Sie sind auf Seite 1von 15

Journal of Structural Engineering Vol. 31, No.1, April–June 2004 pp. 1–15

No. 31-8

Risk-based remaining life assessment of corrosion affected reinforced concrete structural members

K. Balaji Rao , M. B. Anoop , N. Lakshmanan , S. Gopalakrishnan $ and T. V. S. R. Appa Rao #

Remaining life assessment of corrosion affected reinforced concrete structural members is a topic of current R&D worldwide. There is an urgent need to develop more scientific and rational methodologies for this purpose. Number of investigations have already been carried out to assess the damage/distress due to corrosion, that would help in remaining life assessment. While most of these studies are deterministic in nature, efforts are also being made to use probabilistic methods for damage assessment. However, an important aspect in remaining life estimation is the interpretation of the data related to damage/distress and making expert judgement about damage/distress level. Due consideration needs to be given to the quality of the data and the expert interpreting the data. In this paper, expert judgement regarding corrosion damage level is integrated with the structural risk, expressed in terms of probability of attaining a particular damage level, for the remaining life assessment of corrosion affected reinforced concrete structural members. In the proposed methodology, the thinking process of the expert, in corrosion damage assessment, is modelled within a probabilistic framework using Brunswikian theory. The performance of each expert is determined by computing the achievement index. The damage assessment procedure is integrated with Markov Chain model for risk-based remaining life assessment. To illustrate the usefulness of the proposed methodology in determining more rationally the remaining life, an example problem of remaining life assessment of a reinforced concrete bridge girder is considered.

The corrosion of reinforcement in concrete is an issue of major concern, as it affects the safety (due to the reduc- tion in area of reinforcement) and serviceability (due to the formation of rust stains, cracking and spalling) of the structure. The premature deterioration of reinforced concrete (rc) structures has necessitated the need for continual structural health monitoring (SHM) to deter- mine the existence, location and extent (or degree) of corrosion damage, if any, on the structure. SHM is the process of establishing some knowledge of the current condition of the structure or its components 1 , which is required for the performance evaluation of the structure. While SHM is a part of the value chain as proposed by Wong and Yao 2 , the information obtained from SHM needs to be processed further for quantifying the risk, which is an important link in the value chain. As pointed out by Wong and Yao 2 , there exists a gap between where SHM currently stops and where financial deci- sion (remaining useful life) begins, and this gap needs to be filled. This requires the rational assessment of the current damage state based on the data from SHM, and the remaining life estimation using structural risk as the norm. A rational estimation of the current condition and remaining life will help in making engineering decisions

regarding inspection/maintenance activities for the structure. A reliable method for service life estimation of the structure is a pre-requisite for remaining life assess- ment. Systematic approaches/methodologies for ser- vice life prediction have been proposed by various researchers/codes of practice 35 . Expert judgement is iden- tified as an important part of evaluation in most of these methodologies. For instance, the methodology of RILEM TC 31-PCM emphasizes expert judgement as an essen- tial part of evaluation, and points out that ‘the complex- ity of evaluating the interaction of materials or systems with their environment requires expertise that cannot be fully replaced by existing test methods’ 3 . It also identi- fies the use of methods of social-, health-, psychological-, natural- and technical sciences for the expert judge- ment. CIBW80/RILEM 71-PSL 3 pointed out that there is a requirement for an effective mechanism for obtaining and reporting data on in-service performance of structures. Also, it may be difficult to handle the large amount of data acquired from continuous SHM manually. A suitably cre- ated database of KBS would help in handling this amount of data through proper acquisition, representation and

retrieval. Different knowledge-based systems (KBS) have been developed by various agencies involved in the inspec- tion / maintenance of infrastructural facilities (for e.g., Bridge Management Systems such as PONTIS in United States, COSMOS in UK and HiSMIS in Europe). KBS have been developed for damage assessment of the struc- tures subjected to natural hazards such as earthquakes, cyclones (IRAS, WINDRITE, HAZUS, SPERIL). Some of the KBS developed also help in assessing the rate of dete- rioration, which will be useful for remaining life estima- tion. For instance, many of the BMS, including PONTIS, uses a Markov deterioration model to determine the proba- bility of change between consecutive condition states, and thus to assess the degree or rate of deterioration. But the main aim of KBS is to assist the engineer by manipulat- ing large amounts of data and to produce reports to aid the engineer in decision making, the final decisions are left to the engineer 6 . Hence, the recommendations given by the BMS, such as PONTIS, has to be integrated with engi- neering judgement for making rational decisions. Thus, human judgement plays an important role in the dam- age assessment and decision making and in completing the value chain as indicated by Wong and Yao 2 . Human judgmental research is an upcoming field in psychologi- cal sciences, and has found applications in different areas such as settlement of disputes and forensic engineering.

A promising theory which unifies the preferential choice

and judgement is Brunswikian theory, the application of which is researched upon very recently 7 , 8 . For instance, using the concepts of Brunswikian theory, Gigerenzer et al 7 proposed probabilistic mental models (PMM) for mod- elling the human mental process in making decisions. To the authors’ knowledge, none of the existing KBS takes

into consideration the uncertainties associated with the human mental process in decision making. There is a need

to bridge this gap by using suitable models of the human

mental process in the damage assessment and decision making. The authors are involved in carrying out some of the investigations related to the CSIR-IISc collaborative project ‘Structural damage detection using vibration data and probabilistic health assessment’, from 2001. Towards this, a methodology has been developed at SERC for health assessment of bridges using Markov Chain (MC) modelling 9 . The proposed methodology integrates the ear- lier developed MC model with the Brunswikian theory for damage assessment and risk estimation of corrosion affected rc structural members which will help in the remaining life assessment. The article is organised as follows. The proposed proce- dure for corrosion damage assessment using Brunswikian theory is presented in the next section, followed by the Markov Chain model for remaining life assessment of rc structural members subjected to chloride-induced corro- sion of reinforcement by including the results of inspection and the judgement of experts regarding the corrosion dam- age state. An example problem is provided to illustrate the proposed methodology for corrosion damage assessment and remaining life estimation. The concluding remarks are given in the last section.

CORROSION DAMAGE ASSESSMENT

Chloride-induced corrosion of the reinforcement is con- sidered as one of the major mechanisms of degradation of resistance for rc structures located in marine and other aggressive environments. In addition to the reduction in cross-sectional area of steel (resulting in the reduction of resistance), corrosion of reinforcement leads to the crack- ing of cover concrete due to the tensile stresses induced in the cover concrete by the formation of corrosion products. The service life of a rc structural member with respect to corrosion of reinforcement can be divided into two stages:

i) corrosion initiation, and ii) corrosion propagation.

