See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/245492083

Concrete Bridge Management: From Design to Maintenance

Article in Practice Periodical on Structural Design and Construction · May 1998

DOI: 10.1061/(ASCE)1084-0680(1998)3:2(68)

CITATIONS READS
15 426

2 authors:

Jorge de Brito Fernando A. Branco

University of Lisbon Technical University of Lisbon
1,675 PUBLICATIONS 43,685 CITATIONS 305 PUBLICATIONS 5,035 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Jorge de Brito on 16 July 2015.

The user has requested enhancement of the downloaded file.

 CONCRETE BRIDGE MANAGEMENT: FROM DESIGN TO MAINTENANCE

By J. de Brito1 and F. A. Branco2

Abstract: A bridge service life is limited both by its structural deterioration and

its functional obsolescence as traffic demands grow. In order to rehabilitate and

upgrade existing bridges, important investments are necessary. Bridge budgets are

always limited, and frequently only a selection of the problems detected can be dealt

with by the owners. To be aware of existing in-service problems and to help in

rationalizing maintenance decisions, bridge management systems have been developed

and put to use in several countries.

This paper considers concrete bridge management within a global methodology

to be implemented by the authorities, in which the design and construction stages,

where important measures to attain durability must be implemented, are also

considered. A concrete bridge management system, ready to be implemented at the

service stage, is presented, incorporating a knowledge-based system that allows

standardization of inspections and assistance in maintenance / repair decision-making

based on safety considerations and an economic analysis.

INTRODUCTION

The focus of bridge engineering has, in the last few decades, slowly been
changing from the design of new bridges to the maintenance, rehabilitation and
upgrading of existing ones. This is due to the high structural deterioration rates that
have been
_____________________________________________________________________
1
Assistant Professor, Instituto Superior Técnico, Av. Rovisco Pais, 1096 Lisboa Codex, Lisbon,

Portugal
2
Full Professor, Instituto Superior Técnico, Av. Rovisco Pais, 1096 Lisboa Codex, Lisbon,

Portugal
observed in some structures and the functionality loss that has often occurred due to

increasing traffic volumes or axle loads. Causes for structural degradation can be

found in poor durability design, lack of quality control during construction, increasing

levels of pollution and, most importantly, the absence of regular inspection and

maintenance actions.

The most important costs associated with structural deterioration of a bridge

beyond the minimum acceptable level of safety are not the direct costs of repair, even

though these can amount to a very big investment, but the functional costs due to

traffic detours or disruption, specially near important urban or industrial areas.

Therefore, functional costs should be taken into account in the original design,

encouraging options with longer life between major repairs, even if the minimum

initial cost to achieve the target level of service is increased. Such problems as the

need to properly and easily inspect and maintain the bridge on a regular basis

(accessibility) should also be taken into account in a durability minded design.

The functional failure costs mentioned above will also occur due to a deficiency

in long-term traffic planning. In some countries, unexpected general or local economic

booms have led to the functional obsolescence of strategic utilities such as bridges,

long before the predicted end of their structural life. When designing a bridge, two

important points must be taken into account: upgrading a bridge a few years after it has

been built can lead to costs higher than the initial construction; the upgrading always

causes very high functional costs concerning the surrounding utilities and it is

sometimes impossible to implement. Therefore, a conservative traffic design is quite

often the best option.

Such a wide range of potential problems leads to necessities, that easily become

impossible to accommodate within the existing budgets. Therefore, it becomes

necessary to rate the potential works in terms of priority, according to several decisive
parameters. In order to rationalize such maintenance / rehabilitation decisions, bridge

management systems have been developed and are being implemented, mostly in

Europe and North America [de Brito 1992].

The volume of constructions to be inspected and the increasing maintenance /

repair budgets led, as a first step, to a standardization of the inspection procedures,

namely with the development of inspection manuals and the implementation of

databases. The experience thus gained is presently being used to develop new

management systems, in which human-based decisions are progressively being

replaced by expert knowledge-based criteria fed into computers. In these systems, the

main module contains the decision criteria, that will lead to the optimal repair decision

considering safety, durability, functionality and economy [de Brito et al 1994, de Brito

and Branco1994].

Bridge authorities have also begun to understand that bridge management

systems can be used to obtain longer service lives and reduce maintenance costs, if the

management system is implemented during the design phase and continued throughout

both the construction and service stages.

In this paper, a methodology is presented to analyze concrete bridges from design

to service in terms of durability. Firstly, it considers a durability minded design stage,

followed by a construction process in which procedures are adopted to implement that

same design. During the structure life, bridge management is performed through a

knowledge-based system that allows: storage of the design considerations,

construction and in-field data; standardization of procedures and reports related to

inspections, where anomalies are detected possibly resulting in the updating of design

and construction future measures; and assistance in maintenance / repair decision-

making considering both safety and costs analysis.

