Vulnerability, Uncertainty, and Risk: Analysis, Modeling, and Management

In this paper, an attempt is made to apply evolutionary computing for earthquake disaster restoration and prevention problems. The relationships among early restoration, minimization of life-cycle cost and target safety level of road network are discussed by using multi-objective genetic algorithm. A comprehensive disaster prevention program is developed based on the recognition that road networks may be unavoidably damaged when very strong earthquakes occur. By introducing the concept of multi-objective optimization into the restoration and prevention scheduling for earthquake disasters, it is possible to find out several near-optimal restoration and prevention scheduling plans.

Quantitative risk analysis (QRA) enables rational accounting for the effect of spatial scale on estimates of damage to individual (multi-component) structures or to groups of closely spaced structures during severe windstorms. The case of the window breakage on the facade of Boston's John Hancock Tower, which motivated a series of such risk-based studies, is summarized. We further describe and illustrate an integrated vulnerability model to estimate structural damage in clusters of residential buildings due to tropical-cyclone winds, accounting for the "chain reaction" of events involving wind pressure damage and wind-borne debris damage, amplifying aggregate losses. The last part of the paper provides relevant background information on new methodology to analyze multi-scale random fields, presented in the second (2010) edition of the first author's book Random Fields.

In the majority of decision models used in practice all input data are assumed to be precise. This assumption is made both for random results of measurements, and for constant parameters such as, e.g. costs related to decisions. In reality many of these values are reported in an imprecise way. When this imprecision cannot be related to randomness the fuzzy set theory yields tools for its description. It seems to be important to retain both types of uncertainty, random and fuzzy, while building mathematical models for making decisions. In the paper we propose a fuzzy-Bayesian model for making statistical decisions. In the proposed model the randomness of data is reflected in related risks, and fuzziness is described by possibility measures of dominance such as PSD (Possibility of Strict Dominance) and NSD (Necessity of Strict Dominance). The proposed model allows a decision-maker to reflect in his/hers decisions different types of uncertainty.

Two broadly different approaches have been proposed for handling data that contain non-negligible interval uncertainty from censoring, plus-minus digital readouts, and other sources of measurement imprecision or incertitude. Modeling interval data with uniform distributions over their ranges allows relatively straightforward calculation of sample statistics, but does not guarantee these estimates will approach the parameters of the actual distribution, even for asymptotically many random samples. In contrast, modeling interval data as bounds on possible values yields corresponding bounds on sample statistics that are easier to interpret although often more difficult to calculate. We illustrate the approaches in estimating descriptive statistics, empirical distribution functions, and best-fit distributions. Statistical inference under the bounding approach generally yields a class of decisions under the theory of imprecise probabilities. In contrast, the uniforms approach will yield a unique decision (up to indifference), although this decision cannot be said to be implied by the data alone because it depends on ancillary assumptions that may not be tenable.

Recently we proposed a new form of imprecise probability based on the generalized interval, where the probabilistic calculus structure resembles the traditional one in the precise probability because of the Kaucher arithmetic. In this paper, we study the independence properties of the generalized interval probability. It resembles the stochastic independence with proper and improper intervals and supports logic interpretation. The graphoid properties of the independence are investigated.

Whenever a disaster occurs, it's of utmost importance that the rescue system recognizes accurately human behavior and evacuation command in the fire and its black smoke. However, we have infinite pattern for movement instructions by our personality. On the other hand, the singular spectrum analysis method has proposed as analytical method for time-series data. In this paper, we propose a method for acquiring embodied knowledge of human behavior from time-series gesture data using singular spectrum analysis. A behavior is distinguished in terms of gesture characteristic with similarity criteria by interval time-series data. We discuss the usefulness of the proposed method using an example of gesture motion.

Predicting the behavior and reliability of engineering structures and systems is often plagued by uncertainty and imprecision caused by sparse data, poor measurements and subjective information. Accounting for such limitations complicates the mathematical modeling required to obtain realistic results in engineering analyses. The framework of imprecise probabilities provides a mathematical basis to deal with these problems which involve both probabilistic and non-probabilistic sources of uncertainty. A common feature of the various concepts of imprecise probabilities is the consideration of an entire set of probabilistic models in one analysis. But there are differences between the concepts in the mathematical description of this set and in the theoretical connection to the probabilistic models involved. This study is focused on fuzzy probabilities, which combine a probabilistic characterization of variability with a fuzzy characterization of imprecision. We discuss how fuzzy modeling can allow a more nuanced approach than interval-based concepts. The application in an engineering analysis is demonstrated by means of an example.

In dealing with uncertainty and imprecision Fuzzy Project Scheduling (FPS) has been presented as an alternative to well established traditional scheduling methods. The aim of this paper is to propose a process for the estimation of the duration of activities that are to be used in a FPS system. Initially, the sources of uncertainty are distinguished into the project and activity level. Then, the work focuses primarily on the calculation of the duration of activities based on the productivity of resources and the fuzzy parameters that define it. Thereafter, a specific example is presented from earthworks in a motorway project. Finally, the estimated duration can be adjusted based on the perceived risks at the project level. Overall, the paper presents a means for encoding a project manager's perception of uncertainty in the duration of activities that are to be used in FPS.

Despite their limitations as a platform for calculations, Microsoft Excel spreadsheets enjoy widespread use throughout much of engineering and science, and they have emerged as a lingua franca for computations in some quarters. Given their ubiquity, it would be useful if Excel spreadsheets could express uncertainty in inputs and propagate uncertainty through calculations. We describe an add-in for Microsoft Excel that supports arithmetic on uncertain numbers, which include intervals, probability distributions, and p-boxes (i.e., bounds on probability distributions). The software enables native calculations in Excel with these objects and ordinary scalar (real) numbers. The add-in supports basic arithmetic operations (+, –, x, ÷, ˆ, min, max), standard mathematical functions (exp, sqrt, atan, etc.), and Excel-style cell referencing for both function arguments and uncertain number results. Graphical depictions of uncertain numbers are created automatically. Using function overloading, the standard Excel syntax is extended for uncertain numbers so that the software conducts uncertainty analyses almost automatically and does not require users to learn entirely new conventions or special-purpose techniques.

Increasingly, uncertainties are explicitly considered for important engineering tasks. Often, little case-specific information is available for characterizing these uncertainties. Uniform distributions are an easy way to describe errors in absence of more precise information. In many situations, the bounds are fixed based on user experience. The extended uniform distribution (EUD) provides a probability density function that accounts for higher orders of uncertainty (uncertainty of uncertainty) when using a uniform distribution to describe errors. Since the EUD accounts for several orders of uncertainty it is more representative than uniform and curvilinear distributions. The extended uniform distribution helps increase the reliability and robustness of tasks requiring uncertainty combination through better representing incomplete knowledge of parameters.