Time for corrosion initiation

The reinforcement in concrete is normally protected against corrosion by a microscopically thin oxide layer formed on the surface of the reinforcement due to the high alka- linity of the surrounding concrete 10 . Chloride ions diffuse from the surface of the concrete through the cover concrete. When the chloride concentration around the reinforcement exceeds a critical value (critical chloride concentration), the protective oxide layer dissolves and corrosion initiates. The diffusion of chlorides through the cover concrete is generally modelled using Fick’s second law of diffusion. The time for corrosion initiation can be determined from Fick’s second law of diffusion as

2

(1)

t i = c

4 D erf 1 c s c cr

c

s

2

where

c

D

c

c

s

cr

clear cover to reinforcement

diffusion coefficient for chlorides in concrete surface chloride concentration critical chloride concentration

Corrosion propagation

Due to corrosion, cross-sectional area of steel reduces. The remaining diameter of the reinforcing bar at any time t, (t) , can be obtained as

(2)

where ( 0 ) t i

Researchers have proposed different models for deter- mining the rate of corrosion of reinforcing bar. From a brief review of these models 11 , it is found that the model proposed by Andrade et al (given in Rodriguez et al 12 ) is widely accepted. Using this model, the rate of corrosion can be determined as,

(3)

where

r corr = 0 . 0115 I corr α

(t) = ( 0 ) r corr (t t i )

initial diameter of the reinforcing bar, in mm time required for corrosion initiation, years

r corr

I corr

rate of corrosion, in mm/year average value of corrosion current density, in

µ A/cm 2

α factor for including the effect of highly localised pitting normally associated with chloride-induced corrosion (varies from 4 to 8)

0.0115 factor which converts µ A/cm 2 to mm/year

Modelling of corrosion damage

Damage is defined as the physical disruption or change in the condition of a structure or its components brought about by external actions and influences, such that some aspect of either the current or future functionality of the structure or its components are impaired 13 . The loss in the area of steel due to corrosion reduces the load carrying capacity of the rc member. This degradation can be modelled by calculating the ‘capacity ratio’, ν(t) , of the member at time t as (Andrade et al 14 ),

ν(t) = R(t)

S

(4)

where R(t) is the load carrying capacity of the member at any time t , and S is the required capacity for the struc- tural member according to relevant design standards. ν(t) is considered as the measure of corrosion damage to the structural member at time ‘ t ’.

Damage assessment

Damage assessment is the process of collecting and evalu- ating the information about the current condition of a struc- ture or its components. A comprehensive assessment of damage also helps in predicting the future performance of the structure. The need for damage assessment of a struc- ture may arise from any of the following 13 :

- The structure has deteriorated through the effects of external environment and loads

- The structure has suffered damage from external envi- ronmental aspects (such as seismic attacks, floods)

- The structure has suffered damage from internal dete- riorating effects

- The structure is being used or is planned to be used for purposes significantly different from the original, such as application of greater live loads

- The structure is under consideration for redesign or structural alterations

- The structure has been exposed to a number of the above detrimental effects and is under consideration for major changes in purpose and/or structural alter- ations.

The assessment of damage to the structure includes the evaluation of the cause(s) of damage, degree and amount of damage, expected progress of damage with time, and the effect of damage on structural behaviour and service- ability. The damage assessment of rc structural compo- nents/structures subjected to chloride-induced corrosion of reinforcement has received considerable attention in recent years. Some of the common methods and procedures that are used for the assessment of corrosion damage in existing rc structures are 15 :

Visual inspection

Concrete cover depth survey

Delamination survey

Half-cell potential survey

Corrosion current measurement

Visual inspection

A visual inspection of rust stains, the severity, type and

location of cracks and concrete spalls gives a first indication

of the extent of corrosion damage in the structure. This is

the simplest, least expensive and the most common method

in conducting a concrete structure survey, and provides pri-

mary information on the integrity of the structure and cor- rosion state of the reinforcement. But the interpretation of the results depends upon the experience of the inspector. Some of the guidelines available in literature for classify- ing the defects based on visual examination are given in Table 1.

 

TABLE 1

GUIDELINES FOR CLASSIFYING DEFECTS BASED ON VISUAL EXAMINATION 16

Rating

Appearance

Cracks (diagonal, longitudinal and transverse)

1 (very slight)

< 1 mm in width

2 (slight)

1–10 mm in width

3 (moderate)

10–20 mm in width

4 (severe)

20–25 mm in width

5 (very severe)

> 25 mm in width

Spalling

1 (very slight)

Barely noticeable

2 (slight)

Clearly noticeable

3 (moderate)

Holes larger than popout of coarse aggregate

4 (severe)

Holes 150 mm (6 inch) in dia. and at least 150 mm (6 inch) deep

5 (very severe)

Holes larger than 150 mm

Concrete cover depth survey

Covermeter, profometer or pachometer are used to deter- mine the concrete cover depth. These devices work on the principle of applying a magnetic field into the concrete surface and recording a change in response due to the

JOURNAL

OF

STRUCTURAL

ENGINEERING VOL. 31, NO.1, APRIL–JUNE 2004

3

TABLE 2 FEATURES OF THE MOST WIDELY-USED METHODS FOR CORROSION DAMAGE STATE ASSESSMENT OF REINFORCED
TABLE 2
FEATURES OF THE MOST WIDELY-USED METHODS FOR CORROSION DAMAGE STATE ASSESSMENT
OF REINFORCED CONCRETE STRUCTURES 17
Methods
Electro-
Electrochemical
Charac-
Gravimetric
Potential
LPM
Guard
Coulostatic
chemical
Impedance
teristics
test
mapping
(
R p )
ring
method
Noise
Spectroscopy
Harmonics
Speed for
individual
◦ ◦
◦ ◦
⊗ ⊗
measurements
Speed of
response to
◦ ◦
◦ ◦
◦ ◦
changes
Quantitative
• ◦
◦ ◦
⊗ ⊗
information
Non-destructive
◦ ◦
◦ ◦
◦ ◦
Non-disturbing
◦ •
⊗ ⊗
◦ ⊗
Measurement
I
E
I
I
I
I
I
corr
corr
corr
corr
corr
I corr ?
corr
corr
parameter
average
mechanisms

Note : Meaning of the symbols:

The method possesses the listed characteristics in an optimal degree, e.g., the individual measurements is instantaneous. The method possesses the listed characteristics in a less than fully-satisfactory degree, e.g., the individual measure- ment is relatively slow. The method does not possess at all the listed characteristics, e.g., the individual measurement is very slow.

rainforcement. Radar scanning method is also being used for the determination of location of reinforcement and the depth of concrete cover.

Delamination survey

Delamination is the separation of a portion of concrete along a plane parallel to the outer surface of the concrete. It is normally located at the level of reinforcement, and is due to the corrosion of reinforcement which will lead ulti- mately to the spalling of the concrete cover. Chain drag and hammer sounding methods are commonly used for detecting and mapping delamination. While the procedure is simple and offers information relating to the corrosion of reinforcement prior to the occurrence of concrete cracking and spalling, the accuracy of the results depends upon the experience of the person performing the survey.

