Life-Cycle Performance of Deteriorating Structures: Assessment, Design and Management

This paper examines the various aspects of risk, considering in particular how an understanding of these aspects can lead to more effective balancing of risks in the built environment.

This paper describes a probabilistic model for predicting the performance of deteriorating structures under different maintenance strategies. For these structures, the proposed model is used to predict the evolution in time of three performance indicators, namely, condition, safety and cost. The uncertainties associated with these performance indicators are considered. A brief demonstration of the predictive capabilities of the model using a complex data set of random variables characterizing different maintenance strategies for existing concrete bridges is provided.

In this paper, an attempt is made to discuss the relationships among the minimization of Life-Cycle Cost (LCC), the optimal extension of structural service life, and the target safety level by using the multi-objective genetic algorithm.

In this paper is presented a brief review of modern modeling of the reliability profiles for reinforced concrete structures. Deterministic models for the different steps in the deterioration process are discussed. Several of the parameters used in the modeling are so uncertain that a stochastic modeling is natural. By crude Monte Carlo simulation predictions for time to initial corrosion, time to initial cracking, and time to a given crack width may be obtained.

The use of probabilistic — based design of concrete durability has been increased during last ten years. Although the mathematical models and probabilistic characterizations are almost available for several degradation processes, it is still not established what order of probability should be considered as acceptable for design and assessment purposes. In this paper the Life Cycle Cost analysis is used for several structural typologies in buildings and some values of acceptable failure probabilities are obtained. The resulted values are very close to those proposed by the Joint Committee of Structural Safety (JCSS) which are also given in the new version of the Eurocode 0 "Basis of Design" for Serviceability Limit States.

In this paper, spatial variability of structural properties is considered in the reliability analysis of deterioration of concrete structures. It is shown how this model affects the decision making of optimal maintenance strategy for the structures.

An experimental research program to assess the deterioration and the structural performance of reinforced concrete beams is presented. The major focus of this research relates to the identification of a deterioration model that takes into account the concrete quality and different types and extents of environmental attacks (chlorides, sulphates, and carbon dioxide).

Two main factors govern the ingress of chloride ions into concrete reinforced with ordinary steel reinforcement, from de-icing salts: (1) the cover concrete (permeability, thickness), and (2) the microclimatic conditions (humidity, temperature, concentration of de-icing salts) at the concrete surface. A numerical model of chloride transport, taking into consideration environmental conditions (temperature, humidity, snow, rain and salt spreading), was used to predict the chloride profiles in concrete representative of that found in bridge element, for two types of exposure to water (splash, mist). This model was applied for two different regions in Switzerland: on the plateau in Lausanne, where there is a relatively mild winter climate and in the Alps where there is severe winter climate.

Paper is concerned with methodology for design of environmentally sustainable structural systems. It is proposed that whole life performance should be analysed in terms of environmental cost. Due to extensive uncertainties, associated with properties and deterioration models, constrained probabilistic approach is proposed for this life-cycle analysis.

This paper presents the probabilistic extension of a novel approach for the evaluation of the structural performance of concrete structures subjected to the diffusive attack from external aggressive agents. The diffusion process is modeled by using cellular automata and the mechanical damage coupled to diffusion is evaluated by proper material degradation laws. Monte Carlo simulation is applied in order to account for the randomness of the main parameters of the structural problem. The time-variant reliability with respect to proper measures of the structural performance is finally computed.

The use of probabilistic models for deterioration phenomena is presently gaining more and more ground. Typical practical applications are problems dealing with damage accumulation, as induced by e.g. chemical attack (carbonation, corrosion etc.), fatigue or time-dependent resistance (due to loss of prestressing etc.). In most cases the methods applied base on time-invariant reliability theory, treating time as a deterministic palrameter. There are, however, a number of cases, where such an approach would be simplistic and time-variant reliability theory has to be used instead: This is the case, as soon as random processes are involved. The present paper differentiates between the most important situations, starting from the most general case of non-smooth damage accumulation and proceeding to the description of several important special cases, for which it is possible to make some useful simplifications. Special attention is paid to the assumption of smooth deterioration, to models involving an initiation period and to the question, which reliability method has to be applied under which circumstances.