Standard Monte Carlo (sMC) simulation models have been widely used in AEC industry research to address system uncertainties. Although the benefits of probabilistic simulation analyses over deterministic methods are well documented, the sMC simulation technique is quite sensitive to the probability distributions of the input variables. This phenomenon becomes highly pronounced when the region of interest within the joint probability distribution (a function of the input variables) is small. In such cases, the standard Monte Carlo approach is often impractical from a computational standpoint. In this paper, a comparative analysis of standard Monte Carlo simulation to Markov Chain Monte Carlo with subset simulation (MCMC/ss) is presented. The MCMC/ss technique constitutes a more complex simulation method (relative to sMC), wherein a structured sampling algorithm is employed in place of completely randomized sampling. Consequently, gains in computational efficiency can be made. The two simulation methods are compared via theoretical case studies.

Recent natural disasters such as typhoons and heavy rains have frequently occurred in Japan. Moreover, people have failed to escape because of improper evacuation activities and a lack of post-disaster risk awareness; this includes fire or tsunamis caused by earthquakes. Almost all of the victims evacuate using a hazard map, evacuation signs, or with assistance from local residents. However, the local community tends to rarefy than before. Therefore, hazard maps and evacuation signs are more important than ever. In our research, we studied the influence of leaders during evacuation using a multi-agent model based on flood evacuation system for decreasing the number of victims who failed to get out in time by the indication of the leader during evacuation to the escaping residence. Through our implementation of this multi-agent model based on flood evacuation simulation, we discuss the requisite number of leaders during evacuation and the arrangement of such leaders in the city of Takatsuki.

After Hanshin-Awaji huge earthquake disaster in 1994, it has been important for Japanese local Government to build a comprehensive evacuation program of a large earthquake occurrence. In this program, it is substantial to give appropriate information on human behavior for the evacuation time. Thus, a new methodology based on behavior-oriented agent system should be developed. In this study, the production rules of the attributive groups were constructed in terms of the questionnaire survey for civilian return-trips from the working or the shopping places during earthquake disaster. Next, using the set of production rules composed of the questionnaire data, a multi-agent system model for return trips in a hypothetical large-scaled earthquake was built by a MAS method. It comes to the conclusion that the human behaviors during the earthquake impact were constructed by multi-agent system model and the possibility of the return-home was found in view of the conditions of the roads and the human attributes.

Designing marine and maritime systems require the probabilistic characterization of sea waves in the time-history and spectral domains. These probabilistic models include parameters that can be empirically estimated based on limited data in durations, locations and applicability to particular designs. Characterizing the statistical uncertainties associated with the parameters and the models is an essential step for risk-based design methods. This paper introduces proposed statistical methods to assess confidence intervals for prediction of sea conditions at points of interest in the sea environment. The methodology is briefly introduced and demonstrated using illustrative simulation examples based on notional information.

When evaluating the validity of the reasoning process of a transportation plan, a reasoning map facilitates discourse. If the truth values are attached to the premise and each link of the map, the overall truth of the reasoning process can be measured. The map is useful to examine the sensitivity of the truth with respect to the changes in knowledge and opinions. In calculating the truth, Bayesian inference measures it in probability measure. Demspter-Shafer (D-S) theory measures it in Belief and Plausibility. This paper examines the differences between these two methods using an example that deals with selection of the mode of public transportation in a large commercial complex.

This paper presents a simulation framework designed to plan the movements of unmanned autonomous systems (UAS's) in hazardous environments, to coordinate their actions, predict their behavior and evaluate their mission success in various combat situations. Current simulation methods do not predict the complex interrelations among vehicles, operating environments, and paths, thus providing inadequate tests. A family of methods for coordinating, positioning, routing and assessing diverse military units or "agents" (such as unmanned ground vehicles) is described in this paper. The methods are tested and evaluated through computer simulations to ensure suitability for operating unmanned autonomous systems (UAS's). The path evaluation is performed using a dynamic GIS, distance transform, and genetic algorithms. The optimization algorithms for use in testing future unmanned systems are based on multiple objectives and criteria, including: (1) timeliness, (2) detectability & exposure time, (3) probabilities of survival & mission completion; (4) energy use; and (5) obstacle avoidance. A series of tests are presented which mimics real-world combat situation to test the effectiveness of the developed algorithms.

This paper describes the application of fuzzy models of drivers' gap-acceptance behavior at priority intersection; differently from other experiences, in this work gap-acceptance data were collected from experiments of driving behavior performed using a fixed-base driving simulator. The experiments have been conducted at the Transportation Laboratory of the University of Padova, using STSoftware® driving simulator. The proposed fuzzy model allows to overtake problems related to both non-homogeneous explanatory variables and to uncertain and imprecise information on the system. Nevertheless, the findings appear interesting because allow to better understand the effects of explanatory variables not detectable from direct observations (on site) on driver's gap-acceptance behavior with the aim to use this knowledge to perform effective operational analysis (maneuver capacity and LOS assessment).

In this paper basic problems are considered of developing the entropy-probabilistic approach for solving problems of assessing the safety of interdependent CI. This approach allows overcoming the problem of heterogeneity of source data, as all physical parameters of various engineering problems are brought down to one dimensionless measure — the system entropy.

Many critical infrastructure systems are comprised of complex physical, geographical, and logical networks. Such systems include electric power, drinking water, wastewater, cellular communication, internet, and transportation. These systems are vulnerable to hazards, both natural (e.g. hurricanes and earthquakes) and man-made (e.g. terrorism and accidents), which can induce failures in network elements and reduce system performance. In conducting risk and reliability analyses for complex infrastructure systems, network theory has been used to understand the effect of perturbations of individual network elements on overall system performance. In this paper, we present a survey of research that has employed this network theoretic approach and provide a discussion of future research needs in the field.

It is known, that in case of world economic destabilization prognostic mistakes can not only lead to deterioration of a government but also promote strengthening of crisis tendencies in regions. It makes experts to improving algorithms of social and economic forecasting and gradually to replacing ‘static' and ‘quasi-static' models of a society by ‘dynamic', which includes nonlinear feedback through elements of an economic infrastructure. Quite often this feedback becomes "critical", that is it starts process of transition of territory's economic complex to a new condition with unpredictable in advance properties. In these cases we often speak about occurrence of "critical infrastructures", those are capable to affect essentially on social, economic, environmental, etc. safety of territory. In advance, it is not always known, that infrastructures can appear critical, therefore the risk-analysis should include a stage of their detection on the basis of preliminary formed system of criteria. Designing of criteria of criticality and carrying out risk-analysis becomes complex methodological problem in conditions of continuous change of the importance of parameters of regional social and economic system. Their decision demands participation of the big number of experts which should understand each other and for this purpose to apply the uniform approach to definition of concept of "critical infrastructures". Until now different authors define in their own way terms they use. There is a situation then the significant number of infrastructures drops out of consideration as "critical" because they did not render essential influence on social and economic position of region. The role of the specified infrastructures can suddenly increases in conditions of world crisis. The specified circumstance has caused necessity of revealing of such "latent" potential infrastructures and has become the reason of performance of the presented research. In the work we started with discussed before assumption, according to which occurrence of critical infrastructures influences essentially on stability and as a consequence on safety of regional social and economic system (hereinafter — ‘regional SES') [1]. Thus, for revealing conditions of "criticality" of an economic infrastructure it is necessary to make the list and to estimate a level of regional risks. The analysis of publications of various authors has shown that usually basic risks and concept of safety of the regional SES are connected with conditions of existence and ability to react to external influences. Thus existence of regional system during life cycle is postulated as preservation of its qualitative definiteness even at change of structure of system and functions of its elements [2] (figure 1). Thus as one of a condition of occurrence of a "critical" infrastructure it is necessary to consider occurrence of preconditions for change of qualitative definiteness of the regional SES.