Half-cell potential survey

Since corrosion of reinforcement is an electrochemical process, the main techniques used for the inspection of corrosion damage are electrochemical in nature. Some of the commonly used electrochemical methods for corrosion state assessment are given in Table 2. The corrosion poten- tial ( E corr ) measurement usin a half-cell is the method

most frequently used in the field because of its simplic- ity. In this method, an indication of the relative probability of corrosion activity was obtained empirically through the measurement of the potential difference between a stan- dard portable half-cell placed on the surface of concrete and the reinforcement below. The results of the potential survey can be evaluated on the basis of ASTM guidelines or the JSCE guidelines (Table 3).

Corrosion current measurement

The commonly used electrochemical technique for field assessment of corrosion currents in concrete is the linear polarisation. The 3LP and the Gecor are two devices which employ linear polarisation technique to measure the cor- rosion current, I corr . The main difference between these devices is that the Gecor device has a guard ring electrode which is used to confine the influence area of the counter electrode by actively confining the polarisation current dur- ing the measurement process. The general guidelines for interpreting the results of 3LP and Gecor, as supplied by the instrument manufacturers, are given in Table 4. The I corr value can be transformed into the rate of corrosion for determining the loss in reinforcement area using Eq. 3. Corrosion durability of a rc structural member depends on the environment in which the member is located. Thus,

 

TABLE 3

 

EVALUATION OF HALF-CELL POTENTIAL DATA 18

Half-cell potential reading

 

Corrosion

 
 

(

E corr )

level

Remarks

250 mV E corr

 

I

Black surface with no corrosion

350 mV E corr ≤ − 250 mV

II

Only spots of rust on rebar surface

450 mV E corr ≤ − 350 mV

III

Thin rust layer with corrosion products adhering to concrete

E corr ≤ − 450 mV

 

IV

Formation of expansive corrosion products, with little loss in cross-section

V

Formation of expansive corrosion products, with loss in cross-section

 

TABLE 4

GUIDELINES FOR INTERPRETING RESULTS OF 3LP AND GECOR 15

 

3LP device

Gecor device

I corr A/cm 2 )

Corrosion Damage

I corr A/cm 2 )

Corrosion State

< 0 . 2

No damage expected

< 0 . 1

Passive

0.2–1.0

Damage possible within 10 to 15 years

0.1–0.5

Low corrosion

1.0–10.0

Damage possible within 2 to 10 years

0.5–1.0

Moderate

> 10 . 0

Damage is expected within 2 years

> 1 . 0

High corrosion

it is important to characterise the environment in which the

member is located while carrying out the corrosion dam- age assessment. In most of the codes and standards, the exposure conditions are classified in a general and quali-

tative manner, which leads to ambiguities in the selection of exposure condition for a structure/structural member. The importance of specifying the exposure condition by

a parameter, in the context of prediction of service life of

reinforced concrete members, is brought out by Masters and Brandt 3 . Anoop et al 19 proposed a methodology for quantification of environmental aggressiveness taking into consideration the uncertainties associated with the speci- fication of different environmental parameters (which are in general linguistic, viz., high humidity, medium temper- ature). In this method, the environment is characterised by an environmental aggressiveness factor (EAF), on the basis of environmental variables, namely, temperature, rel- ative humidity and degree of wetting and drying which are obtained from the field. To account for the uncertainties in the values of these variables, they are represented by fuzzy sets. The environmental aggressiveness factor, represented by fuzzy sets, is defined on an arbitrary scale of 0 to 6. The defuzzification of the output fuzzy set for EAF will give the crisp value of EAF quantifying the given exposure

condition. Using the information regarding the envi- ronmental aggressiveness together with the information obtained from visual inspection and field measurements, the corrosion damage assessment of the structural member can be carried out. The information from the field surveys are passed on to an expert or a group of experts for making judgement regarding the corrosion state of the structure. It has to be noted that the correctness of the data from visual inspection depends on the evaluation ability of the person carrying out the inspection. Hence, this data has to be corrected for the expertise (or lack of it) of the observer before pass- ing it to expert(s). Also, there can be uncertainties in data measured using instruments in the form of random mea- surement error, systematic errors from imperfectly cali- brated instruments, and recording and other transmission errors 20 . The measurement data should first be filtered and processed to account for these errors before passing it on to the expert(s). Thus, there is a need to validate the observer and the measurement equipments. Once the data has been passed on to the expert, he has to make a judgement regarding the corrosion state of the struc- tural component/structure based on this data. The expert judgement is an essential part of evaluation, and there will

JOURNAL

OF

STRUCTURAL

ENGINEERING VOL. 31, NO.1, APRIL–JUNE 2004

5

be uncertainties associated with the human mental pro- cess in making judgement. A comprehensive framework for damage assessment should take into account the uncer- tainties associated with human mental process. The human mental process can best be desribed in a probabilistic basis and the Brunswikian theory provides a rational framework for taking into account the uncertainties associated with human mental process in judgement and decision making. A brief overview of the relevant Brunswikian and related concepts pertinent to the present study are given below.

Brunswikian Theory

Brunswik 21 pointed out that one’s knowledge of a dis- tal ‘initial focal variable’ is mediated by more proximal ‘cues’ (or information) that one has about it. According to Brunswik 22 , individuals are generally and in many situa- tions competent, but within the framework of such an opti- mistic estimation of the individual’s scope, the levels of capability (the sharpness of their ‘lenses’) differ drastically among various people. The central and typical feature of the organisms coming into grips with his/her environment can be seen as a ’lens’. Brunswik preferred this concise term to express the multifaceted human competence of an individual to reach focal goals in judgements or actions 22 . The lens model proposed by Brunswik 21 conceptually rep- resented the situation wherein one individual has to make a judgement about the true state of the distal variable using multiple pieces of information.

Bruswik lens model

According to Brunswik lens model, the environmental structure and the organismic system are joined with a linking function that incorporates the subject’s beliefs or assumptions about the causal texture of the environment 23 . The environment is defined by its inherent objects, and at the same time, by the subjective transformations or inter- pretations of these objects. In the necessary process of perception, Brunswikian aspects of distality and proximal- ity are components formed of objects. Objects are facts which may be relevant in their autonomous qualities for psychological analysis. Therefore, the processes within the organism are undoubtedly subjective, but a person’s ori- entation within such processes is partially directed to or determined by objects mostly independent of subjective decisions. The paradigm of the ‘lens’ can be regarded as a gener- alised process characteristic for the construction of human judgements and action – individually and in social contexts. On the other hand, the lens shows openness, flexibility and variety, facilitates communication like a network and solves problems of unexpected matters 22 . Its combination of goal direction and virtuosity in perception and action leads to a concentration of essentials.

Generalised linear model

Brunswik methodology of regression together with the gen- eralised linear model constitute a very useful framework to

conceptualise the general research process in psychology 23 . These concepts helps to understand the level of calibration of the expert to the environment. Using these concepts, the correlation between the judgement and the actual environ- ment (represented as the achievement of the expert, r a ) can be determined as follows.