A procedure for the durability design of reinforced concrete structures exposed to de-icing salts is demonstrated and checked setting the focus on site dependent parameters.

This paper presents a probabilistic approach for predicting the chloride contamination of concrete and reinforcing steel corrosion, which takes into account the uncertainty associated with the analytical models of chloride transport, corrosion initiation, as well as damage accumulation, material properties, structural dimensions, and applied environmental and mechanical loads. The proposed approach is illustrated on an aging reinforced concrete bridge deck that has been exposed to chlorides from deicing salts for forty years. An extensive non-destructive and destructive evaluation of the corrosion-damaged deck was undertaken. The field survey data showed a considerable level of variability in all parameters measured with coefficients of variation ranging from 34% for the concrete cover depth to 86% for the diffusion coefficient. The distributions of the chloride concentration at the level of the top reinforcement mat and the time for its corrosion initiation were generated using Monte Carlo Simulation. The simulated results were very close to the field data, which illustrates the prediction capabilities of probabilistic methods as opposed to deterministic methods.

The importance of the role of concrete permeability in the durability of the material was recognized more than a decade ago by Rilem members. A technical committee TC-116-Permeability of concrete as a criterion for its durability was set up and its results were published. This TC-116 promoted an important amount of research and collaborated in advancing the understanding of the relation between concrete permeability and durability. At the time of the initiation of the work of TC-116 durability was a considered from a comprehensive perspective. At the end of the work, it was identified the need to study specifically chloride penetration into concrete, due to the high proportional importance of this type of aggressive in concrete deterioration and the economical relevance of reinforcement corrosion. This was the justification for the aims of present Rilem TC-178.

The causes of deck deterioration include: poor material properties (shrinkage), inadequate pouring procedures, freeze and thaw cycles, truck load spectra (axle loads and groups of axles), fatigue and degradation of concrete under compressive load cycles, corrosion of rebars and usage of deicing chemicals. Various stages of deterioration have been observed (over 50 bridges) and documented. This data serves as a basis for the development of deck performance models as a function of age, type of concrete and traffic volume. Bridge Inventory Data is used to model changes of deck rating during the service years.

The structural performance depends on load and resistance parameters. These parameters are time-varying random variables. Therefore, reliability can be utilized as a rational measure of structural performance. This paper is focused on the effect of material related parameters on the reliability of concrete components. Properties of concrete change with time, depending on many factors including proportions of the mix, initial stages (curing conditions and maturing), concrete volume (effect of hydration heat), and exposure to aggressive environment (pollution, salt).

A new probabilistic fatigue damage model naturally modifying the uniaxial stress-strain relation of the Model Code 90 and the S-N approach by mechanically consistent evolution laws of fatigue is proposed. Its advantage for the assessment of structural degradation and reliability is discussed.

Uncertainties in parameters of any physical model of continuously deteriorating bridges strongly influence management decisions. These uncertainties can be reduced by information obtained through the introduction of health monitoring systems. This paper outlines a Bayesian methodology for performance updating of deteriorating bridges monitored through proactive techniques. The methodology is illustrated with an example of a concrete bridge element subjected to chloride induced deterioration, whereby uncertainties associated with time to corrosion initiation are reduced by the effective integration of monitoring data. This provides a tool to optimise decisions regarding bridge inspection, repair and maintenance activities.

Mechanical degradation of bridge deck, as a result of steel depassivation, using concepts of nonlinear fracture mechanics is discussed. Following sufficient chloride ingress or carbonation, steel corrodes and there is a volumetric expansion which results in radial cracks. Those cracks can then be infiltrated by water which freezes and thus further extend them. This phenomenon severely affects bridge deck serviceability (but not necessarily load carrying capacity) and will eventually result in spalling. In the proposed analysis methodology, the radial cracks are modeled as discrete ones propagating within either a linear elastic concrete surrounding or within a nonlinear one (which also allows formation of secondary smeared cracks).