We consider a new class of integrated network design and scheduling problems, with important applications in the restoration of services provided by civil infrastructure systems after an extreme event. Critical services such as power, waste water, and transportation are provided by these infrastructure systems. The restoration of these services is necessary for the society to recover from the extreme event as quickly as possible. The class of integrated network design and scheduling problems considered by this work focuses on a set of selected arcs to install into an existing network (i.e., network design decisions) and then scheduling these arcs on a set of work groups. Unlike previous network design problems, the network must be operating at intermediate points in time so that the scheduling decisions associated with the design decisions have a significant impact on the objective of the problem. The operations of the network at intermediate points in time will be evaluated by determining the amount of satisfied demand in the network. We also discuss exact methods to solve this class of large scale optimization problems by employing decomposition techniques. Our methods are tested on a realistic data set representing the (disrupted) power infrastructure of New Hanover County, NC. These results indicate that our methods are capable of providing better computational performance to decision-makers.

This paper considers the power system restoration planning problem (PSRPP) for disaster recovery, a fundamental problem faced by all populated areas. PSRPPs are complex stochastic optimization problems that combine resource allocation, warehouse location, and vehicle routing considerations. Furthermore, electrical power systems are complex systems whose behavior can only be determined by physics simulations. Moreover, these problems must be solved under tight runtime constraints to be practical in real-world disaster situations. This work is threefold; It formalizes the specification of PSRPPs, introduces a simple optimization-simulat ion hybridization necessary for solving PSRPPs, and presents a complete restoration algorithm that utilizes the strengths of mixed integer programming, constraint programming, and large neighborhood search.

The reliability of electricity supply is in modern societies generally on a very high level. However, there have been power interruptions that have severely impacted on the functioning of a society, and on the safety and well-being of citizens — and the possibility for severe interruptions remains also in future. Responsibility to prepare against power interruptions does not rest in Finland only with network operators, but responsible are also electricity users and public actors like rescue services and municipalities. The topic of this paper is the preparedness of Finnish electricity users against major disturbances in supply of electric power. The paper is based on a questionnaire study directed to 86 Finnish distribution system operators. According to the questionnaire, with a response rate of about 60 per cent, it seems that there would be a need to enhance the preparedness against major disturbances in almost every electricity user type.

Cargo rail systems entail uncertainties and associated risks, particularly with the transport of hazardous materials. The risk associated with their release is assessed through the development of spatial risk profiles which vary based on several analytical and computational steps. The complete assessment of risk involves scenario identification, evaluation of the consequences based on inventories of assets and a hazard assessment to obtain the final risk profile along the rail. Using this framework any given length of railway can be analyzed for risk given the appropriate spatially mapped information on the areas surrounding the rail tracks. The proposed risk quantification and management framework is consistent with current quantitative risk analysis practices in order to enable all-hazard decision making. The methodology is briefly introduced and demonstrated using illustrative examples based on notional information.

Urban distributed lifelines constitute critical infrastructure elements with their continuous interaction guaranteeing the service and well-being of growing urban populations. However, their interdependence and exposure make them inherently fragile to perturbations. This paper proposes a methodology for the probabilistic study of interdependent urban systems response affected by cascading failures induced by natural hazards. This methodology manages interdependence uncertainty by including a probabilistic parameter, I str, to control intersystemic damage propagation. In parallel, the methodology simulates cascading failures using a load nodal betweenness and a local capacity factor α. The combination of I str and α allows the exploration of different conditions of interdependence coupling and local robustness. A test application on two interdependent networks under seismic hazard revealed that interdependence escalation induces fragility amplification with increases on α having limited effect on systemic fragilities. This result questions the effectivity of local retroffiting on the global fragility of interdependent systems.

Socioeconomic development and sustainability highly depend on the construction and operation of infrastructure networks. Therefore, robustness, reliability, and resiliency of infrastructure networks are vital to the economy, security and wellbeing of any country. When exposed to natural or man-made hazards, the estimation of the vulnerability and the extent of damage, potentially caused to infrastructure networks, is a complex and computationally expensive task. This paper presents a model that combines a systems approach with strategies for detecting community structures within networks to make vulnerability estimates. Then, by means of a clustering-based decomposition, a hierarchical representation of the network is derived. This is used to obtain information (i.e., evidence) at different levels of abstraction avoiding the complexity and computational cost of a full evaluation of the network. This simplified model of the network favors the efficient assessment of risk and vulnerability for decision-making regarding, for instance, resource allocation and risk mitigation. In this paper, the conceptual foundations and the practical implementation of the proposed model are presented and discussed. Furthermore, the applicability and the challenges of the model are presented through a practical application.

Success in the design and use of civil engineering systems often rests on the synergy with ideas that initially appear to be only peripherally related to the main topic. As more ideas are incorporated into a system greater synergies may become possible, but they may become at the same time more elusive because of the increasing system complexity and the difficulty in considering so many ideas. The complexity of the system results from both technical and social considerations. Under these circumstances it is common to simplify the problem by the introduction of rules and regulations. The setting in which decisions are made will compel civil engineers to participate in decision making processes that are subjective and uncertain. This is different than the rational decision making model in which they are trained and with which they are most comfortable. This paper explores how system complexity, regulations, and decisions are related. This is done in the context of sustainability, a theme in contemporary discourse with higher levels of uncertainty and complexity. It is of particular interest to determine how large the system boundary should be for purposes of decision-making. It is proposed that the boundary be governed by the principle of decision invariance, i.e., as the system boundary changes, the decision does not change. The principle of surprise inevitability is then offered as a reminder that unlikely but potentially dangerous outcomes should be expected, planned for, and to the extent possible avoided.

A system is vulnerable if any small damage produces consequences which are disproportionately large. The damage may come from unknown sources. Consequently any inherent weaknesses in the form of the system need to be explored. In this paper, we present a systems approach to analyse the vulnerabilities of a system and hence to manage risks. The form of the system is organized into a hierarchical model that can be systematically examined for weak points. The approach can be applied to many networked systems including lifelines. Here it is briefly illustrated through a simple structural system and a road network.