FIG. 1. SCHEMATIC REPRESENTATION OF LENS MODEL
FIG. 1. SCHEMATIC REPRESENTATION OF LENS MODEL

A simple lens model is shown in Fig. 1. In this figure, Y e denotes the distal variable (criterion), X i ’s, the cues and Y s , the judgement made by the expert. Data ( X i s) have been collected for a number of base line cases with known crite- rion values, Y e s and the expert’s judgement ( Y s s) have been obtained for these cases. Now, linear regression equations can be formulated on the judgmental and environment side as

s

Y

= a +

and

e

Y

= c +

k

i = 1

k

i = 1

b i X i

d i X i

(5)

(6)

respectively, using multivariate linear regression.

Let R s = γ Y s ,Y

s

and R e = γ Y e ,Y be the values of coeffi-

e

cient of determination on the judgement side and the envi-

ronment side, respectively. R s is the correlation between the actual judgement and the predictive judgement, which

is a measure of the cognitive consistency of the expert. R e is the correlation between the actual criterion value and the predicted criterion value, which is an index of how pre- dictable the environment is. The correlation between the

is given by the matching index,

. The value of G denotes how well the

predicted values Y

G, i.e., G = γ Y

s

s

,Y

e

and Y

e

model of expert corresponds to the model of the environ- ment. The achievement index, r a , which is the correlation between the judgement and the criterion is given by

(7)

where C is the configurality index, defined as the correla- tion between the residuals of the two regression equations,

r a = GR s R e + C 1 R 1 R

2

s

2

e

] . The configurality index accounts

for nonlinearities in the environment (criterion) side and the judgement side. The achievement index, r a , can be regarded as a measure of the accuracy of the judgements made by the expert.

i.e., C = γ [ Y s Y

s

]

, [ Y e Y

e

Confidence limits

Based on the assumption that people are good judges of the reliability of their knowledge, Gigerenzer et al 7 proposed probabilistic mental models (PMM) for cognitive processes in judgement. Two important aspects of PMM are that prob- abilistic inference is part of the cognitive process and that uncertainty is part of the outcome. The PMM models are consistent with human thinking process in the way that the expert would assign a confidence level for his judgement. This treatment would enable to characterise the thinking process with respect to various confidence levels. The over- or under-confidence limits associated with an expert for the different confidence levels are determined based on the judgements made on a number of baseline cases. The over- or under-confidence limit for a given confidence level, ‘ i ’, for a given expert ‘j ’, can be determined as

over - or under - confidence, ouc i = n i (p i f i )

n

(8)

where n - total number of decisions made by the expert ‘ j p i - confidence judgement (mean value of confi- dence level) n i - number of times the confidence judgement p i was used by the expert ‘j f i - relative frequency of correct answers for all decisions for which confidence p i was assigned by the expert ‘j ’ The over- or under-confidence takes into account the relative bias of the expert through the term n i . Brehmer and Hagafors 24 expanded the Brunwikian lens model to a multilevel lens model to study the use of experts in complex judgement making. Such a multilevel lens model is used in the present study to model the cor- rosion damage state assessment of rc structural members. The procedure for corrosion damage assessment using Brunswikian theory is given below.

Corrosion

Damage

Assessment

Using

Brunswikian

Theory

The distal stimuli of the multilevel lens model, used in cor- rosion damage assessment of rc structural members, is the corrosion of reinforcement, which gives rise to the proximal stimuli to the observer/instrument in the form of appear- ance and corrosion current/potential. The information on proximal stimuli (such as rust stains, amount of crack- ing and spalling, corrosion current density) are recorded by the observer/instrument (cues). These cues, together with corrosion state of reinforcement are the distal stim- uli for the expert, who is making a decision regarding the

corrosion damage state. The information recorded by the observer/instrument (cues) are corrected for the evaluation ability/human error (in the case of human observer) and for the detection capability and correctness of detection (in the case of instrument). The corrected data is the proximal stimuli for the expert who makes a judgement regarding the distal stimuli, namely, the corrosion state of the rein- forcement. The following cues are used in this study for the corro- sion damage assessment

(i)

Rust stains (visual inspection)

(ii)

Cracking (visual inspection)

(iii)

Spalling (visual inspection)

(iv)

Delamination (chain drag or hammer sounding)

(v)

E corr (Half-cell potential mapping)

(vi)

I corr (linear polarisation)

(vii)

Remaining area of reinforcement (radiography)

(viii)

Cover thickness (cover meter)

(ix)

Temperature

(x)

Relative humidity

(xi)

Degree of wetting and drying (The cues ix–xi are used for determining the EAF quantifying the environment 19 ).

In this study, five corrosion damage states have been

identified for rc structural members subjected to corro- sion of reinforcement based on the guidelines available in literature 14 . The definition of these damage states together with the associated range of capacity ratio for rc struc- tural members in these states are given in Table 5. These ranges of capacity ratio are based on the guidelines given by CEB 25 . By integrating the information required for corrosion damage assessment (Tables 1-5), and supplying the same along with cues, the aim would be to rationally capture the thinking process of an expert in arriving at the judgement regarding the damage state. It is known that the mental process can best be described in the probabilistic basis. A number of experts are asked to make judgement regarding the corrosion damage state independently using the same set of cues. The expert is asked to identify the corrosion damage state(s) in which he believes the member is in, and to attach confidence level(s) for his judgement from a confidence scale. The confidence scale consists of seven categories I to VII representing 0%, 1%–20%, 21%–40%, 41%–60%, 61%–80%, 81%–99% and 100% confidence, respectively. The corresponding confidence judgement p i (mean value of the confidence level) for different confi- dence levels are 0%, 10%, 30%, 50%, 70%, 90% and 100%, respectively. Consistent with probabilistic mental thinking, the experts would judge the probable damage states of cor- rosion affected rc structural members, along with respective confidence levels. Once this information is obtained from different experts for a given case, it remains how to finally judge the damage state. The final decision should take into account in some form the damage information provided by different experts. Instead of classifying judges as experts or non-experts, it is better to consider them rational to differ- ent degrees 26 . This requires that the achievement of experts to be quantified.