DEFINING THE SERVICE LIFE AT THE PLANNING STAGE

The length of the service life will be defined either by physical deterioration or

functional obsolescence of the bridge. If no important structural degradation problems

occur, service life ends when the benefits obtained from the operation are exceeded by

the functional, construction and inspection / maintenance costs. Notwithstanding the

aspects mentioned before, when setting the service life of a bridge careful

consideration should be given to its eventual cultural, scenic or historical value [AIJ

1993]. Taking into account the disruption caused to society in general whenever an

utility is replaced, the planned service life should be specified, in principle, as long as

possible.

The life cycle of a bridge comprises the stages of planning, design, construction,

operation, maintenance / repair and demolition. During these stages, the bridge

authority, the designer, the contractor and the maintenance authorities will be

responsible for the durability and functional behavior of the bridge, each entity with

different specific tasks.

At the planning stage, the authorities should define the global service life that is

intended for the structure. This corresponds to the period of time, from its completion

up to the time when it is supposed to reach a specified state of degradation or

economic / functional obsolescence, on the assumption that normal maintenance is

provided. Current concrete bridges are usually designed for lives of 50 - 60 years, and

important ones are sometimes expected to perform its function for 100 - 120 years.

This period of time is adjusted in order to justify, from an economic point of view, the

construction of the bridge, taking into consideration the associated functional benefits.

Such benefits can be quantified in terms of tolls or simply users time / fuel saving,
even though other benefits such as a reduction in the accident rate or the bridge

environmental impact can also be estimated [de Brito 1992].

The prediction of the end of the functional life of a bridge needs specific traffic

impact studies, that should be prepared during the planning stage. For example, in road

bridges the functional obsolescence is mainly related with restrictions to traffic volume

and maximum axle load, whose evolution can be estimated from statistical analyses or

authorities global development plans. Measures to delay obsolescence may be put into

effect at the design stage, through flexibility which allows for a relatively inexpensive

upgrading of the functionality of a bridge during its life (the possibility of increasing

the number of traffic lanes, for example).

DESIGN FOR DURABILITY

At the design stage, studies of physical deterioration have to be performed to

guarantee that the defined global service life will be met. These studies should be

taken into account in the bridge durability design and the resulting requirements

included in the construction technical specifications.

The service life, defined at the planning stage, is expected to be achieved in most

of the construction elements, mainly in the structural ones, with minor repair costs.

Analysis of durability performed at the design stage considers the service life of

individual bridge components. The result of this analysis is a long-term calendar of

predictable maintenance and repair works throughout the service life. A successful

planning technique is to design those several components that predictably need to be

replaced before the bridge is put out of service with service lives that are sub-multiples

of the bridge global service life.

Estimation of the Service Life

 The estimation of the service life based on physical deterioration is a complex

problem undergoing significant research throughout the world [Mangat and Elgarf

1991, Lopes et al 1995]. It includes the definition of the reference limit states

associated with the end of the service life, the environment characterization, the study

of the degradation phenomena of the materials and components and the definition of

mathematical models for evaluation of the degradation path.

In concrete bridges, the most important degradation mechanisms are associated

with the following situations:

- Carbonation;

- Chloride attack in saline environment;

- Freeze-thaw cycles;

- Chemically aggressive salts.

In order to achieve the expected durability for current structures with service

lives of 50 to 60 years, the European standards [CEN 1995] have defined the specific

recommendations in terms of easily measurable concrete characteristics (minimum

cement content, maximum water / cement ratio, minimum strength) and reinforcement

cover. For important bridges, where greater service lives are specified (100 to 120

years), there are presently no specific code recommendations. In such cases, the study

of the service life must be performed in terms of physical deterioration models

[Mangat and Elgarf 1991], based on local environment conditions experience, and the

limit states adopted for design. Presently, these models have been mainly developed

for carbonation and chloride penetration prediction.

The carbonation of concrete is caused by the reaction of CO2 in the atmosphere

with the Ca(OH)2 of the cement hydration products in the presence of water. The main

result is a loss in alkalinity in the concrete cover causing pH values to approach

neutrality. The penetration of CO2 in concrete pores tends to move as a front which

proceeds at a rate controlled mainly by the CO2 diffusion coefficient [Mangat and

Elgarf 1991]. The evolution of the depth of the carbonation front "d" (mm) can be

estimated by:

d=K t (1)

where t is the time in years and K is a carbonation constant, which depends on the

effective diffusion coefficient of CO2 through concrete and the concentration of CO2

in the immediate surrounding environment and inside the concrete mass. Based on

experimental data, the parameter K has values around 1.0 - 1.5 (mm / year0.5) for

current situations, but can increase to 7.0 - 8.0 for poor concrete and heavily polluted

industrial environments [Mangat and Elgarf 1991].

For chloride attack, its rate of penetration into concrete can be reasonably

predicted by Fick's law of diffusion. The solution of this differential equation, taking

into account the time dependence of the chloride diffusion coefficient Dc (cm2 / sec),

leads to the following equation, giving the chloride concentration C(x, t) inside the

concrete at any depth x (cm) and time t (sec):

C(x, t) = CErro! (2)

where C0 is the equilibrium chloride concentration on the concrete surface (as a

percentage of the weight of cement), erf is the error function and m an empirical

constant (m = 0.4) [Mangat and Elgarf 1991].