Various structural systems are subjected to the risk of fatigue-induced failures. If such structural systems do not have an adequate level of redundancy, local failures may initiate sequential failures toward system-level failures. In order to analyze the reliability of fatigue-induced sequential failures of such structures, the Branch-and-Bound method employing system reliability Bounds method (termed as B³ method) was recently developed. Using a disjoint cut-set formulation, the B³ method identifies critical sequences of fatigue-induced failures in the decreasing order of their likelihood as it systematically updates both lower and upper bounds on the system failure probability without additional system reliability analyses. The updated bounds provide reasonable criteria for terminating the B³ analysis without missing critical sequences or estimating the system-level risk inaccurately. However, the original B³ method is not readily applicable to continuum structures because of its limitations in describing general stress distributions in limit-state formulations, evaluating stress intensity range based on the crack length, and dealing with slow convergence of the bounds for structures with high redundancy. In this paper, the B³ method is generalized to overcome these limitations, and demonstrated by a numerical example of aircraft longeron structure.

Previous papers and presentations by the authors have proposed that robust simulation should be used to define uncertainties in catastrophe risk analyses for portfolios and/or systems. Robust simulation begins with a "preferred" comprehensive model that turns out to be comprised of non-unique solutions for many technical issues. Uncertainties in this "preferred" model can be estimated either endogenously, that is, through alternative distributions used in the simulation process, or exogenously, through alternative comprehensive model simulations. This paper elucidates how these uncertainties are estimated through the examination on the one hand of available earthquake hazard models and (e.g., GMPE, kinematic) and selected uncertainties (e.g., directivity, focal depth), and on the other hand of available building vulnerability models (statistical, opinion-based, engineering) and selected uncertainties (e.g., structural period, strength, ductility). The goal of this paper is thus to define more clearly what count as "alternative credible models" and how they may be used to estimate uncertainties in the resulting portfolio loss distributions.

A formal probabilistic framework for seismic assessment of a structural system can be built around the expression for the probability of exceeding a limit state capacity, a measure of the reliability of system under seismic excitations. Common probabilistic tools are implemented in order to derive a simplified closed-form expression for the probability of exceeding a limit state capacity. This closed-from expression is particularly useful for seismic assessment and design of structures, taking into account the uncertainty in the generic variables, structural "demand" and "capacity" as well as the uncertainty in seismic excitations. This framework implements non linear dynamic analysis procedures in order to estimate variability in the response of the structure ("demand") to seismic excitations. Alternative methods for designing a program of nonlinear analyses and for applying the results of dynamic analysis, particularly as it relates to displacement-based "demand" and "capacity" estimation, are discussed.

In the paper two possible ways are presented for the improvement of the probabilistic seismic safety analysis of nuclear power plants. The fragilities of structures and equipment used in safety analysis are presented as function of peak ground acceleration, although it is not an adequate damage indicator. In the paper an interpretation is given for cumulative absolute velocity as damage indicator. The possibility for derivation of conditional probability of failure for cumulative absolute velocity is also shown. In the paper an attempt is also made for outlining some new options for nuclear power plant seismic fragility development based on the interval and p-box-concept. These theoretical tools might be used in case of large number and variety of types of components, failure modes. The proposed methods can be adopted for simplification of the screening, which requires enormous experience.

Founded in the middle of the 1^st century A.D., Antiocheia ad Cragum was one of the larger Roman cities of the Mediterranean coast region of modern Turkey. This coastal region of Anatolia was known as Rough Cilicia in antiquity. The ancient city, now in a state of ruin, includes an imperial Temple, which was first identified by archaeologists in the 1960s. In 2004, a new project started, with the goal of studying, excavating, and perhaps partially restoring the Temple to a state of "site museum". Several theories have been postulated regarding the collapse of the original temple. Since the temple is located near the East Anatolian Fault, it is highly probable that a seismic event aided in the collapse. In order to better understand the performance of the temple under seismic loading, virtual and physical models of the temple are being created. This paper provides an overview of the project and details the progress being made in seismic analysis. The first author is the architectural engineering director of this project that is conducted in collaboration with art historians and archaeologists, and under the observations and rules of the Turkish Ministry of Culture.

This paper discusses chemical process models for which the only uncertainties of interest are model parameters. In an earlier paper the authors addressed multicriteria optimization in the presence of model and process uncertainty at the design stage. Specifically the authors discussed extensions of the average criterion method, the worst-case strategy and the ε -constraint method under the following conditions: (a) at the design stage the only information available about the uncertain parameters is that they are enclosed in a known uncertainty region T, and (b) at the operation stage, process data is rich enough to allow the determination of exact values of all the uncertain parameters. The suggested formulation assumed that at the operation stage, certain process variables (called control variables) could be tuned or manipulated in order to offset the effects of uncertainty. This formulation made the conventional assumption that there was only one type of uncertain parameters. In this paper, the authors consider the more realistic case, where the uncertain parameters fall under at least two classes at the operation stage, namely (a) those that can be determined with enough accuracy and (b) those that cannot be determined with such accuracy given the available process data. The case study is an application to a direct methanol fuel cell.

The national infrastructure is the most essential society component serving for civil activities, economic prosperity, and public safety every day. Its absence or failure will result in tremendous devastation to nation's growth and competitiveness. Nevertheless we can't manage and drive the infrastructure to a right direction unless the infrastructure conditions and risks are properly evaluated. And current researches mainly focus on single or multiple projects, seldom study the risk evaluation from the overall national level. This paper first constructs a national-level risk evaluation framework from four levels: ultimate outcome, immediate determinants, underlying determinants, basic determinants; then assesses the infrastructure condition based on ASCE quadrennial infrastructure report card case. Various methodologies, assessment objectives, evaluation breakdown structure, gauging criteria and data sources are discussed though the case. The result of this research can be used for investigating national infrastructure risks systematically, supporting the national infrastructure funding allocation, and assisting to policy decisions making.

Forecasting is a critical function of project control and management. Reliable forecasting enables the project manager to make better informed decisions for timely control actions to prevent or mitigate adverse project outcomes, especially schedule delays and/or cost overruns. Recently, a new probabilistic method for project schedule forecasting was developed based on the Kalman filter method and the earned value method. In this paper, the Kalman filter forecasting method for schedule is extended to formulate a consistent and practical method for project schedule and cost performance forecasting. A numerical example is presented to demonstrate how the new method can be efficiently employed in real projects. Monte Carlo simulation is also conducted to evaluate the accuracy of the proposed method.