JOURNAL

OF

STRUCTURAL

ENGINEERING VOL. 31, NO.1, APRIL–JUNE 2004

7

 

TABLE 5

CORROSION DAMAGE STATES FOR THE REINFORCED CONCRETE STRUCTURAL MEMBERS

Corrosion

   

Damage State

 

Description

Capacity ratio

No

No visible signs of corrosion E corr > 250 mV

1.0

Slight

Rust stains barely noticeable; some cracking; no spalling;

0.95

350 mV E corr ≤ − 250 mV

 

Rust stains; several longitudinal cracks; some cracks

 

Medium

in stirrup direction; clearly noticeable spalling in some places;

0.80–0.95

450 mV E corr ≤ − 350 mV

5% loss in area of reinforcement

Severe

Highly noticeable rust stains; extensive cracking and spalling; E corr ≤ − 450 mV 10% loss in area of reinforcement

0.60–0.80

Very Severe

Highly noticeable rust stains; extensive cracking and spalling; steel is no more in contact with concrete at some places; E corr ≤ − 450 mV 25% loss in area of reinforcement

0.35–0.60

The achievement ( r a ) of each expert is determined using the generalised linear model based on a number of base- line cases. Previous data on corrosion damage assessment of distressed rc structural members or data from labora- tory experiments are used for this purpose. Similarly, the values of over- or under-confidence of each expert for different confidence levels are also determined using the PMM theory. Suppose an expert has identified corrosion damage state(s) and assigned confidence level(s) to these corrosion damage state(s) based on the given data. Then the state probabilities (probability that the structure is in damage state k ) can be determined as

P k =

p k

5

k = 1

p k

(9)

where p k is the confidence judgement for the confidence level assigned to the damage state k . For the corrosion dam- age states not identified by the expert, the value of p k is taken as 0%. The state vector for the corrosion damage state based on the judgement of the expert ‘ j ’ is

{ P } j = { P 1 , P 2 , P 3 , P 4 , P 5 }

(10)

If there are n experts who have been asked to make judge- ments independently using the same set of cues, then the state vector for the corrosion damage state combining the judgements of all the experts can be obtained as

{

P c } =

n

j = 1

w j { P } j

(11)

is the corrosion damage state vector based

on the judgement of the j th expert, and w

attributed to the judgement of the j th expert. The weights reflect the accuracy of the expert in making the judgement,

is the weight

where { P }

j

j

and can be obtained from the values of achievement (r a ) for the experts as

w j =

r a

j

m

n

j = 1

r a j m ;

m 0

(12)

where r a j is the achievement of the j th expert and m is a value reflecting the degree of importance. When m = 0, all the experts have been attributed the same weight. As the value of m increases, the degree of importance attached with the achievement of the expert increases. The over- or under-confidence associated with the judge- ment of the j th expert is given by

EC j =

5

k = 1

ouc k

(13)

where ouc k is the over- or under-confidence of the expert for the confidence level associated with damage state k . The value of EC j defines the bounds for the damage state probabilities based on the expert’s judgements. This esti- mate gives a rational way of determining the confidence to

be attached with the judgement of an expert, which is only possible using the PMM by modelling human mental pro- cess on a probabilistic basis. The over- or under-confidence associated with the corrosion damage state obtained by combining the judgements of all the experts is given by

EC c =

n

j = 1

w j EC j

(14)

The value of EC c will be useful as a measure of the confi- dence that can be put on the final corrosion damage state by

processing the judgements of all the experts. A lower value

of EC c denotes tighter bounds on the damage state assessed

and vice versa. The evaluation of the corrosion damage state is useful for predicting the future performance of the

structure, and hence in the remaining life assessment. The procedure proposed in the present study for remaining life assessment of corrosion affected rc structural members is presented in the next section.

REMAINING LIFE ASSESSMENT

A rational estimation of remaining life of corrosion affected

rc structural members is required for making engineering decisions regarding the inspection/maintenance activities.

In this paper, the Markov Chain (MC) model that has been

developed at SERC for health assessment of rc bridges 9 , 27

is used for determining the remaining life of the structural

member. The corrosion damage state (represented by the

capacity ratio) of the rc structural member at a given time is a random variable. Also, due to the corrosion propagation, the corrosion damage state changes with time. Thus, the evolution of the corrosion damage of the structural mem- ber with time has to be modelled as a random process. MC

is a stochastic process model in which both state space and

index space are discrete 28 . While some of existing BMS, such as PONTIS, uses a Markov deterioration model to determine the probability of change between consecutive condition states, and thus to assess the degree or rate of deterioration, in the present study, the MC model is inte- grated with the Brunwikian theory to take into account the probabilistic nature of human mental process. In this case, the state space is the corrosion damage state (Table 5) of the member and the index space is the time. The stochas- tic evolution of the system, modelled by homogeneous MC can be completely described by the Transition Probabil- ity Matrix (TPM), P. Typically, element p ij represents the probability that the corrosion damage state of the structural member will be i in the next year given that the corrosion damage state at present is j . P is a one-step transition prob- ability matrix the elements of which can be evaluated ana- lytically once the rate of corrosion propagation is known. The n -step TPM is given by

P n = P × P ×

× P (n-times)

(15)

where

P = { p ij } i 1 ,N ; j 1 ,N , N = number of states

P n = { p ij (n) } i 1 ,N ; j 1 ,N

where

p ij (n) = Probability { system state will by ‘j ’ after n - transitions, starting from state ‘ i }

If the index space is discretised into 1 year intervals, then

p ij (n) gives the probability of the system being in corro-

sion damage state ‘j ’ at the end of n years given that the

corrosion damage state at the beginning was ‘ i ’.

Life Assessment of a New Structure

For a new structure, the time to corrosion initiation and the rate of corrosion can be determined using Eq. 1 and Eq. 3, respectively, if the values of D, c s , c cr , I corr and α are known. These values are selected based on type and grade of concrete, water-cement ratio and exposure condition 29 . The capacity ratio at any time ‘t ’, determined using Eq. 4, will be a random variable due to the uncertainties associ- ated with the workmanship and the aggressiveness of the environment to which the structural member is exposed to. By computing the values of ν(t) for two consecutive years, the 1-step TPM, P, can be computed. The n -step TPM, P n , can be computed for determining the corrosion dam- age state of the structural member at the end of ‘n ’ years using Eq. 15. The corrosion damage state probabilities at any time can be determined from the n-step TPM for that time period, using the methodology given by Balaji Rao and Appa Rao 27 . By comparing the capacity ratio at any time with a target value, the service life of the structure with respect to safety can be determined. The target value for capacity ratio (min- imum safety level) is fixed based on engineering judgement and/or experience gained based on performance of simi- lar structures. Different options for determining the min- imum safety level have been discussed by Das 30 . CEB 25 gives guidelines for classifying the urgency of repairing or strengthening structural elements undergoing damage. According to CEB, capacity ratio values lower than about 0.5 would require immediate repair action. The probability of capacity ratio being less than or equal to 0.5 at different time instances can be determined from the n-step TPM 27 . This information would help in determining the service life of the rc structural member; the service life being the time at which the probability of capacity ratio being less than or equal to 0.5 becomes equal to a prefixed value, which in the present study is assumed to be 10 2 .