For many reinforced concrete structures corrosion of reinforcement is an important problem since it can result in expensive maintenance and repair actions. Further, a significant reduction of the load-bearing capacity can occur. One mode of corrosion initiation is that the chloride content around the reinforcement exceeds a critical threshold value. In the present paper a stochastic model is described by which the chloride content in a reinforced concrete structure can be estimated. The chloride ingress is modeled by a 2-dimensional diffusion process and the diffusion coefficient, surface chloride concentration and reinforcement cover depth are modeled by stochastic fields. The paper contains a description of the parameters to be included in a stochastic model and a proposal for the information needed to obtain values for the parameters in order to be able to perform reliability investigations. The distribution of the time to initiation of corrosion is estimated by simulation. As an example a bridge pier in a marine environment is considered.

The paper initially reviews sources of structural deterioration, with emphasis on chloride-induced corrosion of reinforced concrete structures. Thereafter, a critical review of deterioration models for corrosion initiation and propagation and their effect on structural performance is given. It is observed that many probabilistic deterioration models exist and that they can produce somewhat divergent predictions of corrosion initiation and propagation. The paper also examines the high temporal and spatial uncertainty and variability of existing deterioration models and the impact this might have on decision-making criteria. Finally a discussion is given in regard to how to proceed in future research and development with the aim of establishing a consistent and operational model basis for the design, assessment, inspection and maintenance of concrete structures.

This paper presents the results of several probabilistic linear elastic fracture mechanics (LEFM) calculations conducted to study the fatigue behaviour of tubular bridge joints with and without post weld treatment. The effect of post weld treatment is considered through the calculation of an effective stress intensity factor which takes into account the presence of residual stresses (due to welding and post weld treatment) along the crack path. The results of large-scale fatigue tests on tubular bridge truss specimens are predicted using this approach. Specifically, the observed shift of the critical crack initiation site after post weld treatment is predicted.

It is well known that the chloride in concrete can result in rebar corrosion. Due to the expansive nature of rust, rebar corrosion deteriorates the reinforced concrete structures by creating damages in the concrete around the rebar such as cracking and spalling, which affect the durability and service of life of reinforced concrete structures. This study presents a comprehensive model for predicting service life of reinforced concrete structures exposed to chloride attack. The model characterizes various stages in the process of corrosion damage: the initiation of corrosion, the initiation of cracks and critical size of cracks. The initiation process is modeled by coupled chloride and moisture diffusion equations. The initiation of cracks and critical crack length are determined by coupling the corrosion rate, the permeation of corrosion product and mechanical restraint of surrounding concrete. Until the initiation of cracks, the mechanical problem is solved by using elastic theory for thick wall cylinder. After the cracking, the theory of thick wall cylinder is combined with nonlinear fracture mechanics. The period of each deterioration stage can be determined using the present model.

We propose a methodology to support decision making for prioritization seismic reinforcement of existing structures based on the concept of Life Cycle Cost taking into account seismic risk. As an example, the methodology is applied to actual road bridges in Hanshin Expressway Public Corporation and the applicability of this methodology examined.

Most seismic risk assessments for economic decision-making of commercial buildings are based on a risk metric called probable maximum loss (PML) that is associated with losses from an earthquake shaking severity with a 500-year return period. For various reasons, PML is a poor metric for economic performance assessment. This paper introduces an analogous measure, the probable frequent loss (PFL), defined as the mean loss resulting from shaking with 10% exceedance probability in 5 years (an approximately 50-year event). It overcomes many of the problems of PML, and offers the advantage that expected seismic life-cycle costs and expected annualized loss are approximately proportional to PFL through a seismic hazard coefficient that depends on site characteristics, fundamental period, and damage shaking threshold, and can be tabulated for ready use. A brief review is given of a building-specific seismic vulnerability method that may be used to calculate PFL.