The purpose of this research is to establish an appropriate program for the long/middle term planning of public works. This planning needs to satisfy various constraints such as the order relations among construction projects and annual budget. Therefore, in order to solve this planning problem as a combinatorial optimization, it is necessary to search for solutions in very complicated solution space. In this optimization, Genetic Algorithm (GA) is more likely to be trapped by a local optimum due to the lack of diversity among individuals. In this study, an attempt is made to develop a decision support system for the optimal planning by using Improved GA. The proposed method adopts a local search which aims at satisfying constraint of the order relations. This local search can prevent that the search of GA tends to converge to a local optimum. Numerical examples are presented to demonstrate the applicability of the proposed method through the comparison with Simple GA computation.

Effective hazard reduction depends to a large degree on informed public related to risks, risk consequences, and risk management practices. Quantitative risk analysis can estimate probabilities of damages considering different variables as well as the costs associated with the damages. In a community, several engineered systems act together interdependently to provide operating functionality. Interdependence among systems is not well understood and thus meaningful measures of system vulnerability are difficult to derive. To minimize the impact of a damaging natural hazard on a community, actions by several levels of decision-makers are necessary. Effective public policies have to consider not only the quantitative analysis of physical systems but also economic consequences for the society, and socio-political rationality. The focus of this paper is on long —term actions taken by the public officials in minimizing the loss of functionality in a community, by combining quantitative analysis with the qualitative aspects of the decision-making reality in the context of public policy.

Based on the mega public building safety management status in China, the related legal system and government supervision mechanism still need to be improved. However, as many a mega public building with high technology is being built or has been operated in China, safety risk may exist to some some extent. Therefore, this article puts forward the establishment of mega public building safety inspection system. The researches will contain legislation, inspection contents, confirmation of inspection methods or procedures, qualification of inspection organization and inspectors.

With rapid development of Chinese urbanization, the number of city buildings is growing rapidly. The original building safety supervision system: supervision functions are divided into many departments, supervision focused on the spot, could no longer be effective. There exist problems with the deficiency of executive resources, blind management zone and unclear responsibility. By employing third-party assessment agencies as a "chain" of building safety management, the original divided and intermittent administration functions can be integrated into an organic network, so the supervision system could evolve into a long-lasting and adaptive management system. This paper mainly researches the construction approach and the operating mechanism of this new regulatory model, clarifies the responsibilities of the parties involved in this model. It is hoped that this model can provide a feasible solution to the public management reform on building safety supervision in China.

Risks from future sea level rise entail significant uncertainties concerning overall potential impacts, the specific threats faced by particular areas, and what benefit or costs are associated with strategies for addressing such risks. Quantitatively assessing these risks requires the development of spatial risk profiles based on several analytical and computational steps of hazard likelihood assessment, scenario identification, consequence and criticality assessment based on inventories of assets along coastal areas particularly of population centers, vulnerability and inundation assessment, and benefit-cost analysis to manage risks. The proposed risk quantification and management framework is consistent with quantitative risk analysis practices in order to enable decision making. The methodology is briefly introduced and demonstrated using illustrative examples based on notional information.

Tornadoes, hurricanes, droughts, and earthquakes are example of natural events that pose risks to society. The first objective of this paper is to demonstrate why we cannot simply apply to risks associated with natural events insights and frameworks for moral evaluation developed in the literature by considering ordinary risks, technological risks and the risks posed by anthropogenic climate change. The second objective is to develop a framework for the moral evaluation of the source of the risks associated with natural events. Our discussion concentrates on the way the construction and modification of the built and natural environments can alter the probability of occurrence of natural events and the character and magnitude of the impact that such events have. We propose a standard of reasonable care for decisions about the built and modified natural environment, which accounts for technical and resource constraints and the place of natural hazard mitigation in public policy.

Hurricane winds are one of the governing design environmental loads for structures. In coastal regions such as Florida, hurricanes cause enormous loss to life and property. Research focusing on the complex interaction between hurricanes and the built environment is therefore needed for developing a cohesive approach to build hurricane resilient coastal communities. At the International Hurricane Research Center (IHRC), Florida International University (FIU), research is going in stages on the construction of a large state-of-the-art Wall of Wind (WoW) facility for potential full- and large-scale wind engineering testing. In this paper, a technique for simulating hurricane winds at the WoW is presented and investigated. Wind profiles were simulated using turning vanes, and/or adjustable planks mechanism with and without grids. Assessments of flow characteristics were performed in order to enhance the WoW's flow simulation capabilities. The full-scale testing facility will be capable of generating hurricane wind and wind-driven rain field with proper characteristics to allow better understanding of category 1 to 4 hurricane (using Saffir-Simpson scale) effects on structures. The facility will be large enough to engulf full- and large-scale models of single-story buildings built using actual construction materials. This will help improve code provisions, innovative hurricane mitigation development, and producing solutions which bridge the disciplines of wind engineering and structural engineering.

Risk Management is one of the vital factors for success in long term projects. According to some researchers which consider the risks in long term and large structures projects, risk management is the order of the day and one of the critical success factors (CSFs) of PPP-BOT projects. The concept of risk in these projects, where the risks are generally high, involves contingency and uncertainty. So it is essential to employ uncertainty, probabilistic and stochastic methods and to use related models to manage risks in these projects. In this paper some Probabilistic and Stochastic methods, tools and software are used in real cases to show their role and benefits in the life cycle risk management of Long Term Infrastructure Projects; namely PPP-BOT projects and these have been compared with current methods. Finally a framework for these methods has been proposed to examine dependability modeling under uncertainty by using proposed global dynamic environment model.

Is it a good idea to mix data that have been collected with painstaking attention to proper protocols and careful measurement using highly precise instruments with data that have been hastily collected by poorly paid, sloppy, and unmotivated people with bad instruments and shoddy protocols under uncontrolled conditions? Interval statistics is one convenient method that accounts for the different qualities of data in an analysis. High quality data have tighter intervals and poor quality data have wider intervals, and the two can be legitimately pooled using interval statistics. Thus, it is possible for an analyst to combine good data and bad, but is it advisable to do so? We provide illustrations that show that under some circumstances, including more data without regard for its quality decreases the amount of uncertainty in the output of an analysis. Ordinarily, statistical judgment would frown on throwing away any data, but as demonstrated by these examples, it seems clearly advantageous to ignore this judgment under certain situations. More data does not always lead to more power, and increasing the precision of measurements sometimes provides a more efficient return on research effort.

A lot of uncertainties can be encountered in the process of building design, construction and use, and they tend to affect the safety of building in the building occupancy phase. The study was launched to select an appropriate risk assessment method which can be used to make a proper evaluation of uncertainties, and it would be better that the assessment of a building's safety risk is convenient and fast while the staff only use some simple inspection methods. The characteristics of risk factors were analyzed, and several risk assessment methods were compared. Finally, a suitable building safety risk assessment framework was put forward. However, the specific risk assessment approach requires further study, and more in-depth analysis should be made in selection principle of selecting the risk assessment methods.

