LIFE CYCLE COST ANALYSIS OF PRESTRESSED CONCRETE

BRIDGES USING CFRP REINFORCEMENT

Nabil GRACE
Dean and Distinguished Professor, College of Engineering
Lawrence Technological University
Southfield, Michigan, USA
nabil@ltu.edu*

Elin JENSEN
Associate Dean of Graduate Studies and Research
Lawrence Technological University
Southfield, Michigan, USA
ejensen@ltu.edu

Christopher EAMON
Associate Professor, Civil and Environmental Engineering
Wayne State University
Detroit, Michigan, USA
eamon@eng.wayn.edu

Tsuyoshi ENOMOTO
Manager
Tokyo Rope Manufacturing Co. Ltd.
Tokyo, Japan
enomoto.tsuyoshi@tokyorope.jp
Xiuwei SHI
Former Graduate Student
Lawrence Technological University
Southfield, Michigan, USA
xshi@ltu.edu

Abstract
A significant cause of prestressed-concrete bridge deterioration is corrosion of the steel
reinforcement, which has led to the consideration of non-corrosive FRP reinforcement
alternatives. To assess the economic performance of FRP, a life cycle cost analysis of
prestressed-concrete highway bridges using carbon fiber reinforced polymer (CFRP)
reinforcement bars and strands was conducted. Side-by-side box beam and AASHTO beam
bridge superstructures are considered over several span lengths and traffic volumes. The
results show that despite the higher initial construction cost of CFRP reinforced bridges, they
can be cost effective when compared to traditional steel reinforced bridges over the lifespan
of the bridge. The most cost efficient design was found to be a medium-span CFRP-
reinforced girder bridge located in a high traffic area. A probabilistic analysis revealed that
there is a probability of more than 95 % that the CFRP reinforced bridge will become the
least expensive option after 20 to 40 years of service depending on traffic volume and bridge
geometry.

Keywords: bridge, prestressed concrete, CFRP, life-cycle cost analysis

Page 1 of 8
1. Introduction
Approximately, 30 % of the bridge inventory in the United States requires immediate repair
due to the reinforcement corrosion issues with an estimated repair bill exceeding $8 billion
[1]. This has long been recognized as a costly maintenance problem for concrete bridge
components.
In light of this problem, the interest in using non-corrosive alternatives such as fiber
reinforced polymer (FRP) composites to replace traditional steel reinforcement has grown
rapidly. Presently, various demonstration FRP-reinforced bridges have been constructed all
over the world. Examples in the U.S.A. include the construction of bridges to carry: Pierce
Street in Lima, OH (1999); Salem Avenue in Dayton, OH (1999); Rollins Road in
Rollinsford, NH (2000); Sierrita de la Cruz Creek in Potter County, TX (2000); 53rd Avenue
in Bettendorf, IA (2001); Bridge Street in Southfield, MI (2001); Highway 151 in Waupun,
WI (2005); Route Y in Boone County, MO (2007).
Because the initial construction cost of an FRP-reinforced bridge is often significantly higher
than that of a steel-reinforced bridge (up to 80 % higher but typically within the range of 25
to 50 %), any potential economic advantages associated with FRP will not be realized unless
the cost over an extended period of time is considered. Thus, a life-cycle cost analysis
(LCCA) should be performed to determine if and when an eventual cost-savings occurs.
Moreover, as construction costs of many maintenance events are rarely known with certainty,
an important component of LCCA is consideration of these uncertainties. Inclusion of
uncertainties, by representing critical LCCA costs as random variables, allows for results to
be expressed in a probabilistic sense; for example, the probability that one reinforcement
alternative is less costly than another as a function of time. The goal of this study is to
determine when prestressed concrete bridges utilizing CFRP reinforcing bars and strands can
represent a cost effective design alternative to conventional steel reinforced prestressed
concrete bridges.