Considering that corrosion initiation in reinforcement usually occurs for values

around C = 0.4 % (of weight of cement), the equation above can be used to estimate

the time for the corrosion initiation for given reinforcement covers (x) and surface

chloride concentration levels (Table 1 [Mangat and Elgarf 1991]), using concrete

chloride diffusion coefficients obtained from experimental data.

 After initiation, the deterioration of the bars diameter Dt (at time t) due to

corrosion can be estimated by [Mangat and Elgarf 1991]:

Dt = Di - 0,023 t Ic (3)

where Di is the initial bar diameter in mm, t is the time in years, and Ic (µA/cm2) is the

corrosion rate with values varying from 1x10-1 to 1x102 µA / cm2. The scattering of

these values makes the estimation of the corrosion rate difficult, if not based on in-situ

measurements. It must be noted that the corrosion rate is much higher in chloride

initiated than in carbonation initiated processes, resulting in shorter service lives after

initiation for chloride initiated processes (usually not exceeding 5 years [Tuutti 1982]).

With these models, a design estimation of the degradation process may be

achieved for specific concrete characteristics and reinforcement cover. Usually, several

alternatives may be used in the choice of the concrete components and bars cover, and

a cost efficiency analysis performed to reach a decision. It should be noted that for

very long service lives the minimum concrete strength to be adopted is usually

controlled by durability requirements.

Durability design of concrete bridges may also consider other particular analyses

that affect their service life. These can include, for example, a fatigue analysis of some

components or the study of steel elements corrosion (submerged piles, steel bearings,

etc.) in which protection systems must be prescribed or alternatively an additional

thickness must be adopted to consider the corrosion loss over time.

Bridge Monitoring

For important bridges and because mathematical models, at the present stage of

knowledge, do not yet give sufficiently accurate results, a monitoring system should

also be planned for at the design stage. Monitoring during construction and service life

(associated with periodic inspections) will provide, for important structural elements,
the main parameters that control the deterioration mechanisms, namely in concrete

structures, the carbonation front depth, the corrosion level, etc., and will allow the

confirmation or adjustment (using artificial intelligence) of the deterioration rates

assumed in the design.

Besides the deterioration problems, important bridges should also be monitored to

follow their structural behavior in service. This will enable the prevention of major

accidents and the performance of a quick assessment of the structure after an accident

or unusual environmental action (cyclone or earthquake).

To monitor the functional service life of the bridge, periodic traffic surveys

should also be implemented in order to re-assess the assumptions used in the design

and help in eventual repair decisions, considering the most relevant functional,

economic and social issues.

Flexibility

Components, such as bearings, joints, rails, pavement, etc., that predictably will

need repair or replacement prior to reaching the defined global service life should be

designed with flexibility, meaning that their replacement / repair can be performed

with minor disruption to the facility’s normal operation [AIJ 1993]. If a later

upgrading of the structure functionality is foreseeable, this should also be considered at

the design stage in terms of flexibility.

Technical Specifications and Basic Maintenance Plan

The previous aspects of the durability design should be presented in a set of

technical specifications (durability specifications) for the materials and components,

including the definition of durability tests to be adopted during construction and the

corresponding expected durability parameters for the materials. Based on the design

estimation of the service life of the components, a basic inspection and maintenance
plan for the structure, including the durability and structural monitoring, should be

presented.

In this technical specifications the main parameters related to durability control

should be identified and critical levels defined. The possibility of these levels being

reached during early stages of the service life should trigger off emergency plans in

order to prevent further middle-term effects in terms of both structural and functional

performance of the bridge.

Re-use of materials

In the choice of the materials, the possibility of their re-use at the end of their

service life should also be considered. Guidelines for demolition operations at the end

of the service life may also influence the options at the design stage [AIJ 1993].

CONSTRUCTION FOR DURABILITY

During the construction stage, effective quality control is the best way to attain

the planned service life. The construction quality control is typically performed by a

supervision team, in contact with the designer, and should include the main activities

described below.

Initial Characterization of the Material Properties

Before any concrete mixing begins, the contractor should study the concrete’s

ability to achieve the durability parameters defined at design stage. This is specially

important because durability tests take time to provide results and, if they are not

performed in advance, the construction may suffer some delays. Besides this initial

study, a control must also be implemented at the reception of all the materials, and in

particular for the structural ones, whose composition needs to be periodically checked

with durability tests to guarantee their behavior [CEN 1995]. Alkali-silica related

problems are a good example of the need of materials quality control.

In-situ Control

Implementation of procedures for in-situ measurements of the behavior of the

materials and characteristics of the structure is necessary to guarantee the durability

specifications. This quality control is one of the most important activities and must be

done in a systematic way during the construction stage, according to the monitoring

plan defined at the design stage.