Remaining Life Assessment Including the Inspection Results

The MC model can also be used for determining the remaining service life of existing rc structural members, for which inspections have been carried out and expert judgement information regarding damage state are avail- able. For existing structural members, the corrosion rates can be determined by measuring corrosion currents in the field with linear polarization instruments. The state vector for the corrosion damage of the structural member at the time of inspection is obtained from the judgements given

JOURNAL

OF

STRUCTURAL

ENGINEERING VOL. 31, NO.1, APRIL–JUNE 2004

9

by experts using Brunswikian theory. Let { P c } be the corro- sion damage state vector obtained using Brunswikian the- ory, and P m be the 1-step TPM for the structural member for one year determined using the measured corrosion rate. The corrosion damage state vector after n -years from the time of inspection is given by

{P c (n) }={ P c } T ( P m ) n

The probability of capacity ratio being less than or equal to 0.5 at different time instances can be determined from the corrosion damage state vector corresponding to that time using the methodology proposed by Balaji Rao and Appa Rao 27 , and the remaining service life of the structural member can be determined by comparing with the target probability. A schematic representation of remaining life assessment without inspection and including the informa- tion from inspection is shown in Fig. 2. The practical appli- cation of the proposed methodology is illustrated through an example problem below.

(16)

FIG. 2. PROBABILITY OF THE STRUCTURAL MEMBER BEING IN A STATE REQUIRING IMMEDIATE REPAIR ACTION
FIG. 2. PROBABILITY OF THE STRUCTURAL MEMBER
BEING IN A STATE REQUIRING IMMEDIATE REPAIR
ACTION ACCORDING TO CEB [23] (SCHEMATIC
REPRESENTATION)
EXAMPLE

Service life of a reinforced concrete T -beam for a bridge is estimated using the proposed methodology. The beam is to be designed for a bending moment of 1252.6 kNm to carry the loads. The beam is exposed to the environ- mental conditions characterized by: temperature of around 35 C, RH of about 70% and degree of wetting and dry- ing around 0.6. The given environment can be defined as exposure class 4a, i.e., sea-water environment (due to high value of degree of wetting and drying), according to Euro Code 2 classifications of exposure. The minimum grade of concrete and maximum water-cement ratio for this class of exposure is 30 MPa and 0.55 respectively and the minimum required cover thickness is 40 mm. The value of environ- mental aggressiveness factor (EAF) for a temperature of around 35 C, RH of about 70% and degree of wetting and drying around 0.6 is obtained as 4.784 using the method- ology proposed by Anoop et al 19 . The beam is designed satisfying the safety and ser- viceability requirements specified in the code of prac- tice. The cross sectional details of the beam are given in

Fig. 3 . The mean values for D, c s and c cr are selected as 5 × 10 8 cm 2 / s, 0.30% by weight of concrete and 0.125% by weight of concrete, respectively, which are represen- tative of the values reported for similar environment for concrete with similar properties 31 , 32 . To account for the variations in the values of these variables due to uncer- tainties arising due the workmanship and aggressiveness of environment, they are considered to be random. The values of coefficient of variation (COV) chosen for these variables are given in Table 6 33 . Using these values, the mean and COV of time for corrosion initiation are determined using Eq. 1 as 8 years and 0.20, respectively.

FIG. 3. CROSS SECTION OF T-BEAM CONSIDERED
FIG. 3. CROSS SECTION OF T-BEAM CONSIDERED

The value of I corr and α are taken as 8. 75 µ A/cm 2 and 5.75 based on water-cement ratio and exposure condition 29 . The rate of corrosion is determined as 0.58 mm/year using Eq. 3. The rate of corrosion is considered as a random vari- able with the mean as 0.58 and COV as 0.30 to account for the uncertainties arising due the workmanship and aggres- siveness of environment. The cover to reinforcement (d ), compressive strength of concrete (f ck ) and yield strength of steel ( f y ) are also considered as random variables with mean and COV values as given in Table 6 to account for the uncertainties in the workmanship. The 1-step TPM, P, is computed and the corrosion damage state probabilities are determined. The values of corrosion damage state proba- bilities plotted against the age of thestructural member are shown in Fig. 4, and the probability{capacity ratio 0 . 5} with age of the structural member is shown in Fig. 5. From Fig. 5, it is noted that at 19 years of age, the prob- ability of capacity ratio being less than or equal to 0.5 becomes 0.01. An inspection is carried out at this time. The information (cues) obtained from the inspection is given in Table 7. It is assumed that these information have already been corrected for the evaluation ability/human error (in the case of human observer) and for the detection capabil- ity and correctness of detection (in the case of instrument). This information (cues) are passed on to five experts, who have been asked to make judgements regarding the corro- sion damage state and to assign confidence levels for their judgements from the confidence scale of I-VII.

Evaluation of the Experts

The experts have been evaluated based on their perfor- mance in making judgement regarding corrosion damage state for rc structural members for which actual corrosion damage states are known. One hundred baseline cases are

 

TABLE 6

RANDOM VARIABLES CONSIDERED IN THE EXAMPLE PROBLEM

 

Variable

 

Mean

COV

Diffusion coefficient, D

5 × 10 8 cm 2 /s

0.10

Surface chloride concentration, c s

0.30% by wt. of concrete

0.10

Critical chloride concentration, c cr

0.125% by wt. of concrete

0.05

Cover thickness, d

 

40

mm

0.05

Rate of corrosion, r corr

0.58 mm/year

0.30

Compressive strength of concrete, f ck

 

30

MPa

0.18

Yield strength of steel, f y

 

415 Mpa

0.12

(Note- The values of COV are from Enright and Frangopol 33 )

FIG. 4. CORROSION DAMAGE STATE PROBABILITIES FOR THE RC STRUCTURAL MEMBER IN CONSIDERED THE EXAMPLE
FIG. 4. CORROSION DAMAGE STATE PROBABILITIES FOR THE RC STRUCTURAL MEMBER IN
CONSIDERED
THE
EXAMPLE
PROBLEM
FIG. 5. PROBABILITY OF THE STRUCTURAL MEMBER BE- ING IN A STATE REQUIRING IMMEDIATE REPAIR
FIG. 5. PROBABILITY OF THE STRUCTURAL MEMBER BE-
ING IN A STATE REQUIRING IMMEDIATE REPAIR
ACTION ACCORDING TO CEB [23]

assumed for this purpose, out of which 8 were in corrosion damage state 1, 21 were in corrosion damage state 2, 36 corrosion damage state 3, 24 were in corrosion damage

were in state 4 and 11 were in corrosion damage state 5. Five experts are considered in this study. All these five experts are considered to be equally informed and are equally capable in terms of qualification to assess

the damage. The information (cues) on the 100 struc- tural members (baseline cases) were passed on to these experts, who have been asked to make judgements regard- ing the corrosion damage state and to assign confidence levels for their judgements from the confidence scale of I-

VII.

The over- or under-confidence (ouc i ) for the confidence judgement p i for the different experts can be determined using Eq. 8. Detailed calculations for determining ouc i have been carried out typically for expert 1 (Table 8). It is assumed that similar computations have been made for the other four experts, and the results are plotted in Fig. 6. The values of ouc i reflect the probabilistic nature of mental process in judgemental decision making.