2. Structures considered
The current study considered a total of 26 cases for bridge superstructures with different
configurations, spans, and traffic volumes. The cases addressed side-by-side prestressed
concrete box beam and AASHTO beam bridge superstructures. The bridge superstructures
were assumed to accommodate two lanes with shoulders and span over a 4, 6, or 8-lane
highway (which corresponded to bridge spans of 14, 18, or 37 m, respectively). The bridge
superstructure was composed of two simply supported spans with a total deck width of 14 m.
Figure 1 shows the cross section of the selected bridge superstructures.
The side-by-side box beam bridge superstructures were composed of eleven side-by-side
precast prestressed box beams and were provided with a 150-mm thick deck slab reinforced
with a single reinforcement mesh. Box beams with sizes of 910 × 510, 910 ×710, and 1200 ×
1200 mm2 were selected to model bridges with spans of 14, 18, and 37 m, respectively.
The AASHTO beam bridge superstructures were composed of spaced precast prestressed
AASHTO beams and were provided with a 230-mm thick deck slab reinforced with a two
reinforcement meshes. AASHTO beams Type I, II, and IV with a beam spacing of 1.7, 2.4,
and 2.9, m were selected to model bridges with spans of 14, 18, and 37 m, respectively.
Two traffic volumes were considered: a low volume with an initial annual average daily
traffic (AADT) of 1,000 and a high volume with initial AADT of 10,000. The annual growth
rate was taken as 2 % and limited to a maximum AADT of 26,000, which was calculated
from the free flow lane capacity of the roadways. Below-bridge traffic volumes ranged from

Page 2 of 8
10,000 to 20,000 for low volume, 30,000 to 100,000 for medium volume, and 100,000 to
140,000 for high volume, depending on bridge span, with the annual growth rate taken as 1
%. These combinations of bridge configurations, span, and traffic volumes resulted in the 26
cases for deterministic LCCA. For each case, three reinforcing alternatives were considered:
(a) black steel reinforcement without epoxy-coating with cathodic protection, (b) epoxy-
coated steel reinforcement, and (c) CFRP reinforcement. The CFRP bridge is designed based
on ACI 440 [2[3] design guidelines. The CFRP is designed such that it has the same flexural
and shear design capacities as the steel reinforced bridges.

Figure 1. Bridge sections

3. Activity timing
The period of the analysis must extend to cover major rehabilitation actions for each
reinforcement alternative. To satisfy this requirement, the LCCA was conducted for a period
of 100 years. For consistent LCC comparison among cases, it is important that the
maintenance actions are scheduled such that the expected bridge condition, at any year, is the
same for all three reinforcement alternatives. In order to maintain the same performance
level, different operation, maintenance and repair (OM&R) strategies may be defined for
each type of bridge reinforcement alternative considered.
For the black steel (with cathodic protection) and epoxy-coated steel bridges, the OM&R
strategies in this study were based on MDOT practices for the time intervals for inspection,
deck and beam-related maintenance work, and superstructure demolition and replacement.
The maintenance activities over the 100 year LCCA period for steel reinforced bridges
included superstructure replacement (year 65); deck and beam replacement (year 40); deck
overlay and beam end repair (years 20, 55, 85); deck patch (years 8, 16, 28, 36, 48, 73, 81,
and 93); and cathodic protection for black steel (years 25, 50, and 90). As MDOT has no
CFRP reinforced bridges in their inventory, the OM&R strategies of existing CFRP bridges
in Japan [4, [5] and Canada [6] under similar climatic conditions as Michigan were consulted
to establish an expected maintenance schedule for timing activities. Based on these
schedules, the CFRP bridge is expected to require only one deck shallow overlay and one
deck replacement during its service life, which are projected to occur at years 50 and 80,
respectively. This greatly-reduced maintenance activity is expected, as the purpose of using
CFRP is the elimination of corrosion-induced concrete component deterioration. Based on

Page 3 of 8
MDOT practices, routine bridge inspections occur every other year, and a detailed inspection
occurs every five years for the steel reinforced bridges and every 10 years for the CFRP
bridge (but not during years of superstructure replacement).