This paper presents a risk-based analysis method aiming at defining the risk profile associated with an offloading operation. For offloading operations the risk profile is usually evaluated considering that the environmental condition will not suffer considerable changes during offloading that has an approximate duration of 24 hours, varying based on the tanker size. The method follows four basic steps: Accident Modeling, Failure probability assessment with Bayesian techniques, Evaluation of consequences, and Markovian process to aid decision making. The method is applied to evaluate the risk profile of an offloading operation in Campos Basin, Brazil. The method is used to model the risk scenario associated with shuttle tanker main engine failure as initiating event. The changes in environmental conditions have great influence in risk profile and increase the probability of disconnection.

Nowadays, LNG Import Terminals (where the storage and regasification process is conducted) are mostly onshore; the construction of these terminals is costly and many adaptations are necessary to abide by environmental and safety laws. Moreover, an accident in one of these plants might produce considerable impact in neighboring areas and population; this risk may be even worse due to the possibility of terrorist attack. Under this perspective, a discussion is conducted about a vessel known as FSRU (Floating Storage and Regasification Unit), which is a storage and regasification offshore unit, that can work miles away from de coast and, because of this, can be viewed as an option for LNG storage and regasification. The goal is to develop a Preliminary Risk Analysis, which will map potential hazardous events, equipment and operation of critical points at the FSRUs Regasification System, based on the Reliability Theory and the Risk Analysis Theory. This analysis is essential to define a maintenance plan based on the Reliability Centered Maintenance. The results intend to clarify the operational risks of the system and might improve the development of an effective maintenance plan, which can provide good operability with appropriate safety levels.

A probabilistic framework is discussed in this paper for assessment and sensitivity analysis of seismic risk, characterized as the life-cycle repair cost. A comprehensive methodology is initially discussed for earthquake loss estimation that uses the nonlinear time-history response of the structure to estimate the damage in a detailed, component level. A realistic, stochastic ground motion model is then adopted for describing the acceleration time history for future seismic excitations. In this setting, the life-cycle repair cost can be quantified by its expected value over the space of the uncertain parameters for the structural and excitation models. Estimation of this cost through stochastic simulation is suggested and a probabilistic sensitivity analysis approach is also presented, aiming to identify the structural and excitation properties that probabilistically contribute more to this cost.

Results of previous research of shadow economy show powerful impact of crisis tendencies in economy on shadow economic activity. Moreover, during crises a scale of shadow economy grows considerably, therefore an importance of learning this part of economy is increasing. So far as economic crisis cause activation of all forms of shadow activities an estimating quantitative influence of socio-economic consequences of the world economic crisis on a shadow economy is of interest. In order to achieve goal the economy-mathematical model was designed. The economy-mathematical model of a shadow economy in the conditions of the world economic crisis was used to forecast a development of the regional shadow economy under the influence of the world economic crisis. Short-term forecasts of the development of a regional shadow economy were obtained for the Russian regions.

Engineers must deal with risks and uncertainties as a part of their professional work and, in particular, uncertainties are inherent to engineering models. Models play a central role in engineering. Models often represent an abstract and idealized version of the mathematical properties of a target. Using models, engineers can investigate and acquire understanding of how an object or phenomenon will perform under specified conditions. This paper defines the different stages of the modeling process in engineering, classifies the various sources of uncertainty that arise in the each stage, and discusses the categories into which these uncertainties fall. The paper then considers the way uncertainty and modeling are approached in science, in order to highlight the very different criteria appropriate for the development of models and the treatment of the inherent uncertainties in engineering. Finally, the paper puts forward nine guidelines for the treatment of uncertainty in engineering modeling.

In view of all the damage caused by flooding that has affected large numbers of regions throughout the world over the last ten years, urban areas appear to be little prepared for facing up to this type of catastrophe. Today, improving their resilience, i.e. their capacity to recover rapidly after flooding, appears to be a real issue at stake in societies' sustainable development. Due to their organization in the form of subsystems , the multiple aspects of their functions and the dynamics that drive them, these urban territories must be considered as complex systems . Within these urban systems, technical networks are physical links between inhabitants and the actors involved, the symbolic links of belonging to the same community, to the same organized territory (Lacoste quoted by Dupuy, 1991). As supports and even objects of interactions between the different sub-systems in the urban system and with the outside environment, they supply, unify and irrigate all the constituent elements of urban territories. In this way, networks participate in organizing and regulating the system by being the vector of relations between its different constituent elements. "Physically connecting the elements in the system unifies them and creates the network's operating conditions at the same time. In the same way, it makes a certain mode of operation and evolutions in the system possible" (Dupuy, 1984). This strategic position makes networks extremely influent in the dynamics of maintaining the global urban system. In turn, they can be generators of incidents by interrupting flows or vectors in the propagation of unforeseen turns of events. Therefore, characterizing their resilience to flooding may prove to be interesting in for providing a better understanding of urban resilience. In this context, we have decided to work on the resilience of waste management networks. Because they raise essential questions on sanitation and public health and because they often have a strong visual and psychological impact, these networks appear to be real issues at stake in crisis management. After flooding, the volume of waste generated is often significant and of a different sort (mixed, even polluted wet waste). Faced with this situation, waste management poses a real problem. What should be done with this waste? How should it be collected? Where can it be stored? How should it be processed? Who is in charge? Providing answers to these questions is all the more strategic inasmuch as post-flood waste is the visual sign of the catastrophe. As a result, cleaning up is populations' and local actors' first reflex in order to forget what has happened, but also in order to start up again as quickly as possible. Therefore, it would appear primordial to improve waste management networks' resilience to flooding (Beraud et al., 2010). First and foremost, improving the resilience of an organization requires understanding the way it operates in order to identify what dysfunctions it may contain. A methodology needs to be developed for this purpose, capable of analyzing a waste management network's way of operation under normal and crisis conditions. In this way, risks and potential dangers resulting from the urban system being flooded can be identified and the means of prevention for improving the waste management network's resilience to flooding can be brought out. Choosing the right methodology is not an easy task. Numerous methods of risk analysis exist. For the most part they are of industrial origin. As a result, it is not always easy to use them for studying social systems such as urban technical networks, as these systems possess characteristics that differ considerably from industrial systems. "Multiple responsibilities with regard to the design, build, operation and maintenance of networks, separated amongst numerous actors who do not regularly communicate together and share information" are, for example, one of the particularities of risk control in an urban environment (Prost, in Blancher, 1998). As a result, the way in which networks operate may appear to be extremely complex: the diversity and involvement of actors, different scales and territories to be taken into account, the issues at stake concerning the public service mission, catastrophic consequences, that are immediate and those with important repercussions when they are interrupted, etc. This complexity makes it extremely difficult to model urban networks (Maiolini, 1992), whereas applying methods used in the world of industry requires that models are created beforehand. Therefore, there are real methodology stakes in play when transferring these methods from industrial engineering to urban engineering. This article will present the methodology set up followed by initial results.