This paper provides a model for the reliability analysis of bridge piers subjected to local foundation scour. The reliability analysis shows that current scour analysis models provide lower levels of safety than those observed for bridges subjected to other types of extreme events. Additional safety factors are proposed to increase current scour safety levels to acceptable levels.

A systematic procedure is developed for the reliability-based seismic safety level for bridge life-cycle cost design criteria of continuous PC bridges. In the paper, a set of cost function models for life-cycle cost analysis of bridges is proposed. The total life cycle cost functions consist of initial cost and direct/indirect damage costs considering repair/replacement costs, human losses and property damage costs, road user costs, and indirect regional economic losses. The damage costs are successfully expressed in terms of Park-Ang median global damage indices and damage probabilities. To compute Park-Ang median global damage indices and damage probabilities, SMART-IDARC-BRIDGE, a modified version of the well-known IDARC-BRIDGE, incorporating the simplified Monte Carlo simulations and system reliability for multi-unit PC bridge is used. Earthquake ground motions used to simulate the realistic loading condition are obtained from non-stationary filtered Gaussian processes with both frequency and amplitude modulations based on the Yeh-Wen stochastic ground input motion model. The proposed approach is applied to model bridges of both moderate seismicity regions like Korea and high seismicity regions like Japan and various sensitivity analyses are performed to identify governing parameters. From the results of the numerical investigation, the proposed approach can be effectively utilized for optimal seismic safety level for bridge life-cycle cost design criteria.

Structures should be optimal with respect to the economic investment, the benefits derived from their existence, the expected consequences in case of failure and the degree of protection to human life and limb. This paper presents the implications of a new optimization strategy for the seismic design of structures. A renewal model for the sequence of structural failures is used to define the objective function for optimizing the design of structures in seismic regions. The Life Quality Index, which is a compound social indicator, is included in the optimization for the efficiency of the measures to save human lives. This criterion balances quality-adjusted life years saved against the associated cost for society. The results show that safety standards used in current practice in earthquake engineering should be reviewed in the light of optimization of resources and saving human lives. They also show the importance of different socio-economic characteristics in the definition of risk acceptability.

Reinforced concrete pipes (RCP) are sometimes used to distribute aggressive water in industrial systems. Such networks of RCP deteriorate with time due to internal corrosion. Due to the low O₂ — content of aggressive water, a slow corrosion speed is expected for such applications. If the RCP in these networks are not periodically replaced, they will eventually fail. Strategies for replacement for these pipes depend on (1) the risks associated with failure of a water distribution network, and (2) the costs associated with replacing the pipes, including the cost of removing existing pipes, the cost of the new pipes and the cost of service interruption of there is any temporary closure. Due to the lack of statistical data in regards with RCP failure, the development of a risk-based replacement strategy is not an easy task. This paper gives an example of how predictive models of the deterioration of RCP and the consequences of failure can be used to develop risk-based replacement strategies for RCP networks. Replacement strategies are determined for a series network consisting of 251 RCP. The main contribution of this paper is to provide a theoretical approach based on (1) the results of a reliability study, the use of (2) random drawing (binomial law) and (3) the decision tree method to determine optimal replacement strategy.

For a specific large group of bridges, riveted steel bridges, fig. 1, which were built during the period of time from 1850 to 1940, a set of new methods has been developed and approved which promise to improve considerably the diagnosis, rating and restoration of old steel bridges. Following the often used terminology of medicine, the term minimal invasive methods has been introduced comprising all the relevant methods and techniques. Up to now, the different methods have been part of adjacent fields like non-destructive testing or material testing or experimental structural analysis as well as fracture mechanics as a theoretical tool. However, all of them, in a concerted action, lead to a new quality of investigation and rating. The different methods are presented and some examples of application are given.