The concrete durability properties of the bridge and the environmental parameters

should be controlled, using the in-situ tests results which will also allow the updating

of the analysis performed at design, in terms of expected service life.

Structural monitoring may also be implemented during construction related to the

evolution of material properties (in-situ concrete shrinkage and creep, Young modulus,

etc.) or to the structure behavior (deck deflections, vibrations, settlements, etc.). In

important bridges or at the authorities request, static and dynamic tests should also be

performed to check the bridge behavior against the design structural models.

Construction and Repair Procedures

Construction methods and repair procedures should be analyzed, specified and

implemented in order to guarantee the best procedures in terms of achieving good

durability for the materials [CEN 1995]. Activities such as concrete compacting and

curing, heat of hydration control, etc. are examples of procedures that are important for

concrete durability.

Knowledge-based systems are also being developed to perform quality control

during construction. Besides the adoption of rationalised procedures for control, if

anomalies are detected, the system gives the inspector suggestions concerning the best

procedures to overcome each problem [Branco and de Brito 1995].

The “Reference State”

 At the end of the bridge construction an overall inspection should be performed

defining the "reference state" that will be used for the service life [Andrey 1987]. Later

on, if this information is not available before an inspection of the bridge (because the

bridge construction precedes the management system implementation or because the

information has been lost), a thorough inspection should be performed, in order to

have a new reference state. The same should happen, when the bridge is reclassified in

terms of traffic usage [de Brito 1992].

MANAGEMENT DURING THE SERVICE LIFE

To guarantee a good behavior of the structure during its service life, several

maintenance and repair measures will need to be implemented. To be aware of existing

problems and to help in rationalizing maintenance decisions (technically and in terms

of costs), management systems have been developed. Initially, they considered a

standardization of the procedures, namely with the development of inspection

manuals, and the implementation of databases. Recently, this information together

with expert knowledge has been fed into computers leading to knowledge-based

systems.

Within a global strategy for management, these systems should include the basic

maintenance plan and service life estimation, defined at the design stage, and all the

durability and structural performance results obtained from monitoring the bridge

during construction, which will be updated throughout the service life.

When planning for the maintenance of a structure, the following parameters

should be taken into account [AIJ 1993]:

- reduction of the consequences of a disaster and assurance of safety;

- continual effective use of the utility;

- effective use of resources and energy saving;

 - minimization of maintenance costs (direct and indirect);

- sustenance of a comfortable and hygienic environment.

The global architecture of a management system of this type consists of three

different modules (Fig. 1 [Branco and de Brito 1995]):

I - Database ;

II - Inspection module;

III - Decision module.

The database must include such information as general description of every

bridge within the net, the design basic maintenance plan, the durability and structural

tests and inspections results from the construction phase, the defect classification

system and correlated knowledge-based data described below, inspection manuals,

costs lists, inspection files, etc. [Branco and de Brito 1995].

The inspection module handles the data acquisition at the structure during its

service life at intervals defined by the inspection calendar. The decision module uses

the data and expert knowledge to help in all the decisions that must be made during

that time concerning the maintenance / repair activities. The decision system is divided

in two sub-modules (Fig. 2 [Branco and de Brito 1995]):

III.1 - Maintenance;

III.2 - Repair .

The maintenance sub-module concerns current maintenance work that is

performed quite frequently. The repair sub-module is related to important structural

repair work and considers the durability and structural updated information obtained

from the inspections. Its use allows the authorities to choose one of several options

(wait and see, repair, build a new structure [de Brito 1992]), resorting when necessary

to a long term present value economic analysis.

THE INSPECTION MODULE

Methodology

The management system functionality is based on a standardized inspection

strategy. It consists of a periodic set of inspections with a fixed timetable, in which

some flexibility is allowed to take into account a plausible global allocation of the

inspection resources, complemented with special inspections when something serious

is detected or suspected.

Typically, a periodic inspection framework consists of two types of inspections

[Andrey 1987]: current and detailed. It is proposed that current inspections have a

period of around 15 months (in order to check the influence of the weather throughout

the year) and that a detailed inspection replaces a current inspection at 5 year intervals.

In the current inspections, serious defects are not expected to be detected. For this

reason, the inspection will be focused mainly on visual observation of the most

exposed areas and, usually, special means of access and non-portable equipment will

be dispensed with. Results from monitoring equipment should also be obtained. The

defects found are registered, as well as their location, and classified in terms of

rehabilitation urgency. A report is prepared and sent to the periodic maintenance

division with all the defects that need short or middle-term action.

The detailed inspection differs from the current inspection in that all details that

are susceptible of raising future problems are investigated. Besides visual observation,

durability in-situ tests should be performed to continue the updating of the service life

estimation.