4. Costs
The agency costs include material, personnel, and equipment costs associated with OM&R,
demolition, and replacement. The total initial construction cost of the epoxy-coated steel
reinforced bridge was estimated based on the general MDOT cost estimate scheme ($110 per
square foot bridge deck area). Costs of the two alternative bridges (black steel and CFRP)
were based on the cost of the epoxy-coated steel reinforced bridge, accounting for the
construction and material cost differences. Material costs such as concrete, steel
reinforcement, and CFRP were based on current (2009) estimates from MDOT and CFRP
fabricators. The cost of OM&R included routine inspection, detailed inspection, cathodic
protection, deck patch, deck shallow overlay, deck deep overlay, deck replacement, beam end
repair, beam replacement, superstructure demolition, and superstructure replacement. These
costs were based on MDOT assessment as well as other sources for verification [7].
During construction and maintenance, traffic delays as well as increased accident rates
occur. The resulting delay costs caused by construction work include the value of time lost
due to increased travel time as well as the cost of additional vehicle operation. Therefore,
user cost is the sum of travel time costs, vehicle operating costs, and crash costs. Equations
(1) to (3) were used to calculate these costs [8].
 L L 
Travel time cos ts =  −  × AADT × N × w (1)
 Sa Sn 
 L L 
Vehicle operating cos ts =  −  × AADT × N × r (2)
 Sa Sn 
Cash cos ts = L × AADT × N × ( Aa − An ) × C a (3)
Where, L is the length of affected roadway over which cars drive (0.8 to 3.2 km, depending
on activity); Sa is the traffic speed during road work (48 km/hr on bridge and 72 km/hr below
bridge); Sn is the normal traffic speed (72 km/hr on bridge and 113 km/hr below bridge); N
is the number of days of road work (4 hours to 5 months, depending on activity); w is the
hourly time value of drivers ($13.61); r is the hourly vehicle operating cost ($11.22); Ca is
the cost per accident ($99,560); Aa (2.58 %) and An (1.56 %) are the accident rate during
construction and normal accident rate per million vehicle-miles, respectively. The annual
average daily traffic (AADT) value for each year of the analysis period is computed from the
initial AADT and the traffic growth rate (given earlier), as limited by a maximum AADT of
120,000 to 250,000 depending on initial traffic volume. Other parameter values are taken
from the available literature [8, [9, [10].
The total life cycle cost is the sum of all yearly partial agency and user costs. Because dollars
spent at different times have different present values (PV), future costs at time t, Ct, are
converted to consistent present dollar values by adjusting future costs using the real discount
rate r, and then summing the results over T years:
t =T
Ct
LCC = ∑ (4)
t =0 (1 + r ) t

Page 4 of 8
The real discount rate reflects the opportunity value of time and is used to calculate the
effects of both inflation and discounting, and is taken as 3 % [1]. For this study, the initial
construction cost occurs in year 0, while the first year after bridge construction is defined as
year 1. The costs associated with any subsequent activity are presented in terms of present
value considering the real discount rate.

5. Deterministic results
A typical result is given in Figure 2, which is for the medium span (18 m) box beam bridge
with a high traffic volume both on and below the bridge. For this case, two lanes pass under
each of the two 18-m spans. As expected, the initial construction cost of the CFRP
reinforced bridge is higher than the traditional steel reinforced bridges. However, in year 20,
when the first significant deck repair takes place on the steel bridges, their cumulative cost
exceeds the cost of the CFRP bridge. The final life-cycle costs are $5.98 million for the
bridge with black steel reinforcement, $5.63 million for the bridge with epoxy-coated steel
reinforcement, and $2.23 million for the bridge with CFRP reinforcement. Similar results
were achieved for the AASHTO beam bridges with the same span and traffic level, with final
costs at 100 years slightly lower than the box beam bridges ($5.39, $5.04, and $1.77 million
for the black steel, epoxy-coated, and CFRP AASHTO beam alternatives, respectively). The
most significant contributor to LCC is user cost, which contributes from 50 % to 78 % of the
total project cost for the different alternatives. It can be noted that the LCC of the steel
reinforced bridges are about three times the LCC of the CFRP reinforced bridge.
Furthermore, the agency life-cycle cost is reduced by 12 % if CFRP reinforcement is selected
over epoxy-coated reinforcement, and by 23 % if CFRP is selected over black steel. The
economic benefit is achieved from the reduced maintenance requirements associated with
CFRP (i.e. there is no corrosion-associated deterioration).
Life-cycle cost (millions of dollars)

$7.0
Black Steel Bridge
$6.0
Epoxy-Coated Steel Bridge
$5.0 CFRP Bridge
$4.0
$3.0
$2.0
$1.0
$0.0
0 10 20 30 40 50 60 70 80 90 100
Year

