. Since the dead load and the live load

are both random, each of the family of beams will experience a different lifetime

maximum moment. Each of the beams concerned has a different moment capacity, R,

due to random variations in the concrete and steel strengths, effective depth, etc. As a

result, the family of Q and R have the distribution plotted in Fig.1. The 45-degree line in

Fig. 1 represents the case where the load effect, Q, equals the strength, R.

Combinations of Q and R that fall above this line, results in failure (Q>R).

-8-
Resistance Factor for Reinforced Concrete Qi Zhang

If we know the means and standard deviations of R and Q (Ro,

R,

Q0 and Q) we can

define a new function:

[2] Y= R – Q>0 or Y=R/Q>1 safety margin

with mean

[3] Y0 = R0 / Q0

and standard deviation

2 2 1/2
[4]

Y =( R + Q )

This is plotted in Fig.2. The shaded portion of this figure represents cases where R-Q<0,

in other words, where failure occurs. The probability of failure is the probability that

R-Q<0, indicated by the shaded portion of the curve. If the type of distribution is known,

the probability of failure can be computed from the number of standard deviation by

which Y0 exceeds zero. This is shown in Fig.2 as 

Y, where is referred to as the

safety index:

-9-
Resistance Factor for Reinforced Concrete Qi Zhang

[5] 

= Y0 / Y

If  is increased by increasing the value of Y0, the shaded area in Fig.2 is reduced and

the probability of failure is reduced. Thus  is a measure of the reliability of a structural

member,  c and s are determined by the reliability target  . Literature suggests that

a desirable range of  values is from 3.0 to 3.5. A target reliability index,  , of 3.00 to

3.25 for ductile failure and 3.25 to 3.5- for brittle failure was selected. These values 
apply to loads that are defined at a probability of exceedance of 0.033 and for normal

structures with a design life of 50 years.

Using the equation from MacGregor (1976),  is shown in a practical format:

[6] γ R = R0 / R

[7] γ U = U0 / U

Where R and U are design strength and design loading respectively

[7] Rγ R [e − βαVr ] ≥ Uγ U [e βαVu ]

Rearranging this gives,

[8]  =ln(R γ R /U γ U ) / ( VR+

VQ)

Where
is a “separation function” having value 0.75 (Lind 1971); VR and VQ are the

coefficients of variation of R and Q.

From “Concrete Design Handbook”, we can get the factored resistance (for example,

- 10 -
Resistance Factor for Reinforced Concrete Qi Zhang

detailed in Appendix). Given the fixed crossing section, the placement of reinforcement

and yield strength of reinforcement and concrete, the factored resistance is determined

by  
c and s.

Preliminary analysis indicated that the reliability in flexure depends strongly on  s and

not very much on  c. Similarly, the reliability of columns that develop compression

failures is more sensitive to c than   s. Therefore, flexural members are used to deduce

 s while columns are used to deduce  c.

Researchers (Ellingwood et al. 1980) have found resistance factors to be relatively

insensitive to the variable loads. Therefore, we do not have to consider all of the many

combinations of variable loads in the deduction of resistance factors. In the report, the

resistance factors are deduced for the dead load (D) and live load (L) combinations.

Other researchers (Kariyawasam et al. 1997) have indicated beams R-0.71 (rectangular

section with  
=0.71 bal) and T-0.14 (T beam with  
=0.14 bal) are the most critical

(have the lowest  values in the different L/D ratios). This would be true even if the

load and resistance factors were changed. Therefore, the two critical beams are chosen

to investigate.

Fig. 3 and 4, for beams R-0.71 and T-0.14 respectively, give the ranges of  s possible

- 11 -
Resistance Factor for Reinforced Concrete Qi Zhang

with different c and L/D values to give  values within the target range of 3.00 to 3.25.

From Table 4,  s=0.85 is chosen, as this would provide the necessary reliability in all

cases considered. This is the value currently specified in CSA A23.3-94.

- 12 -
Resistance Factor for Reinforced Concrete Qi Zhang

In the representative columns studied, the analysis for  c is given in Table 5. The

analysis suggests that when  

s=0.85, c may have a value of 0.675. In Kariyawasam’

study, c =0.65 is chosen , as this would provide the necessary reliability in all cases

considered. This is higher than the c =0.60 given in CSA A23.3-94 for cast-in-place

concrete.

