BRIDGE ENGINEERING

3 General Design Considerations

Objective:

The justification stage of design can begin after the selection of possible alternative bridge types
that satisfy the function and aesthetic requirements of the bridge location has been completed. As
discussed in the opening pages of Chapter 2, justification requires that the engineer verify the
structural safety and stability of the proposed design. Justification involves calculations to
demonstrate to those who have a vested interest that all applicable specifications, design, and
construction requirements are satisfied. In this chapter, we discuss general design considerations
that range from the limit state philosophy of structural design to the calibration of the LRFD
specification to the practical matter of horizontal and vertical clearances. All of these elements
make up the design experience and must be understood.

3.1 Introduction

A general statement for assuring safety in engineering design is that the resistance of the
components supplied exceed the demands put on them by applied loads, that is,

When applying this simple principle, both sides of the inequality are evaluated for the same
conditions. For example, if the effect of applied loads is to produce compressive stress on a soil,
this should be compared to the bearing resistance of the soil, and not some other quantity. In other
words, the evaluation of the inequality must be done for a specific loading condition that links
together resistance and the effect of loads. Evaluating both sides at the same limit state for each
applicable failure mode provides this common link.
When a particular loading condition reaches its limit, failure is the assumed result, that
is, the loading condition becomes a failure mode. Such a condition is referred to as a limit state
that can be defined as:
A limit state is a condition beyond which a bridge system or bridge component ceases to
fulfill the function for which it is designed.

Examples of limit states for girder-type bridges include deflection, cracking, fatigue,
flexure, shear, torsion, buckling, settlement, bearing, and sliding. Well-defined limit states are
established so that a designer knows what is considered to be unacceptable.

An important goal of design is to prevent a limit state from being reached. However, it is
not the only goal. Other goals that must be considered and balanced in the overall design are
function, appearance, and economy. To design a bridge so that none of its components would ever
fail is not economical.

Therefore, it becomes necessary to determine what is an acceptable level of risk or probability

of failure. The determination of an acceptable margin of safety (how much greater the resistance
should be compared to the effect of loads) is not based on the opinion of one individual but on the
collective experience and judgment of a qualified group of engineers and officials.

To design a bridge like you need to take into account all the forces acting on it:

• The friction of the earth on every part

• The strength of the ground pushing up the supports
• The resistance of the ground to the pull of the cables
• The dead weight and all vehicle loads
• Then there is the drag and lift produced by wind and water
• The turbulence as fluids pass the towers
• Need to use appropriate materials and structural shapes in the cheapest way, yet
maintaining a certain degree of safety.
• To account for natural disasters, engineers design bridges with a factor of safety:
usually around 3 or 4.

Instructor: ENGR. JOSHUA CASTILLO JUNIO 1

 BRIDGE ENGINEERING

3.2 Development of Design Procedures ASD and LRFD Design Philosophies

3.2.1 Allowable Stress Design

The earliest numerically based design procedures were developed with a primary focus on
behavior of metallic structures. Safety in the design was obtained by specifying that the effect
of the load should produce stresses that were a fraction of the yield stress fy, say one-half.
This value will be equivalent to providing a safety factor F of 2; that is,

Since the specification set limits on the stresses, so this became known as Allowable
Stress Design.
For steel bridge design, the required net area of a tension member is selected by:

For compression members, the required area is given by:

For beams in bending, a required section modulus ‘S’ is determined as:

In regard to uncertainties in design, one other point concerning the ASD method needs to
be emphasized. Allowable stress design does not recognize that different loads have different
levels of uncertainty. Dead, live, and wind loads are all treated equally in ASD. The safety
factor is applied to the resistance side of the design inequality, and the load side is not
factored. In ASD, safety is determined by:

ASD is not well suited for design of modern structures due to the following
shortcomings:

1. The resistance concept is based on the elastic behaviour of homogeneous materials.

2. It does not give reasonable measure of strength which is more fundamental measure of resistance
than as allowable stress.
3. The safety factor is applied only to the resistance and loads are considered to be
deterministic (i.e., without variation).
4. Selection of a safety factor is subjective and it does not provide a measure of reliability
in terms of probability of failure.