Chloride-induced corrosion has been identified as one of the main causes of deterioration of concrete structures, such as highway bridges, etc. The development of a performance-based approach is critical to ensure adequate safety, serviceability and durability of concrete structures, built in chloride-laden environments, as well as to help identify appropriate maintenance strategies to extend their service life. This paper presents a polynomial chaos response surface approach for the probabilistic modeling of chloride-induced corrosion of carbon steel reinforcement in concrete structures that takes into account the uncertainties in the parameters that govern the physical models of chloride ingress into concrete and corrosion of carbon steel, including concrete diffusivity, concrete cover depth, surface chloride concentration and threshold chloride level for onset of corrosion. A case study of highway bridge deck was used to illustrate the applicability, accuracy and computational efficiency when compared to crude Monte Carlo simulation.

The general scheme for processing the results of the area risk monitoring, requirements for the data bank structure and functioning are presented. The algorithms of processing statistical and expert information for obtaining quantitative estimates of the area risks and creating thematic maps of different detailization of the considered territories are given.

Accurate and time-efficient methods for computing annual failure probabilities are important for effective implementation of performance-based and reliability-based optimal design. This work provides a general approximate closed-form solution to compute the annual failure probability of components or systems. We show an application of the proposed approach to an example elasto-plastic SDOF system and compare the solution with those obtained numerically and using an alternative approximate closed-form solution currently available in the literature. We find that the proposed approximate closed-form performs better than the one in particular at low ductility levels. The proposed solution also gives insights into the physical phenomena by expressing failure probability in the terms of key parameters.

Civil engineers, in particular, have the primary responsibility for the design and planning of civil structures, including protective systems to minimize losses of lives and economies during extreme hazard events. Structural failures in recent earthquakes and hurricanes have exposed the weakness of current design procedures and shown the need for new concepts and methodologies for building performance evaluation and design. Although the uncertainty of seismic load has been well recognized by the profession, the incorporation of uncertainty in most building code procedures have been limited to the selection of design loads based on return period. To strike a balance between the possible high initial cost and potential large losses over the structure's lifetime, the life-cycle cost and the uncertainty in the hazards and system capacity need to be carefully considered. Moreover, to strictly enforce performance goals, the target probabilities need to be set directly for the limit states rather than for the design earthquake or hurricane etc. Therefore, the total cost over the life time of the structure should be considered when optimizing the design for natural hazard. The future or damage cost of the structure due to natural hazard is calculated based on quantitative risk assessment (QRA). This paper presents a methodology for the design optimization based on the QRA. A case study for a three story residential wood building is presented in order to demonstrate the concept.

By quantifying economic risk due to damage to building stock, regional loss models for natural hazards are critical in the creation of regional policies, including evacuation strategies and zoning. The increasingly complex interaction between natural hazards and human activities requires more and more accurate data to describe the regional exposure to potential loss from physical damage to buildings and infrastructure. While databases contain information on the distribution and features of the building stock, infrastructure, transportation, etc., it is not unusual that portions of the information are missing from the available databases. Missing or low quality data compromise the validity of regional loss projections. Consequently, this paper uses Bayesian Belief Networks and Classification and Regression Trees to populate the missing information inside a database based on the structure of the available data. A case study is presented to evaluate results.

The purpose of this exercise was to applied the risk management concepts for assessing the project's risk profile, quantified the risks in terms of cost, validate the cost estimate and to propose mitigation strategies for managing the risks. The California Department of Transportation (CALTRANS) has developed a Project Risk Management Handbook; the guidelines of this document were followed on this study. The Cabrillo Bridge is located in Balboa Park and was built for the 1915 Panama-California Exposition. Currently the project is in the environmental phase and is set for completion in spring 2010. The project team was faced with a decision to move forward with a replacement or with rehabilitation. It was believed that significant public controversy and opposition would arise with a replacement alternative because the Cabrillo Bridge is an integral part of the City's Balboa Park, a public facility which is listed as a National Historic Landmark. A Risk Management Team was formed and its members represent the project different functional units. The results show the possibility to obtain a cost contingency based on specific risks within the risk register and uncertainty with the items of the project cost estimate, instead of assigning a contingency value without any reference to identified risks or uncertainty conditions. In addition, the results validated the alternative recommended by the value analysis study. This paper demonstrates the viability for performing concurrently risk management and value analysis, having done this; it will provide very valuable feedback to the value analysis study plus will contribute with quantifiable data for sustaining a decision over a selected alternative. In addition, it will formalize the use of risk management with CALTRANS current process for delivering projects.

Both Haiti and Chile were subjected to major earthquakes less than two months apart. The loss of life and devastation in Haiti, however, was much worse, despite the magnitude of the earthquake being smaller. This paper investigates the underlying relationship between the implicit level of risk accepted for natural hazard vulnerability, and the level of economic, social and political development of the country. In particular, it reports on a study of seismic hazard and code development/enforcement for two countries at very different levels of development, illustrated through a case study.

Uncertainty analysis is an area of vital interest to all engineers, including those engaged in policy formation as well as those charged with design or construction of novel solutions. While an array of tools has arisen to help assess and manage uncertainty, many fundamental challenges remain, among them the role of human judgment in risk analysis. Heuristics theory is a theory of human judgment and perception, made famous by 2002 Nobel Prize winner Daniel Kahneman, and his associate Amos Tversky. An understanding of heuristics theory can help us leverage quantitative uncertainty analysis by making the role of human judgments manifest, and by giving us techniques to reduce bias in those judgments. This paper reviews recent research in the area of heuristics and bias and proposes applications to civil engineering domains. Case studies of actual projects and situations are reviewed to highlight the interaction of heuristics with uncertainty modeling and analysis. The paper also highlights the challenges of communicating uncertainties, in particular how to close the gap between perceived risk and quantified risk and how perceived benefits keenly influence perceived risk.

In accordance with the ASCE Standard 7-05, in regions subjected to wind and earthquakes, structures are designed for loads induced by wind and, separately, by earthquakes, and the final design is based on the more demanding of these two loading conditions. Implicit in this approach is the belief that the Standard assures risks of exceedance of the specified limit states that are essentially identical to the risks inherent in the provisions for regions where only wind or earthquakes occur. We show that this belief is, in general, unwarranted, and that ASCE 7 provisions are not risk-consistent, i.e., in regions with significant wind and seismic hazards, risks of exceedance of limit states can be up to twice as high as those for regions where one hazard dominates. This conclusion is valid even if the limit states due to wind and earthquake are defined differently, as is the case in ASCE 7. We also describe an optimization approach to multi-hazard design that achieves the greatest possible economy while satisfying specified safety-related and other constraints. We present two applications to illustrate our approaches.