The detailed inspections should also consider the basic maintenance plan defined

at design stage, according to the estimated service life of the bridge components,

checking the need for their replacement / maintenance, and updating that same plan.
The Inspection Interactive Module

To standardize the procedures during concrete bridge visual inspections and to

help the inspectors with technical information about the defects found, an interactive

knowledge-based system was developed for the inspection module of the bridge

management systems. This module uses a defect classification system considering the

most probable defects liable to be found in concrete structures (totalling 94 entries),

classified according to a geographical / functional / materials criteria in 9 different

groups [de Brito et al 1994].

All the possible causes (direct or indirect) of these defects (117 entries) were

classified according to a chronological criteria in 9 different groups. The in-situ

diagnosis methods used to detect or analyse the defects (81 entries) were also

classified according to the functioning principle and the type of results provided, in 14

different groups. The repair techniques used to eliminate or prevent the defects listed

above (69 entries) were themselves classified in the same groups as the defects [de

Brito et al 1994]. This information was implemented in a prototype system, limited at

the time to the main reinforced concrete corrosion related defects.

For each of these defects, a defect form [de Brito et al 1994] was prepared in

order to be included in the inspection manual, complementing the system. Each of the

forms includes the following information: short description of the defect, possible

causes, possible consequences, inspection parameters to investigate and a defect rating.

To help the inspector in making decisions at the structure site, the system has had

knowledge-based data introduced through correlation matrices relating defects and

causes, defects and diagnosis methods as well as defects and repair techniques [de

Brito et al 1994]. Each of these matrices is organized so that each line represents a

defect and each column a possible cause (or diagnosis method, or repair method). In

the intersection of each line and column, representing the correlation between each
defect and the other element, a classification was introduced representing the

knowledge information.

To help the inspector in making decisions concerning the defects detected during

the inspection, the system is then used as a memo aid giving hints as to what should be

done. The inspector selects the defect that has just been detected and has access to the

following standard help [de Brito et al 1994] (Fig. 3):

1.- Diagnosis Methods;

2.- Probable Causes;

3.- Associated Defects;

4.- Related Repair Techniques.

Using the expert system, the inspector is also able to record the results of the

inspection in a Inspection Database as a Provisional Defect Report. This will then be

used to help in the preparation of the definitive inspection report, which is done at the

headquarters.

THE MAINTENANCE SUB-MODULE

The maintenance activity deals with minor repair problems that do not affect the

structural reliability of the structure. It includes also the bridge elements substitution,

defined in the design basic maintenance plan, which is updated during service life as a

function of the inspections.

Every time a periodic inspection is performed, maintenance works must be

planned and performed at least until the next periodic inspection [de Brito 1992].

Budget limitations are not of a paramount importance as the yearly maintenance

present value costs are more or less stable and therefore are easy to predict and include

in the next budget. However, some type of criteria must be used in order to define
which defects will have to be eliminated as soon as possible and which can wait until

the next inspection report.

The rating criteria to be implemented in the management system usually takes

into account the following three basic aspects [de Brito 1992]:

- Rehabilitation urgency;

- Importance to the structure's stability;

- Functionality affected.

An example of maintenance defect rating to be used by the inspector, is presented

in Table 2 [Branco and de Brito 1995] being the corresponding points considered by

the system to obtain a global rating of the defect. The defects will be then included in

groups of priority of action for maintenance, according to the number of points

assigned to each one. The defects rating allows the system to prepare a list of all the

defects detected according to its priority of action (number of points assigned).

An example of the procedures of the inspector at the bridge site, is presented in

Fig. 4 [de Brito 1992], associated to the rating of a defect (bar with reduced cross

section).

COST FUNCTION

For decision-making concerning important repair work, it is necessary to be able

to quantify the total cost associated with the several stages of the structure life. A

global bridge cost function [de Brito 1992, de Brito and Branco 1994] must include the

structural costs CST and the functional costs and benefits CFU, during the structure life

cycle.

C = C ST + C FU (4)
 The structural costs include the initial costs (C0) (design and construction),

inspection (CI), current maintenance (CM), repair (CR) and structural failure costs

(CFSF).

C ST = C 0 + C I + C M + C R + C FSF (5)

The functional costs (CFFF) are associated with reductions in the operation

conditions of a structure, such as speed limitation, live load reduction, etc. The

benefits (B) correspond to negative functional costs as they are associated with an

improvement of the service level in the structure.

C FU = C FFF − B (6)

Structural Costs

These costs are mainly associated with the civil engineering works and can be

approximately predicted, based on the experience of the owners / authorities. The

initial costs (C0) are the ones involved in designing and building the structure. They

can be easily predicted for new structures, based on current construction costs.

The inspection costs CI are the ones involved in inspecting the structure

regularly. They can be estimated based on the structure dimensions and location, the

authorities current costs (labor and equipment) and a pre-fixed calendar of inspections

[de Brito and Branco 1994].

The maintenance costs CM are the ones involved in keeping the structure at its

design level of service and exclude any main structural work. The yearly maintenance

costs can be predicted as a percentage of the construction cost, or using past

experience. Typical average annual values of these costs for bridges vary between 1

and 2% of the initial costs of the structure [de Brito 1992].