JOURNAL

OF

STRUCTURAL

ENGINEERING VOL. 31, NO.1, APRIL–JUNE 2004

11

 

TABLE 7

DATA FROM INSPECTION FOR THE EXAMPLE PROBLEM

From Visual Inspection:

 

Rust stains

Highly noticeable rust stains

Cracking

Several longitudinal cracks; some cracks in stirrup direction

Spalling

clearly noticeable spalling

From Field Measurements:

 

I corr (3LP)

 

6

. 0 A/cm 2

E

corr

 

450 mV

Cover depth

 

38 mm

Remaining diameter of reinforcement

 

32.0 mm

 

TABLE 8

 

RESULTS OF ASSESSMENT MADE BY EXPERT 1 (CHARACTERISED IN TERMS OF ouc i )

Confidence

       

Level (i)

p

i

n

i

f

i

ouc i

I 0.0

   

0

0

 

0

II 0.1

 

30

0

 

0.88

III 0.3

 

19

0.105

 

2.52

IV 0.5

 

16

0.438

 

0.68

V 0.7

 

22

0.682

 

0.27

VI 0.9

 

24

0.958

0.95

VII 1.0

 

53

1.0

 

0

The achievement of the expert (defined by the achieve- ment index, r a ), reflecting the correlation between the judgement and the environment, can be determined using the linear regression model (Eq. 7) for each expert. In the present study, it is assumed that such an analysis has been carried out and the values of r a have been obtained for all the five experts. It is assumed that the experts have similar tendencies in viewing the environment and making judge- ments (viz. all having positive correlations). The values of weight attributed to the judgement of each expert are deter- mined using Eq. 12. The values of achievement index and

FIG. 6. OVER- OR UNDER-CONFIDENCE ASSOCIATED WITH DIFFERENT CONFIDENCE LEVELS FOR THE EX- PORTS CONSIDERED
FIG. 6. OVER- OR UNDER-CONFIDENCE ASSOCIATED WITH
DIFFERENT CONFIDENCE LEVELS FOR THE EX-
PORTS CONSIDERED
 

TABLE 9

 

ACHIEVEMENT INDEX AND WEIGHTS FOR THE EXPERTS

Expert No. (j)

r

a j

w

j

1 0.93

 

0.215

2 0.90

 

0.208

3 0.85

 

0.197

4 0.81

 

0.188

5 0.83

 

0.192

the weights for the five experts considered are given in

Table 9.

Corrosion Damage Assessment

The judgements regarding the corrosion damage state and corresponding confidence levels for these judgements given by all the five experts are presented in Table 10. Using these values, the corrosion damage state probabilities are determined using Eq. 9, and the state vector for the cor- rosion damage state combining the judgements of all the experts is obtained using Eq. 11. These probabilities are given in Table 11. The over- or under-confidence associ- ated with the corrosion damage state obtained by combin- ing the judgements of all the experts is also determined using Eq. 14 as 1.12% (the positive sign denoting over- confidence).

Remaining Life Assessment Including the Inspection Results

The 1-step TPM, P, for the structural member for one year is determined using the measured corrosion rate. Using the corrosion damage state vector obtained using the judge- ments of the experts and P, the corrosion damage state vec- tors at different time instances are determined using Eq. 16. The values of corrosion damage state probabilities plotted

 

TABLE 10

 

EXPERTS’ JUDGEMENT ON CORROSION DAMAGE STATE AND ASSOCIATED CONFIDENCE LEVEL FOR THE EXAMPLE PROBLEM

Corrosion

 

Confidence Level

 

Damage State

Expert 1

Expert 2

Expert 3

Expert 4

Expert 5

1

-

-

-

-

-

2

-

-

-

-

-

3

III

IV

VI

III

V

4

VI

VI

V

V

IV

5

-

-

-

-

-

(Note: ‘-’ indicates not applicable)

 

TABLE 11

CORROSION DAMAGE STATE PROBABILITIES BASED ON EXPERTS’ JUDGEMENT FOR THE EXAMPLE PROBLEM

Corrosion

 

Damage state probabilities

 

Damage

           

State

Expert 1

Expert 2

Expert 3

Expert 4

Expert 5

Combined

1

0

0

0

0

0

0

2

0

0

0

0

0

0

3

0.25

0.357

0.563

0.30

0.58

0.407

4

0.75

0.643

0.437

0.70

0.42

0.593

5

0

0

0

0

0

0

against the age of structural member are shown in Fig. 4, and the probability{ capacity ratio 0 . 5} with age of the structural member is shown in Fig. 5. It can be noted from Fig. 5 that the probability {capacity ratio 0 . 5} = 0 . 01 when the structural member is 22 years of age. Thus the remaining life of the structural member can be consid- ered to be 3 years from the time of inspection against the limit state of probability{capacity ratio 0 . 5} = 0 . 01. It is noted that the remaining life, estimated using the corrosion inspection results and expert judgement, has an over-confidence of about 1.21%. Thus, by carrying out an inspection, the engineer has now the option to postpone the repair activities upto a period of three years for the prob- lem considered. This type of information can be generated using the proposed methodology, which will be useful for making decisions regarding repair, and hence the decisions taken not only completes the value chain as proposed by Wong and Yao 2 , but also is an improvement over the engi- neering decisions obtained using PONTIS.

CONCLUSIONS

This paper presents a methodology for damage assess- ment and risk-based remaining life estimation of corrosion affected rc structural members, by integrating research on modelling thinking process with expert judgement. The proposed methodology tries to capture the thinking pro- cess of the expert on a probabilistic basis which has been embedded through the expert’s confidence in his judge- ment. One of the highlights of the methodology is that those involved in making judgement are considered to be rational, rather than classifying them as experts and non-experts, and the performance of each expert is determined by comput- ing the achievement index. This is in line with the current thinking in risk perception and risk communication 26 , and would help in creating a more effective KBS for damage assessment. The damage assessment methodology is inte- grated with Markov Chain model, which is currently being researched upon at SERC for risk analysis. An example

JOURNAL

OF

STRUCTURAL

ENGINEERING VOL. 31, NO.1, APRIL–JUNE 2004

13

problem of a rc bridge girder is presented, which illustrates the usefulness of the proposed methodology in facilitating decision making regarding repair.

ACKNOWLEDGEMENT

This paper is published with the kind permission of the Director, Structural Engineering Research Centre, Chen- nai. The research work presented in this paper is carried out under the CSIR-IISc Collaborative project “Structural damage detection using vibration data and probabilistic health assessment” (D.O. No. 70(0039)/00/EMR-II dated

11.01.2002).

REFERENCES

1.

Beck, J.L, Au, S.-K and Vanik, M.W., “Monitor- ing structural health using a probabilistic measure”, Computer-Aided Civil and Infrastructure Engineer- ing, Vol. 16, No. 1, 2001, pp. 1–11.

2.

Wong, F.S. and Yao, J.T.P., “Health monitoring and structural reliability as a value chain”, Computer- Aided Civil and Infrastructure Engineering, Vol. 16, No. 1, 2001, pp. 71–78.