3.2.2 Load and Resistance Factor Design

Limit States:
Service Limit State
Strength Limit State
Fatigue and Fracture Limit State
Extreme Event Limit State

To overcome the deficiencies of ASD, the LRFD method was developed which is based on

a. The strength of material

b. Considers variability not only in resistance but also in the effect of loads.
c. Provides a measure of safety related to probability of failure.

The basic design expression in the AASHTO (2004b) LRFD Bridge Specifications that must be
satisfied for all limit states, both global and local, is given as:

Instructor: ENGR. JOSHUA CASTILLO JUNIO 2

 BRIDGE ENGINEERING

Where Φ is the statistically based resistance factor applied to nominal resistance, Rn is the
nominal resistance, γi is the statistically based load factor and Q i is the effect of load
and ηi is the load modification factor.
This equation involves both load factors and resistance factors. The design method is
called Load and Resistance Factor Design (LRFD).

The load modifier ηi takes into account the ductility, redundancy and operational
importance of the bridge. It is given by for loads for which a maximum value of γ i is
appropriate by:

And for loads for which a minimum value of γi is appropriate by:

Where ηD is the ductility factor, ηR is the redundancy factor, and ηI is the operational
importance factor. The first two factors refer to the strength of the bridge and the third
refers to the consequence of a bridge being out of service. For all nonstrength limit states
ηD = ηR = 1.0.

3.2.3 DUCTILITY FACTOR, ηD

Ductility is important to the safety of the bridge. If ductility is present overloaded
portion of the structure can redistribute the load to other portions that have reserve
strength. This redistribution is dependent on the ability of the overloaded component and its
connections to develop inelastic deformations without failure. Brittle behavior is to be
avoided, because it implies a sudden loss of load carrying capacity when the elastic limit
is exceeded. The values to be used for the strength limit state, ductility factors are

ηD ≥ 1.05 for nonductile components and connections

ηD = 1.00 for conventional designs and details complying with the specifications
ηD ≥ 0.95 for components and connections for which additional ductility-enhancing
measures have been specified beyond those required by the specifications

For all other limit states:

ηD = 1.00

3.2.4 REDUNDANCY FACTOR

A statically indeterminate structure is redundant, that is, it has more restraints
than necessary to satisfy conditions of equilibrium.
For example, a three span continuous bridge girder would be classified as statically
indeterminate to second degree. Any combination of two supports or two moments or one support
and one moment could be lost without immediate collapse, because the loads could find
alternative paths to the ground.
Redundancy in a bridge system will increase its margin of safety and this is reflected
in the strength limit state redundancy factors given as
ηR ≥ 1.05 for nonredundant members
ηR = 1.00 for conventional levels of redundancy
ηR ≥ 0.95 for exceptional levels of redundancy
For all other limit states:
ηR = 1.00

3.2.5 OPERATIONAL IMPORTANCE FACTOR

Bridges can be considered of operational importance if they are on the shortest path
between residential areas and a hospital or a school or provide access for police, fire, and
rescue vehicles to homes, businesses, industrial plants, etc. It is difficult to find a
situation where a bridge would not be operationally important.
One example of a non-important bridge could be on a secondary road leading to a remote
recreation area that is not open year around.

Instructor: ENGR. JOSHUA CASTILLO JUNIO 3

 BRIDGE ENGINEERING

In the event of an earthquake, it is important that all lifelines, such as bridges

remain open. Therefore, following requirements apply to the extreme event limit state as well
as to the strength limit state.
ηI ≥ 1.05 for a bridge of operational importance
ηI = 1.00 for typical bridges
ηI ≥ 0.95 for relatively less important bridges
For all other limit states:
ηI = 1.00

3.2.6. ADVANTAGES OF LRFD

1. LRFD accounts for both variability in resistance and load
2. It achieves fairly uniform factor of safety for different limit states.
3. It provides a rationale and consistent method of design.