The repair costs CR are the ones involved in doing main structural work (repair,

strengthening) and include the repair costs themselves and all the costs of the
corresponding structural assessment. In the long-term, these costs can be roughly

predicted using a percentage of the construction costs for each year that tends to grow

with the structure's age [de Brito and Branco 1994]. Typical average annual values for

bridges of up to 5% of the initial cost may be considered.

The structural failure costs (CFSF) include all the costs resulting from a structural

collapse of the structure. Even though collapse will not occur under normal

circumstances, these costs can still be considered in an economic analysis as insurance

costs. The cost associated with the structural failure can be obtained from the

probability of failure Pf and the cost of the actual collapse CFF:

C F S F = Pf C F F (7)

The cost CFF is mainly related to the construction of a new structure, but costs

associated with the time in which the existing structure is replaced and the

classification of the structure in terms of patrimony, can also be added by using

empirical coefficients [de Brito 1992].

Functional Failure Costs and Benefits

The functional failure costs (CFFF) are associated with a reduction of the

functionality of the construction. Benefits (B) are the values corresponding to an

enhancement of the structure in order to provide a better service (or be of a wider

utility) than the one provided at a certain standard situation (usually the design stage).

They are necessarily associated with functional failure costs (a benefit is equivalent to

a negative functional failure cost) and are therefore measured with the same

parameters. The functional failure costs have the following components:

a) The costs due to traffic delayed CFFFD that are the ones caused by the slowing

down of the traffic crossing the bridge, specially during rush hours. They are estimated

considering the average delay time and the average user's hour value;
 b) The costs due to traffic detoured in terms of volume CFFFV that are the ones

caused by the traffic being detoured from one particular bridge to others nearby

because of the saturation of the bridge in terms of traffic flow. They are estimated in a

similar way considering the costs associated with the additional travel time, the vehicle

running expenditures and traffic accident rate increase [de Brito and Branco 1994];

c) The costs due to heavy traffic detoured in terms of load CFFFL that are the ones

caused only by a certain margin of exceptionally heavy traffic having to be detoured

from one particular bridge to others nearby because of its insufficient structural

capacity. They are estimated considering the additional total running costs (including

personnel) and traffic accident costs [de Brito and Branco 1994].

Each of these items can be divided in costs in terms of time wasted by the

drivers, fuel costs, vehicle maintenance costs and traffic accident costs. For an

economic analysis, these costs have to be computed using several data, namely, traffic

surveys (yearly and daily), service design level of the road, future traffic estimates,

existing alternatives to each bridge, its traffic and structural capacity, energy and

vehicle maintenance average costs, etc.. The periodic updating of the traffic results,

obtained from the monitoring plan, are fundamental for this analysis.

Fig. 5 [de Brito 1992] presents the estimation of all functional failure costs

during the life time of a bridge case study. The traffic delayed costs CFFFD are nil until

year 11 (time zero being the year in which the bridge is put in-service) in which the

traffic capacity of the bridge is reached, and these costs start to grow at a rather fast

rate. The traffic detoured costs CFFFV are nil during the whole period of the economic

analysis (up to year 23) because no repair works are planned and the bridge is never

saturated in terms of traffic (even though delays exist). Additionally, the detour length

was too long to compensate for the alternative delay. The exceptionally heavy traffic
detoured costs CFFFL exist every year because of the bridge structural capacity

limitation to very high truck loads. The fact that they decrease in time at present value

prices shows that the discount rate is higher than the detoured traffic annual increase

rate.

REPAIR SUB-MODULE

If the inspection detects an important structural anomaly (Table 2) arising from a

natural cause or traffic accident (or detected by the structural monitoring) or if the

updating of the durability properties shows that the bridge safety is getting close to

minimum acceptable levels, an important bridge repair may be envisaged. The repair

solution may also include situations where important traffic saturation of the bridge

occurs (obtained from the traffic monitoring), which may lead to a bridge widening or

even substitution.

In these situations, and to complement the monitoring results, a thorough

structural assessment must be performed, providing data to define and quantify the

repair options. For structural anomalies, this assessment should include a comparison

with the bridge behavior obtained in the initial testing of the bridge after construction.

The decision concerning the repair option to be chosen can then be reached using this

sub-module, through rational criteria and based on the above economic analysis [de

Brito 1992].

Using this analysis, the repair decision can be made according to the cost

effectiveness index (CEI) of each option. The CEI coefficient indicates how well the

proposed work-plan compares to the no-action option. The bigger the coefficient for a

particular option, the better investment that option is. In the calculation of CEI, the

repair costs (CR), the failure costs (CF) and the benefits (B), described above, are

considered.
 CEI = Erro! (8)

To help in the selection of the best structural repair technique the repair module

can be implemented with expert knowledge (in terms of flowcharts - Fig. 6 [Luikens

1993]) to eliminate the repair techniques that are not appropriate for the structural

defect found. In this system the module will ask a set of parameters that characterize

the defect (for example, its location, the repair area and others) and, with them, the

possible repair methods will be pointed out. Some of these parameters, to be defined

by the inspector, will allow an estimation of the costs of the selected technique. If

more than one technique is considered possible, the system uses an optimization

procedure based on the CEI coefficient. Each technique has an associated cost and

estimated service life.