3.

Masters, L.W. and Brandt, E., “Prediction of ser- vice life of building materials and components - CIB W80/RILEM 71-PSL-Final Report”, Materials and Structures, RILEM, Vol. 20, 1987, pp. 55–77.

4.

ACI 365.1R-00, “Service life prediction - State-of- the-art report”, ACI Manual of Concrete Practice,

2000.

5.

ASTM E632-81, “Standard practice for developing accelerated tests to aid prediction of service life of building components and materials”, American Soci- ety for Testing of Materials, 1981.

6.

Ryall, M.J., “Bridge management”, Butterworth- Heineman, 2001.

7.

Gigerenzer, G., Hoffrage, U. and Kleinbolting, H., “Probabilistic mental models: A Brunswikian theory of confidence”, Psychological Review, Vol. 98, No. 4, 1991, pp. 506–528.

8.

Adelman, L., Miller, S.L., Henderson, D.V. and Scholles, M., “Using Brunswikian theory and a lon- gitudinal design to study how hierarchical teams adapt to increasing levels of time pressure”, Acto Psy- chologica, Vol. 112, 2003, pp. 181–206.

9.

Balaji Rao, K. and Appa Rao, T.V.S.R., “A method- ology for condition assessment of rc bridge girders with limited inspection data”, The Bridge & Struc- tural Engineer, Ing-IABSE, Vol. 29, No. 4, 1999, pp. 13–26.

10.

Kropp, J. and Hilsdorf, H.K., “Criteria for concrete durability”, E & FN Spon, London, 1995.

11.

Balaji Rao, K., Anoop, M.B. and Appa Rao T.V.S.R., “A critical review of corrosion propagation model and results of probabilistic analysis of resistance degradation of reinforced concrete flexural mem- bers due to corrosion”, Project Report No. SS-OLP 07741-RR-2000-1, Structural Engineering Research Centre, Chennai, May 2000.

12.

Rodriguez J., Ortega L.M., Casal J. and Diez J.M., “Assessing structural conditions of concrete struc- tures with corroded reinforcement”, in ’Concrete repair, rehabilitation and protection’, R.K. Dhir and M.R. Jones (Eds.), E&FN Spon, London, 1996, pp. 65–78.

13.

fib, “Structural Concrete: Textbook on behaviour, design and performance Updated knowledge of the CEB/FIP Model Code 1990 - Volume 3”, Interna- tional Federation for Structural Concrete, 1999.

14.

Andrade, C., Alonso, M.C. and Gonzalez, J.A., “An initial effort to use the corrosion rate measurements for estimating rebar durability”, in ’Corrosion rates of steel in concrete’, N.S. Berke, V. Chaker, and D. Whiting (Eds.), ASTM STP 1065, American Soci- ety for Testing and Materials, Philadelphia, 1990, pp. 29–37.

15.

Gu, P., Ramachandran, V.S. and Beaudoin, J.J, “Cor- rosion of steel in concrete: assessment techniques”, in ’Testing and Quality Control in Cement Industry:

Progress in Cement and Concrete Vol. 3’, Akademia Books International, New Delhi, 1997.

16.

RILEM Draft Recommendations, “Draft recommen- dations for damage classification of concrete struc- tures”, Materials and Structures, RILEM, Vol. 27, No. 170, 1994, pp. 362–369.

17.

Gonzalez, J.A., Feliu, S., Rodriguez, P., Lopez, W., Ramirez, E., Alonso, C. and Andrade, C., “Some questions on the corrosion of steel in concrete. Part II: corrosion mechanism and monitoring, service life prediction and protection methods, Materials and Structures, Vol. 29, No. 186, 1996, pp. 97–104.

18.

Japanese Society of Civil Engineers, “Present status and future trend in research related to reinforcement corrosion, corrosion protection and repair”, Con- crete Engineering Series 26, 1997.

19.

Anoop, M.B., Balaji Rao, K., and Appa Rao, T.V.S.R., “Application of fuzzy sets for estimat- ing the service life of reinforced concrete structural members in corrosive environments”, Engineering Structures, Vol. 24, No. 9, 2002, pp. 1229–1242.

20.

fib, “Monitoring and safety evaluation of exist-

ing

concrete

structures:

state-of-the-art

report”,

International Federation for Structural Concrete,

2003.

21.

Brunswik, E., “The conceptual framework of psy- chology”, University of Chicago, 1952.

22. Wolf, B., “Processes of constructing judgments and actions by competent individuals with respect to object orientation: Programmatic ideas in the tradition of Brunswikian thoughts”, Essay #7, The Brunswik Society, http://brunswik.org,

2000.

23. Alvarez, M.M.R., “Contingency judgments”, Article #31, The Brunswik Society Newsletter, http://brunswik.org, 1999.

24. Brehmer, B. and Hagafors, R., “Use of experts in complex judgment decision making: A paradigm for the study of staff work”, Organizational Behaviour and Human Decision Processes, Vol. 38, 1986, pp. 181–195.

25. CEB, “Assessment of concrete structures and design procedures for upgrading”, CEB Bulletin No. 162, 1983, pp. 87–90.

26. Reid, S.G., “Perception and communication of risk and the importance of dependability”, Structural Safety, Vol. 21, No. 4, 1999, pp. 373–384.

27. Balaji Rao, K. and Appa Rao, T.V.S.R., “Stochas- tic modelling of crackwidth in reinforced concrete beams subjected to fatigue loading”, Engineering Structures, Vol. 26, No. 5, 2004, pp. 665–673.

28. Benjamin, J.R. and Cornell, A.C., “Probability, statistics and decision for civil engineers”, Mc-Graw Hill, 1970.

29. Anoop, M.B., Balaji Rao, K. and Appa Rao, T.V.S.R., “A methodology for durability-based service life design of reinforced concrete flexural members”, Magazine of Concrete Research, Vol. 55, No. 3, 2003, pp. 289–303.

30. Das, P.C., “Application of reliability analysis in bridge management”, Engineering Structures, Vol. 20, No. 11, 1998, pp. 957–959.

31. Bamforth, P.B., “The derivation of input data for modelling chloride ingress from eight-year UK coastal exposure trials”, Magazine of Concrete Research, Vol. 51, No. 2, 1999, pp. 87–96.

32. Hobbs, D.W., “Chloride ingress and chloride- induced corrosion in reinforced concrete members”, Proc. of 4 th International Symposium on Corro- sion of reinforcement in concrete construction, Cam- bridge, 1996, pp. 124–135.

33. Enright, M.P. and Frangopol, D.M., “Probabilistic analysis of resistance degradation of reinforced con- crete bridge beams under corrosion”, Engineering structures, Vol. 20, No. 11, 1998, pp. 960–971.

JOURNAL

OF

STRUCTURAL

ENGINEERING VOL. 31, NO.1, APRIL–JUNE 2004

15