In establishing resistance factors for shear, only beams that have a longitudinal

reinforcement ratio greater than 1.0% are considered. Members without stirrups that are

not governed by the limitation that Vf shall be less than or equal to Vc/2 (clause 11.2.8.1

of A23.3-94) are also not considered in Kariyawasam’ study. The most critical beams are

investigated with different combinations of factors. The results are given in Table 6. It is

seen that together with  c =0.65 and  s =0.85, the constant in the Vc equation should

be equal to 0.17 to give  values within the target range.

- 13 -
Resistance Factor for Reinforced Concrete Qi Zhang

Conclusion

Limit State Design uses partial safety factors including the load factors, , the load

combination factor,  , the importance factor,  , and the resistance factor,  . The

safety and serviceability are also controlled by defining specified loads and material

properties statistically, in terms of the probability level or the return period. The

resistance factor of ACI slightly increases simplicity in application; the others can

account for the effect of varying the portion of the load assigned to the steel and concrete.

 s=0.85 and c =0.65, which are higher than the value given in CSA A23.3-94 for

cast-in-place concrete, could provide the necessary reliability in all cases of reinforced

concrete structures.

- 14 -
Resistance Factor for Reinforced Concrete Qi Zhang

References:

Arthur H. Nilson, George Winter, “Design of concrete structures” – 11th Edition, 1991,
McGraw-Hill, Inc., NY.
Canadian Portland Cement Association, “Concrete Design Handbook,” (1998): 1.1.4,
Canada
CSA, 1994. “Design of concrete structure --- structures (design).” CSA Sstandard A23.3- 94,
Canadian Standards Association, Rexdale, Ont.
Dennis E. Becker, “ Eighteenth Canadian Geotechnical Colloquium: Limit States Design for
Foundations. Part 1. An overview of the foundation design process,” Can. Geotech. J. 33: 956-983
(1996).
David E.Allen, “Limit states design, a probabilistic study,” Canadian Journal of Civil Engineering,
Vol. 2, pp 36-49. 1974
Ellingwood, B., Galambos, T.V., MacGregor, J.G., and Cornell, C.A.,”Development of
a probability based load criterion for American National Standard A58.” Special
Publication 577, National Bureau of Standard, Gaithersburg, MD. 1980
James G. Macgregor, “Safety and Limit States Design for Reinforced Concrete,” Canadian
Journal of Civil Engineering, Vol. 3, No.4, December 1976, pp. 484-513.
Lind, N. C. 1971. “Consistent partial safety factors. Proc. ASCE, 97(ST6), pp. 1651-1670
Russell S. Fling, “Practical design of reinforced concrete.” 1987, John Wiley & Sons, Inc. NY.
Samuel E. French, “Design of reinforced concrete.” 1988, Prentice Hall, Englewood Cliffs, NJ
Sher Ali Mirza and James G. MacGregor, “Probabilistic study of the strength of reinforced
concrete members,” Canadian Journal of Civil Engineering, Vol. 9, No. 3, September 1982, pp
431-457.
Shilin Shu, “Structures of reinforced concrete.” 1996, Zhejiang University, Hangzhou, China
S. N. Kariyawasam, D. M. Rogowsky, and J. G. MacGregor, “Resistance factors and
companion-action load factors for reinforced concrete building design in Canada,” Can. J. Civ. Eng.24
(1997): 593 – 602

- 15 -
Limit State Design
Calculate the Moment of resistance, Mr, with CSA A23.3

Given:
f'c’ = 30 Mpa
fy = 400 Mpa


f = 0.85
c = 0.6

1.Calculate 1:
1 
= 0.85-0.0015 f'c’ = 0.85-0.0015*30 = 0.805

2.Calculate the Area of reinforcement As:

As = 3* *(25/2)2 = 1473 mm.2

3.Calculate the depth of rectangular concrete stress block, a:
a= f*As*fy /( 1* c* f'c’*b)



= 0.85*1473*400 / (0.805*0.6*30*300)
= 115 mm

4.Calculate resisting moment Mr:

Mr = As* f*fy *(d-a/2)


= 1473 * 0.85 * 400 * ( 600- 115/2) * 10-6
= 272 KN*m

- 16 -