Recently, the demand on the practical application of life-cycle cost effectiveness for design and rehabilitation of civil infrastructure is rapidly growing unprecedently in civil engineering practice. Accordingly, in the 21st century, it is almost obvious that life-cycle cost together with value engineering will become a new paradigm for all engineering decision problems in practice. However, in spite of impressive progress in the researches on the LCC, the most researches have only focused on the theoretical point but did not fully incorporate the critical issues for the practical implementation. Thus, this paper is intended to suggest a systemic integrated approach to the practical application of various LCC methodologies for the design and rehabilitation of civil infrastructures. For that purpose, hierarchical definitions of LCC models are presented to categorize the level of LCC assessment applicable for the practical implementation. And then, an integrated LCC system model is introduced with an emphasis on data uncertainty assessment and user-friendly knowledge-based database for its successful implementation.

The use of service life models for reinforced concrete structures encounters several problems in order to be accepted by design engineers and owners as a practical tool for design and redesign. The application a probabilistic service life models for the design of tunnel structures and for the redesign of an existing marine structure has created some doubt regarding the objective choice of parameter values. Moreover, a discrepancy is observed between the output of a probabilistic approach and the owner's objectives. Future developments should be directed at the owner's requirements to supply reliable and understandable results which can serve as an input for effective maintenance planning. This paper addresses some of the aspects regarding the practical relevance of design for durability for steel reinforced concrete structures which have been experienced at the Ministry of Transport in The Netherlands.

Currently available management systems often suggest maintenance strategies that are suboptimal. This is due to inadequate consideration of user costs, mostly in the worst condition state. In this paper an approach is suggested, which introduces reliability aspects into bridge management systems without abandoning the current widely accepted condition-based approach.

This paper presents a practical Life-Cycle Cost (LCC) formulation and procedure for the LCC-effective optimum design of steel bridges. A general LCC model for optimum design of steel bridge consists of those of initial cost and direct/indirect rehabilitation costs of a steel bridge including repair/replacement costs, loss of contents or fatality and injury losses, road user costs, and indirect socio-economic losses. Thus, this paper focuses on the presentation of a general formulation of LCC models and LCC effective design system models for steel bridges suitable for practical implementation. And then, in order to demonstrate the LCC-effectiveness for design of steel bridges, the applications of LCC-effective optimum design to actual steel bridges as illustrative examples are performed with discussion. From the results of the numerical investigation, it may be stated that the optimum design of steel bridges based on the LCC approach proposed in this paper will lead to more rational, economical and safer design compared with initial cost optimum design as well as conventional design.

The identification of cost efficient maintenance strategies for concrete structures usually takes basis in condition assessments achieved through inspections, tests and monitoring. It is well recognized that such condition assessments are subject to significant uncertainties and in general at best provide indications rather than observations about the condition of the structure. The present work proposes a probabilistic framework for the quantification of inspection results for concrete structures in regard to their actual condition and the predicted future degradation. Within this context it is shown how condition indicators, based on non-destructive evaluation methods, can be formulated to enable quality control and inspection and maintenance planning. Furthermore it is investigated how the "strength" of the indicators not only depends on the inspection method but moreover may be significantly dependent on the age of the structure at the time of the inspection.

Load-bearing components of existing panel buildings like balconies deteriorate and need regular inspections and maintenance. Although the current design requirements of national standards were satisfied, the initial reliability of some balcony components might be insufficient. Probabilistic assessment based on ISO 13822 provides effective tools for making decision concerning further use of balconies.

Bridge monitoring system via information technology is capable of providing more accurate knowledge of bridge performance characteristics than traditional strategies. Applying the system to monitor the actual state of the existing bridge promotes effective management and opens up prospect for future application. This article not only presents a Stand-alone Monitoring System (SMS) but also introduces a further effective web-based Internet Monitoring System (IMS) developed for remote monitoring. Based on this integrated system, it is possible to measure and manage bridge performance data such as strain, cable force, displacement and ambient temperature for further evaluation and diagnosis. The SMS and IMS described here were successfully applied to measure a cable-stayed bridge model. The experimental results demonstrate the validity of the system.