3.2.7 DISADVANTAGES OF LRFD

1. It requires a change in design philosophy (from previous AASHTO methods).
2. It requires an understanding of the basic concepts of probability and statistics.
3. It requires availability of sufficient statistical data and probabilistic design
algorithms to make adjustments in the resistance factors to meet individual situation.

3.3 Relevant Portions of AASHTO

3.3.1 General

LOAD DESIGNATION [A3.3.2]

Permanent and transient loads and forces that must be considered in a design are
designated as follows:

Permanent Loads
DD Downdrag
DC Dead load of structural components and non-structural attachments
DW Dead load of wearing surfaces and utilities
EH Horizontal earth pressure load
EL Accumulated locked-in force effects resulting from the construction process,
including the secondary forces from posttensioning
ES Earth surcharge load
EV Vertical pressure from dead load of earth fill

Transient Loads
BR Vehicular braking force
CE Vehicular centrifugal force
CR Creep
CT Vehicular collision force
CV Vessel collision force
EQ Earthquake
FR Friction
IC Ice load
IM Vehicular dynamic load allowance
LL Vehicular live load
LS Live load surcharge
PL Pedestrian live load
SE Settlement
SH Shrinkage
TG Temperature gradient
TU Uniform temperature
WA Water load and stream pressure
WL Wind on live load
Instructor: ENGR. JOSHUA CASTILLO JUNIO 4
 BRIDGE ENGINEERING

WS Wind load on structure

LOAD COMBINATIONS AND LOAD FACTORS

The load factors for various load combinations and permanent loads are given in Tables
3.1 and 3.2, respectively. Explanations of the different limit states are given in the
sections that follow.

3.3.2 Service Limit State

The service limit state refers to restrictions on stresses, deflections, and crack
widths of bridge components that occur under regular service conditions [A1.3.2.2]. For the
service limit state, the resistance factors φ = 1.0, and nearly all of the load factors γi
are equal to 1.0. There are four different service limit state load combinations given in
Table 3.1 to address different design situations [A3.4.1].

Service I
This service limit state refers to the load combination relating to the normal
operational use of the bridge with 55-mph (90-km/h) wind, and with all loads taken at their
nominal values. It also relates to deflection control in buried structures, crack control in
reinforced concrete structures, concrete compressive stress in prestressed concrete
Instructor: ENGR. JOSHUA CASTILLO JUNIO 5
 BRIDGE ENGINEERING

components, and concrete tensile stress related to transverse analysis of concrete segmental
girders. This load combination should also be used for the investigation of slope stability.

Service II
This service limit state refers to the load combination relating only to steel
structures and is intended to control yielding and slip of slip-critical connections due to
vehicular live load. It corresponds to the overload provision for steel structures in past
editions of the AASHTO Standard Specifications.

Service III
This service limit state refers to the load combination for longitudinal analysis
relating to tension in prestressed concrete superstructures with the objective of crack
control and to principal tension in the webs of segmental concrete girders. The statistical
significance of the 0.80 factor on live load is that the event is expected to occur about
once a year for bridges with two traffic lanes, less often for bridges with more than two
traffic lanes, and about once a day for bridges with a single traffic lane. Service I is used
to investigate for compressive stresses in prestressed concrete components.

Service IV
This service limit state refers to the load combination relating only to tension in
prestressed concrete substructures with the objective of crack control. The 0.70 factor on
wind represents an 84-mph (135-km/h) wind. This should result in zero tension in prestressed
concrete substructures for 10- year mean reoccurrence winds.