The final economic analysis is then performed, in which all the costs mentioned

before are modelled in order to allow long-term cost prediction. The results can be

very sensitive to some parameters, specially the ones connected with traffic evolution.

Therefore, it is possible to perform a sensitivity analysis, in which a set of parameters

are individually made to vary a certain percentage from a medium value, and the

consequences on the final conclusions give the user an approximate idea of the need to

acquire more accurate predictions of those parameters [de Brito 1992].

CONCLUSIONS

Bridge management must be implemented based on a global methodology,

considering a durability minded design stage and a construction stage in which

procedures are adopted to implement the durability characteristics.

During the bridge life, the management system can include knowledge-based

modules in order to perform and optimize structure inspection strategies and

maintenance / repair policies, updating the basic durability characteristics estimated at

design and construction. This should be implemented supported on a monitoring plan

including also the structural behavior and the traffic evolution with time. Both at the

inspection level and at the decision level, expert technical information helps the

system's user in making a rational management of the resources assigned to each

structure. The use of a global economical analysis is an important tool for repair

decision making.

ACKNOWLEDGEMENTS

Part of the research work presented in this paper was developed at Instituto

Superior Técnico and CMEST - the Structures Research Centre of the Technical

University of Lisboa. This paper also presents results of the EC supported research

project BRITE / EURAM P3091 "Assessment of Performance and Optimal Strategies

for Inspection and Maintenance of Concrete Structures Using Reliability Based Expert

Systems".

REFERENCES

AIJ (Architectural Institute of Japan) “The English Edition of Principal Guide for

Service Life Planning of Buildings”, 1993, Tokyo.

Andrey, D. "Maintenance des Ouvrages d'Art: Méthodologie de Surveillance", Ph.D..

Thesis, École Polytechnique Fédérale de Lausanne, 1987, Lausanne.

Branco, F. and de Brito, J. “Decision Criteria for Concrete Bridge Repair”, Structural

Engineering International, J. Association for Bridge and Structural Engineering, V. 5,

No. 2, pp 92-95, May 1995, Zurich.

CEN (Comité Européen de Normalisation) "Concrete - Performance, Production,

Placing and Compliance Criteria", European Pre-Standard ENV 206 , 7/95, Brussels.
de Brito, J. "Development of a Concrete Bridge Management System" (in Portuguese),

Ph.D.. Thesis in Civil Engineering, Technical University of Lisbon, 1992, Lisbon.

de Brito, J. and Branco, F. "Bridge Management Policy Using Cost Analysis", Civil

Engineering, Structures and Buildings, Journal of the Institution of Civil Engineers,

No. 104, pp 431-439, 1994, London.

de Brito, J.; Branco, F.; Ibañez, M. "A Knowledge-Based System for Concrete Bridge

Inspection", Concrete International - Design & Construction, American Concrete

Institute, pp 59-63, Feb. 1994, Detroit.

Lopes, J.; Branco, F.; Bento, J. "Pre-cast Quality Control Software Manual", CRAFT

CT93-0764 Report T.13, 1995, Axis, Lisbon.

Luikens, G. "Quantification of and Parameters for Defects, Repairs and Repair Costs

for Implementation", BREU P3091 Report T4-3-21, 1993, CSR (Computational Safety

& Reliability), Aalborg.

Mangat, P. S. and Elgarf, M. S. "The Effect of Reinforcement Corrosion on the

Performance of Concrete Structures", BREU P3091 Report T1.4-03, 1991, CSR

(Computational Safety & Reliability), Aalborg.

Tuutti, K. "Corrosion of Steel in Concrete", Swedish Cement and Concrete Research

Institute, 1982, Stockholm.

 Structure Environment C0
(%)
Bridge deck Air zone 1.6
Bridge columns Splash zone 2.5
Bridge column Tidal zone 5.0
Bridge deck De-icing salt 1.6
Bridge column De-icing salt 5.0
Table 1 [1] - Typical chloride surface concentration
CRITERIA CLASSIFICATION POINTS

Rehabilitation 0 immediate action required 30

Urgency 1 short-term (6 months) action required 25
2 medium-term (15 months) action required 15
3 long-term action required 5
_____________________________________________________________________________________________________

Importance to A structural defect in main structural elements 40

the Structure's B quasi-structural defect in main or secondary structural elements 25
Stability C non-structural defect 15
_____________________________________________________________________________________________________