In our country, because there are a huge number of civil infrastructure systems, it will be becoming a major social concern to develop an integrated lifetime management system for such infrastructures in the near future. Then we need to develop an innovative system for long-term lifetime management engineering as Doctor for Infrastructure Systems. Namely, it needs to develop an integrated lifetime management system for civil infrastructure systems combined with the latest information processing technologies and intelligent health monitoring techniques, and also to establish the Doctor Degree for Civil Infrastructure Systems (INFRADOCTOR). This paper describes our Japanese strategy of life-cycle management in civil infrastructure systems for the 21st Century as a Center of Excellence (COE) Project in Japan.

The purpose of the paper is to describe and compare deterioration and maintenance models for civil infrastructures. These models can be applied to determine the best maintenance strategy to insure an adequate level of safety at minimal life-cycle cost while taking the uncertainties in the deterioration process into account. Without being complete, a time-dependent deterioration process can be modelled as a failure rate function, a Markov model, a stochastic process or a time-dependent reliability index. The pros and cons of the different models considered are discussed.

The existing nonlinear Finite Element program ATENA and the software FREET, a Probabilistic software for the treatment of Reability problems, were starting point of the development of the method in SARA. These two modules together shall allow a bigger user circle to use the probabilistic assessment technology and to get insight in the deterioration processes of structures.

Civil infra-structures such as roads, bridges, life lines, etc. can fail upon extreme loading requiring replacement, must be replaced after severe deterioration and can become obsolete which usually also requires replacement. Considering the scarcity of resources, especially public money, they should be cost-optimal including initial planning and erection cost, maintenance cost and, possibly, failure, demolition and reconstruction cost. Since time is involved these cost must be discounted appropriately. Here we propose a well-known renewal model and modify it appropriately to cover the majority of cases met in practice, especially extreme value failure, obsolescence and aging.

In order to make a rational decision about a cost-effective solution the life-cycle cost of a structure including cost of structural failure needs to be considered. Due to uncertainties associated with structural properties, loads and environmental conditions the cost of structural failure is a random variable. The paper derives probability distributions of the cost of failure of a single structure and a group of identical structures when single or multiple failures are possible during the service life of a structure. The probability distributions are based on cumulative probabilities of failure of a single structure over its service life. The probability distributions can be employed to evaluate the expected utility, which is then used in decision making. An example, which considers the selection of durability specifications for a reinforced concrete structure built in marine environments, illustrates the use of the derived probability distributions.

This paper presents a model to predict the progress of deterioration of harbor and coastal structures based on a probabilistic approach. In the presented model, the mathematical principle named Markov-chain is applied as the basic theory and the values of transition probabilities are found to fit to the past progress of deterioration. The degree of deterioration is visually judged by the appearance of structures and their ages are also taken into account. It was made clear that the model enables to simulate the past progress of deterioration with high accuracy. Moreover, with using the model, the future progress of deterioration can be predicted. The future costs, a kind of lifecycle cost for remedial actions, were also calculated with the respective average transition probability. It was concluded that strategic regular maintenance (preventive maintenance) quantitatively shows a great advantage over corrective maintenance from the viewpoint of lifecycle costs.

Engineers have been dealing with deterioration in concrete structures since shortly after their initial use. Before 1980 deterioration was taken into consideration predominantly using rules of thumb developed through experience. After 1980 simple engineering models, such as Fick's law, the Hetek's model and time dependent models, were developed and became more widely used. With time these models, both deterministic and stochastic, are becoming increasingly more sophisticated and are increasingly based on "physical" parameters. This trend will continue and future models will be based even more so on the physical, chemical and mechanical laws and principles that govern material and structural behaviour.