3.3.3 Fatigue and Fracture Limit State

The fatigue and fracture limit state refers to a set of restrictions on stress range
caused by a design truck. The restrictions depend on the number of stress-range excursions
expected to occur during the design life of the bridge [A1.3.2.3]. They are intended to limit
crack growth under repetitive loads and to prevent fracture due to cumulative stress effects
in steel elements, components, and connections. For the fatigue and fracture limit state, φ
= 1.0. Because the only load effect that causes a large number of repetitive cycles is the
vehicular live load, it is the only load effect that has a nonzero load factor in the fatigue
limit state (see Table 3.1). A load factor of 0.75 is applied to vehicular live load, dynamic
load allowance, and centrifugal force. Use of load factor less than 1.0 is justified because
statistics show that trucks at slightly lower weights cause more repetitive cycles of stress
than those at the weight of the design truck [C3.4.1]. Incidentally, the fatigue design truck
is different than the design truck used to evaluate other force effects. It is defined as a
single truck with a fixed axle spacing [A3.6.1.4.1]. The truck load models are described in
detail in Chapter 4.
Fracture due to fatigue occurs at stress levels below the strength measured in uniaxial
tests. When passing trucks cause a number of relatively high stress excursions, cumulative
damage will occur. When the accumulated damage is large enough, a crack in the material will
start at a point of stress concentration. The crack will grow with repeated stress cycles,
unless observed and arrested, until the member fractures. If fracture of a member results in
collapse of a bridge, the member is called fracture critical.

3.3.4 Strength Limit State

The strength limit state refers to providing sufficient strength or resistance to

satisfy the inequality for the statistically significant load combinations that a bridge is
expected to experience in its design life [A1.3.2.4].
Strength limit states include the evaluation of resistance to bending, shear, torsion,
and axial load. The statistically determined resistance factor φ will usually be less than
1.0 and will have different values for different materials and strength limit states.

Strength I
This strength limit state is the basic load combination relating to normal vehicular
use of the bridge without wind [A3.4.1].

Instructor: ENGR. JOSHUA CASTILLO JUNIO 6

 BRIDGE ENGINEERING

Strength II
This strength limit state is the load combination relating to the use of the bridge by
owner-specified special design vehicles, evaluation permit vehicles, or both without wind.
If a permit vehicle is traveling unescorted, or if the escorts do not provide control, the
basic design vehicular live load may be assumed to occupy the other lanes on the bridge
[A4.6.2.2.4].

Strength III
This strength limit state is the load combination relating to the bridge exposed to
wind velocity exceeding 55 mph (90 km/h). The high winds prevent the presence of significant
live load on the bridge [C3.4.1].

Strength IV
This strength limit state is the load combination relating to very high dead/ live
load force effect ratios. The standard calibration process used to select load factors γi and
resistance factors φ for the strength limit state was carried out for bridges with spans less
than 200 ft (60 m). For the primary components of large-span bridges, the ratio of dead- and
live-load force effects is rather high and could result in a set of resistance factors
different from those found acceptable for small- and medium-span bridges. To avoid using two
sets of resistance factors with the load factors of the strength I limit state, the strength
IV limit state load factors were developed for large-span bridges [C3.4.1].

Strength V
This strength limit state is the load combination relating to normal vehicular use of
the bridge with wind of 55-mph (90-km/h) velocity. The strength V limit state differs from
the strength III limit state by the presence of live load on the bridge, wind on the live
load, and reduced wind on the structure (Table 3.1).

3.3.5 Extreme Event Limit State

The extreme event limit state refers to the structural survival of a bridge during a
major earthquake or flood or when collided by a vessel, vehicle, or ice floe [A1.3.2.5]. The
probability of these events occurring simultaneously is extremely low; therefore, they are
specified to be applied separately. The recurrence interval of extreme events may be
significantly greater than the design life of the bridge [C1.3.2.5]. Under these extreme
conditions, the structure is expected to undergo considerable inelastic deformation by which
locked-in force effects due to TU, TG, CR, SH, and SE are expected to be relieved [C3.4.1]
(see Chapter 6). For the extreme event limit state, φ = 1.0.

Extreme Event I
This extreme event limit state is the load combination relating to earthquake. This
limit state also includes water load WA and friction FR. The probability of a major flood and
an earthquake occurring at the same time is very small. Therefore, water loads and scour
depths based on mean discharges may be warranted [C3.4.1]. Partial live load coincident with
earthquake should be considered. The load factor for live load γEQ shall be determined on a
project-specific basis [A3.4.1]. Suggested values for γEQ are 0.0, 0.5, and 1.0 [C3.4.1].