Volume of a t.v. x d.l. x k ≥ 15.000 vehicle km / day 30

Traffic Affected b 15.000 vehicle km / day > t.v. x d.l. x k ≥ 3.000 vehicle km / day 20
by the Defect g t.v. x d.l. x k < 3.000 vehicle km / day 10
t.v. - average daily traffic volume over the bridge (in both directions) [vehicle / day]
d.l. - detour length caused by the total disruption of the bridge [km]
k - degree of obstruction to normal traffic caused by each defect

Table 2 [4] - Example of a maintenance rating criteria for concrete bridges

 D AT A
S T O RAG E

M AN AG EM EN T PRO C ES S ES
S YS T EM S T AN D ARIZAT IO N

D EC IS IO N -
-M AK IN G

Fig. 1 [4] - Main functions of a typical management system

 MAINTENANCE

DEC IS IO N
S YS TEM
INS PEC TIO N
S TRATEG Y
REPAIR

REPAIR W O RK
S ELEC TIO N

Fig. 2 [4] - Organization of the decision system

 DEFECT: A_D05 Bar with reduced cross-section
CROSS-SECTION: deck A-2
TYPE OF INSPECTION: Current Inspection
Please select the Diagnosis Methods you used to conclude the defect:
M_A01 Unaided direct visual observation
M_C01 Galvanic cell test
M_K01 Phenolphthalein
M_A02 Using binoculars, micrometer, camera or video equipment
M_A04 Using special means of aerial access
M_A05 Underwater / on water
M_K02 Silver nitrate
M_K03 Rapid chloride test
i) Suggested diagnostic methods for the defect

DEFECT: A_D05 Bar with reduced cross-section

CROSS-SECTION: deck A-2
TYPE OF INSPECTION: Current Inspection
Please select the probable Causes of the defect:
C_A14 Insufficient reinforcement / pre-stressing design cover
C_A24 Drainage directly over concrete, joint, bearing or anchorage
C_B09 Deficient concrete compacting / curing
C_B11 Inaccurate reinforcement / pre-stressing positioning / detailing
C_F01 Water (wet / dry cycles)
C_F02 Natural carbon dioxide
C_F03 Salt / salty water (chlorides)
C_G01 Water (man-caused)
C_G02 Man-caused carbon dioxide
C_G03 Man-caused de-icing salts
C_A20 Excessive exposed areas in structural elements / faulty geometry
C_A23 No prevision of a minimum inclination in quasi-horizontal surfaces
C_A25 Other drainage design faults
C_A26 Lack of waterproofing membrane
C_A28 Incomplete / contradictory / over compact drawings
C_B01 Wrong interpretation of the drawings
More ↓
ii) Suggested causes of the defect

DEFECT: A_D05 Bar with reduced cross-section

CROSS-SECTION: deck A-2
TYPE OF INSPECTION: Current Inspection
The related repair techniques for the defect are:
(A) HIGH CORRELATION
1. R_D01 Concrete Patching (with reinforcement / pre-stressing cleaning)
2. R_D02 Concrete Patching (with reinforcement / pre-stressing splicing /
replacement)
(B) LOW CORRELATION
not specified
iii) Suggested repair techniques for the defect
Fig. 3 - Example of the use of the inspection interactive module applied to
concrete bridges
DEFECT: A_D05 Bar with reduced cross-section

REHABILITATION URGENCY:

0. Mainly black rust in areas of maximum moments with a local loss over 3%
1. Mainly black rust in areas of maximum moments with a local loss under 3%
2. Predominantly black rust in intermediate areas
3. Predominantly reddish rust

OPTION [0 TO 3]

DEFECT: A_D05 Bar with reduced cross-section

IMPORTANCE TO THE STRUCTURE'S STABILITY:

A. Reinforcement in the deck, main beams, columns, abutments or foundations

C. Reinforcements in the auto-safes, parapets, sidewalks surface and approach
slabs

OPTION [A TO C]

DEFECT: A_D05 Bar with reduced cross-section

AVERAGE DAILY TRAFFIC OVER THE BRIDGE: 20.000 vehicles
DETOUR LENGTH: 5,0 km

VOLUME OF TRAFFIC AFFECTED BY THE DEFECT:

k- degree of obstruction of normal traffic over the bridge caused by the defect

k VALUE [0.0 TO 1.0]

Fig. 4 [5] - Rating the anomalies for maintenance classification

 (x 5) ECU's
200000
CFFFD (traffic delayed)
x 1000$
CFFFV (traffic detoured)
x 1000$
CFFFL
x 1000$(heavy traffic)

100000

0
1990 2000 2010 2020
YEAR

Fig. 5 [5] - Example of the evolution of functional failure costs

 A_D01 Exposed bar
A_D04 Corroded bar
A_D05 Bar with reduced cross-section

yes
Bar corroded ? Cross-section loss > 20% ?

no no unknown yes

R_C02 Concrete patching (with

deteriorated concrete removal)

R_D01 Concrete patching (with R_D02 Concrete patching (with

reinforcement cleaning) reinforcement splicing/
/ replacement)

Fig. 6 [9] - Example of a flow chart for repair selection

View publication stats