Figure 2. Medium span bridge deterministic LCC

The variables that have the highest influence on the life cycle cost were determined with a
sensitivity analysis. The four most significant variables for all cases were: normal driving
speed (Sn) below the bridge, real discount rate (r), driving speed reduction during roadwork
(Sn-Sa) below the bridge, and traffic volume (AADT).
In all cases, as traffic volume increases, the CFRP bridge becomes more cost-effective. This
is because the maintenance-related user cost differences between the CFRP and steel
reinforced bridges are magnified at higher traffic volumes. Among all bridge cases, the

Page 5 of 8
medium-span CFRP bridge with high traffic levels below and on the bridge was found to be
most cost effective in the long term. Moreover, the CFRP reinforcing was found to be more
cost effective when applied to the AASHTO beam bridge than to the box beam bridge. In
summary, the medium span AASHTO beam bridge with high traffic volume below and
above was found to be the most cost-effective case for CFRP reinforcement.

6. Probabilistic analysis
To account for cost uncertainties, a probabilistic analysis was performed to evaluate the
probability that CFRP is the most cost-effective solution throughout the analysis period. All
major cost items were taken as random variables (RVs), except for the agency cost associated
with inspection, which was taken as deterministic. This resulted in ten agency and nine user
cost RVs. RV means were taken as the deterministic cost values, while coefficients of
variation (COV) were taken from the available literature and varied from 0.20 to 0.60.
Insufficient data were available to obtain exact distributions, therefore, RVs were assumed
normal. Agency cost statistics were divided into two categories: construction costs and
repair/maintenance costs. Construction cost COVs were based on an analysis of bridge and
building project cost variances [11[12], where repair and maintenance cost COVs were taken
from DOT bridge repair cost records [13]. Travel time cost COV was based on an analysis of
USDOT-compiled data [14], while vehicle operating cost COV was computed from average
operating costs of different types of vehicles [15]. COV of vehicle crash costs was taken
from FHWA-compiled data of crash geometries pertinent to bridge work sites [16].

1.0
Probability that CFRP Bridge Costs Less

0.9
0.8
0.7 AASHTO Black steel
AASHTO Epoxy-coated
0.6
Box Black steel
0.5 Box Epoxy-coated
0.4
0.3
0.2
0.1
0.0
0 10 20 30 40 50 60 70 80 90 100
Year

Figure 3. Medium span bridge probabilistic LCC

Monte Carlo Simulation (MCS) was used to simulate cumulative bridge costs each year. For
each of the 26 analysis cases, 100,000 simulations per bridge per year were conducted. Figure
3 presents a typical result (medium span, low traffic volume below and high traffic on the
bridge). The figure gives the probability that the cumulative yearly discounted cost of the
black steel and epoxy-coated reinforcement bridges will exceed the cost of the CFRP
reinforced bridge. As time progresses, the probability that CFRP will become the cheapest
option increases. Until year 20, there is a low probability that this will occur, given the high
initial cost of CFRP relative to the other options. However, at year 20, after the first deck
shallow overlay for the steel bridges, the trend reverses where now CFRP has a 0.81
(compared to epoxy-coated) to 0.93 (compared to black steel) probability of being the

Page 6 of 8
cheapest option. At year 40, there is less than a 1 in 1000 probability that CFRP will be a
more expensive option for this case.

7. Conclusions
The LCCA suggests that bridges constructed with CFRP reinforcement may become more
cost effective than steel reinforced concrete bridges by the first major maintenance event (at
about year 20 based on the MDOT maintenance schedule used in this study). Specific results
are:
• Traffic volume on and below the bridge significantly affects the life cycle cost. The cost
effectiveness of the CFRP reinforced bridge is greater when located in an area with a high
traffic volume.
• Among bridges with different span lengths, the CFRP reinforced medium-span bridge is
generally the most cost efficient.
• It is more cost efficient when CFRP reinforcement is applied to AASHTO beam bridges
than to side-by-side box beam bridges regardless of span length and traffic volume.
• The four variables that have the highest influence on LCC in this study are: traffic speed
on the roadway below, real discount rate, speed reduction during construction, and traffic
volume. This was found for all bridge alternatives. The significance of additional
variables depends on the bridge case considered.
• The probabilistic analysis confirmed deterministic results. There is a probability of more
than 0.51 that CFRP will be the most cost-effective option by year 20 for all cases
considered. It was found that for the majority of cases considered, there is a probability
of more than 0.90 that CFRP will be the most cost-effective option by year 20.