Extreme Event II
This extreme event limit state is the load combination relating to ice load, collision
by vessels and vehicles, and to certain hydraulic events with reduced live load. The 0.50
live-load factor signifies a low probability of the combined occurrence of the maximum
vehicular live load, other than CT, and the extreme events [C3.4.1].

3.4 Geometric Design Considerations

In water crossings or bridges over deep ravines or across wide valleys, the bridge engineer
is usually not restricted by the geometric design of the highway. However, when two highways
intersect at a grade separation or interchange, the geometric design of the intersection will often

Instructor: ENGR. JOSHUA CASTILLO JUNIO 7

 BRIDGE ENGINEERING

determine the span lengths and selection of bridge type. In this instance, collaboration between
the highway engineer and the bridge engineer during the planning stage is essential.

The bridge engineer must be aware of the design elements that the highway engineer considers
to be important. Both engineers are concerned about appearance, safety, cost, and site conditi ons.
In addition, the highway engineer is concerned about the efficient movement of traffic between the
roadways on different levels, which requires an understanding of the character and composition of
traffic, design speed, and degree of access control so that sight distance, horizontal and vertical
curves, super elevation, cross slopes, and roadway widths can be determined. The document that gives
the geometric standards is A Policy on the Geometric Design of Highways and Streets, AASHTO (2004a).
The requirements in this publication are incorporated in the AASHTO (2004b) LRFD Bridge Design
Specification by reference [A2.3.2.2.3]. In the sections that follow, a few of the requirements
that determine the roadway widths and clearances for bridges are given.

Fig. 3.10 Typical overpass structure. (Courtesy of Modjeski & Masters, Inc.)

3.4.1 Roadway Widths

Crossing a bridge should not convey a sense of restriction, which requires that the
roadway width on the bridge be the same as that of the approaching highway. A typical overpass
structure of a four-lane divided freeway crossing a secondary road is shown in Figure 3.10.
The recommended minimum widths of shoulders and traffic lanes for the roadway on the bridge
are given in Table 3.10.

Instructor: ENGR. JOSHUA CASTILLO JUNIO 8

 BRIDGE ENGINEERING

Table 3.10 Typical roadway widths for freeway overpasses

From A Policy on Geometric Design of Highways and Streets. Copyright ©

2004 by the American Association of State Highway and Transportation Officials,
Washington, D.C.

Fig. 3.11
Cross section for elevated freeways on structure with frontage roads (AASHTO Exhibit 8–10).
(From A Policy on Geometric Design of Highways, and Streets, Copyright © 2004 by the American
Association of State Highway and Transportation Officials, Washington, DC.

A median barrier must separate the traffic for two-way elevated freeways in urban settings
(Fig 3.11). The width of the barrier is 2 ft (0.6 m). The minimum median width is obtained by adding
two left shoulder widths in Table 3.10 to give 10 ft (3.0 m) for a four-lane and 22 ft (6.6 m) for
six- and eight-lane roadways.
If a highway passes under a bridge, it is difficult not to notice the structure and to get a
sense of restriction. As was discussed in the aesthetics section of Chapter 2, it is possible to
increase the sense of openness by placing stub abutments on top of the slopes and providing an open
span beyond the right shoulder. The geometric design requirements are stated in A Policy on Geometric
Design of Highways and Streets, AASHTO (2004a) as follows:

Overpass structures should have liberal lateral clearances on the roadways at each
level. All piers and abutment walls should be suitably offset from the travelled way. The
finished underpass roadway median and off-shoulder slopes should be rounded and there should
be a transition to backslopes to redirect errant vehicles away from protected or unprotected
structural elements.

In some areas it may be too costly to provide liberal lateral clearances and minimum dimensions
are often used. The minimum lateral clearance from the edge of the traveled way to the face of the
protective barrier should be the normal shoulder width given in Table 3.10. This clearance is
illustrated in Figure 3.12 for a typical roadway underpass with a continuous wall or barrier. If
the underpass has a center support, the same lateral clearance dimensions are applicable for a wall
or pier on the left.

Instructor: ENGR. JOSHUA CASTILLO JUNIO 9

 BRIDGE ENGINEERING

Fig. 3.12
Lateral clearances for major roadway underpasses (AASHTO Exhibit 10–6). (From A Policy on
Geometric Design of Highways and Streets, Copyright © 2004 by the American Association of State
Highway and Transportation Officials. Washington, DC.

3.6.2 Vertical Clearances

For bridges over navigable waterways, the U.S. Coast Guard establishes the vertical
clearance [A2.3.3.1]. For bridges over highways, the vertical clearances are given by A Policy
on Geometric Design of Highways and Streets, AASHTO (2004a) [A2.3.3.2]. For freeways and
arterial systems, the minimum vertical clearance is 16 ft (4.9 m) plus an allowance for
several resurfacings of about 6 in. (150 mm). For other routes, a lower vertical clearance
is acceptable, but in no case should it be less than 18 in. (0.5 m) greater than the vehicle
height allowed by state law. In general, a desired minimum vertical clearance of all structures
above the traveled way and shoulders is 16.5 ft (5.0 m).
3.6.3 Interchanges
The geometric design of the intersection of two highways depends on the expected
volumes of through and turning traffic, the topography of the site, and the need to simplify
signing and driver understanding to prevent wrong-way movements. There are a number of tested
interchange designs, and they vary in complexity from the simple two-level overpass with
ramps shown in Figure 3.10 to the four-level directional interchange of Figure 3.13. In
comparing Figures 3.10 and 3.13, note that the bridge requirements for interchanges are
dependent on the geometric design. In Figure 3.10, the bridges are simple overpasses with
relatively linear ramps providing access between the two levels. In Figure 3.13, the through
traffic is handled by an overpass at the lower two levels, but turning movements are handled
by sweeping curved elevated ramps at levels three and four. The geometric design of the
highway engineer can strongly influence the structural design of the bridge engineer. These
engineers must work in concert during the planning phase and share one another’s needs and
desires for integrating the bridge structures into the overall mission of the highway system.

Fig. 3.13 Four-level directional interchange.

Instructor: ENGR. JOSHUA CASTILLO JUNIO 10

 BRIDGE ENGINEERING

3.7 Remarks

In this chapter, we discuss general design considerations that range from the limit state
philosophy of structural design to the calibration of the LRFD specification to the practical matter
of horizontal and vertical clearances. All of these elements make up the design experience and must
be understood by the bridge engineer. For certain, a bridge engineer is an analyst and must be able
to justify a design by making calculations to show that the probable strength is greater than the
probable effect of load by an acceptable safety margin; however, a bridge engineer is more than an
analyst and a number cruncher. A bridge engineer is also concerned about the appearance of the
bridge and whether it can be safely travelled. As mentioned in Chapter 1, a bridge engineer is in
a unique position of responsibility that helps to affect not only the design project from its
beginning to end but also the operation of the structure throughout its life.

REFERENCES

1. Richard M. Barker and Jay A. Puckett, Design of Highway Bridges: An LRFD Approach, Second
Edition. © 2007 John Wiley & Sons, Inc. ISBN: 978-0-471-69758-9

2. AASHTO (2004). LRFD Bridge Design Specification, 3rd ed., American Association of
State Highway and Transportation Officials, Washington, DC.

PROBLEMS

3.1 What are the main reasons for choosing the probabilistic limit states philosophy of LRFD over
the deterministic design philosophy of ASD?

3.2 Discuss the influence that residual stresses had on the selection of a limit states design
philosophy.

3.3 The AASHTO LRFD basic design expression includes a load modifier term η. What is the purpose of
this modifier? Why is it on the load side of the inequality?

3.4 How does the amount of ductility in structural members, represented by ηD, affect the reliability
of bridge structures?

3.5 Why are the live-load factors in service II and service III not equal to 1.0?

Instructor: ENGR. JOSHUA CASTILLO JUNIO 11