8. References
[1] OFFICE OF ASSET MANAGEMENT, “Life-Cycle Cost Analysis Primer”, Report,
U.S. Department of Transportation, Federal Highway Administration (FHWA),
Washington D.C., U.S.A., August 2002, pp. 1-25.
[2] ACI COMMITTEE 440, “Prestressing Concrete Structures with FRP Tendons”, ACI
440.4 R-04, Farmington Hills, Mich., U.S.A., 2004, pp.1-36
[3] ACI Committee 440, “Guide for the Design and Construction of Structural Concrete
Reinforced with FRP Bars”, ACI 440.1 R-06, Farmington Hills, Mich.,, U.S.A., 2006,
pp. 1-44
[4] ADVANCED COMPOSITE CABLE (ACC) CLUB, “Report of Study Group on
Application of Life Cycle Cost”, 2002, (in Japanese)
[5] NISHIZAKI, I., TAKEDA, N., ISHIZUKA, Y., SHIMOMURA, T., “A Case Study of
Life Cycle Cost based on a Real FRP Bridge”, 3rd International Conference on FRP
Composites in Civil Engineering (CICE 2006), Miami, Florida, U.S.A., 2006, pp. 99-
102.
[6] FAM, A., RIZKALLA, S., TADROS, G, “Behavior of CFRP for prestressing and shear
reinforcements of concrete highway bridges.” ACI structural journal, Vol. 94, No. 1,
1997, pp. 77-86.
[7] HUANG,Y., ADAMS, T., PINCHEIRA, J., “Analysis of life-cycle maintenance
strategies for concrete bridge decks.” Journal of Bridge Engineering, Vol. 9, No. 3,
May/June 2004, pp. 250-258.
[8] EHLEN, M., “Life-Cycle Costs of Fiber-Reinforced-Polymer Bridge Decks”, Journal
of Materials in Civil Engineering, Vol. 11, No. 4, 1999, pp. 224-230.

Page 7 of 8
[9] EHLEN, M., MARSHALL, H., "The Economics of New-Technology Materials: A
Case Study of FRP Bridge Decking", NISTIR 5864, National Institute of Standards and
Technology (NIST), Gaithersburg, MD, 1996, pp. 1-80.
[10] ANALYSIS DIVISION, FEDERAL MOTOR CARRIER SAFETY
ADMINISTRATION, “2000 Work Zone Traffic Crash Facts”, U.S. Department of
Transportation, Washington, DC., U.S.A., April 2002.
[11] SAITO, M., KUMARES, C., ANDERSON, V., “Bridge Replacement Cost Analysis”
Transportation Research Record 1180, 1988, pp. 19-24.
[12] SKITMORE, M., NG, T., “Analytical and Approximate Variance of Total Project
Cost”, ASCE Journal of Construction Engineering and Management, Sept/Oct 2002,
pp. 456-460.
[13] SOBANJO, J., THOMPSON, P., “Development of Agency Maintenance, Repair, and
Rehabilitation (MR&R) Cost Data for Florida’s Bridge Management System”, final
report, Research canter, Florida Department of Transportation, July 2001, pp. 100-301.
[14] OFFICE OF THE SECRETARY OF TRANSPORTATION, “The Value of Saving
Travel Time”, Departmental Guidance for Conducting Economic Evaluations, U.S.
Department of Transportation, April 1997, pp. 1-33.
[15] U.S. DEPARTMENT OF TRANSPORTATION, “Annual Vehicle Distance Traveled
in Miles and Related Data - 2007 1/ By Highway Category and Vehicle Type”,
Highway Statistics 2007, Federal Highway Administration (FHWA), Washington, DC,
U.S.A., April 2011.
[16] COUNCIL, F., ZALOSHNJA, E., MILLER, T., PERSAUD, B., “Crash Cost Estimates
by Maximum Police-Reported Injury Severity Within Selected Crash Geometries”,
Final Report, No. FHWA-HRT-05-051, Office of Safety Research and Development,
Federal Highway Administration (FHWA), Washington D.C., U.S.A., October 2005,
pp. 1-69.

Page 8 of 8