The safety and mobility of the population is dependent on the structural integrity of existing infrastructures. This requires a continuous monitoring program for condition assessment of infrastructures to guarantee their safe and reliable operation. A comprehensive structural health monitoring system involves a combination of advanced structural analyses integrated with field investigation using sensors and performance data acquisition. In this work a new engineering decision-making method for condition assessment of infrastructures is developed that is based on a hybrid experimental structural health monitoring (performance of nondestructive tests for structure's response) and theoretical structural uncertainty analyses (interval finite element analysis of structures) procedure. In many cases, the application of results obtained from the health monitoring program of a structure is limited. Furthermore, the conventional deterministic structural analyses are incapable of considering uncertainties in experiments, system properties and loads, thereby resulting in error when estimating the state of health of a structure. This work proposes a new methodology that utilizes the results from structural health monitoring applications, combined with uncertainty analyses, in an effort to offer a more robust process for estimating the condition, energy conservation, and sustainable design of a given structure. On a broad sense, the results of this study will be applicable to the decision making process in determining a structure's demand for retrofit or reconstruction. The developed methodology along with an illustrative example is presented.

A novel nondestructive Structural Health Assessment (SHA) technique, known as the Generalized Iterative Least-Squares Extended Kalman Filter with Unknown Input (GILS-EKF-UI) method, is being developed at the University of Arizona. The procedure can detect defects in new, deteriorated or rehabilitated existing structures or just after large natural or manmade events. Most SI-based SHA approaches use excitation and response information to identify a structure. Excitation information is not available in most cases. It could be noise-contaminated and the SI concept may not be applicable. For large complicated real structures, it may not be possible to measure responses at all dynamic degrees of freedom and they always contain noise. Addressing all the issues, the GILS-EKF-UI method is being developed and is presented here.

Cluster detection is essential for recognizing design flaws and cryptic common-mode or common-cause dependencies among events such as component failures, but when such events are rare, the uncertainty inherent in sparse datasets makes statistical analysis challenging. Traditional statistical tests for detecting clustering assume asymptotically large sample sizes and are therefore not applicable when data are sparse—as they generally are for rare events. We describe several new statistical tests that can be used to detect clustering of rare events in ordered cells. The new tests employ exact methods based on combinatorial formulations so that they yield exact p-values and cannot violate their nominal Type I error rates like the traditional tests do. As a result, the new tests are reliable whatever the size of the data set, and are especially useful when data sets are extremely small. We characterize the relative statistical power of the new tests under different kinds of clustering mechanisms and data set configurations.

The ground movements induced by the construction of supported excavation systems are generally predicted in the design stage by empirical/semi-empirical methods. However, these methods cannot account for the site-specific conditions and for information that become available as an excavation proceeds. A Bayesian updating methodology is proposed to update the predictions of ground movements in the later stages of excavation based on recorded deformation measurements. As an application, the proposed framework is used to predict the three-dimensional deformation shapes at four incremental excavation stages of an actual supported excavation project.

Bayesian networks (BNs) have characteristics and capabilities that make them useful for assessing and responding to risk due to natural and man-made hazards posed to infrastructure systems. In this paper, we emphasize the use of the max-propagation algorithm of the BN for identifying critical components in a system subject to an arbitrary hazard. The paper begins with a brief introduction to BNs. This is followed by brief descriptions of inference and max-propagation in a BN, and use of the latter for identifying critical components. Limitations of the methodology are discussed. Two simple examples demonstrate the use of the proposed algorithm.

When a natural or man-made hazard occurs, it is essential to detect damaged components in lifeline networks to enable rapid recovery of the utility service in the impacted areas. However, inspections of individual network components such as buried pipes are often impractical due to exceedingly large costs and time. This paper presents a new system reliability method using a Bayesian method developed for identifying network components with higher conditional probabilities of damage given post-disaster network flow monitoring data. This method achieves an optimal matrix-based representation of the problem for efficient damage detection. The developed method is demonstrated by a water pipeline network consisting of 15 pipelines. The conditional probabilities of damage in 15 pipelines given post-disaster network flow observations are obtained by a Bayesian method for damage detection purpose. The results of the post-disaster damage detection by the proposed system reliability method are compared to those by Monte Carlo simulations and by the matrix-based system reliability method without selective expansion scheme in order to demonstrate the accuracy and efficiency.

In this paper the problem of assessing the condition and making decisions for strategic structures, e.g. hospitals, police and fire stations, and highway/railway bridges, immediately after an earthquake is investigated. Nowadays it is possible to monitor such structures with placement of sensors, such as accelerometers, strain gages and seismographs. Due to their low cost and easy operation, it is possible to place a large number of such sensors in various parts of a structure or its surrounding ground. The question then arises as to how to post-process the large amount of data obtained from such sensors and from seismological sources that become available shortly after the earthquake. A Bayesian network (BN) framework for processing such evolving information is proposed herein. This probabilistic information, together with information on the costs associated with inspection and closure of the structure, is then used to formulate a decision problem regarding the optimal action to take in the aftermath of an earthquake. An example demonstrates the methodology for a typical building structure.

The phrase "stream restoration" has become commonplace to describe a wide variety of stream intervention practices. Rather than using the term "restoration" to cover the broad range of stream interventions, many of which are not restoration at all, the Center for Watershed Protection has moved to using the phrase "stream repair". In this paper, a risk-based framework is presented for planning and designing stream repair projects, particularly in the vicinity of infrastructure. The method incorporates a wide variety of uncertainties, including non-quantifiable types, such as aesthetics, failure consequences, etc., providing a systematic approach to reducing risk prior to implementation. Individual components of a stream repair project, as well as the overall design, are considered. The results provide a basis for decision-making in determining the best option for stream repair and provides a powerful tool for communication between practitioners engaged in design, environmental funding and permitting agencies, infrastructure owners, and other stream stake-holders.

Climate change is without a doubt one of the greatest sources of uncertainty today. Even among the majority of scientists who agree that it is happening and will continue to happen, there is a tremendous range of projections as to the specific impacts, their timing and their geographic extent. There's not even agreement over the magnitude of uncertainty surrounding the various projections. All of this uncertainty poses a huge challenge for many organizations, but for none more so than water supply agencies, since climate change may directly impact both water supply and water demand. East Bay Municipal Utility District (EBMUD) is a major water utility in California that recently considered climate change along with other future unknowns when it updated its water supply program through 2040. This paper discusses the process by which EBMUD considered the massive uncertainty posed by climate change in the context of other unknowns to formulate a flexible, defensible water supply strategy that would be robust in the face of the continually evolving scientific data over the next two decades. During this process, EBMUD considered not only the range of scientific projections, but also their possible reliability; and considered not only what we might do if various projections are correct, but what we might do if they are low or high. The discussion of the process and final outcomes should be of interest to all practitioners charged with helping society manage the tremendous uncertainties associated with climate change.

Applying surge response functions (SRFs) in the estimation of peak hurricane surge is valuable to coastal management and safe-evacuation planning. These SRFs make use of the meteorological characteristics for expected storms as input, and were developed by Irish et al. (2009) using generalized dimensionless scaling laws and optimally selected sets of hydrodynamic hurricane simulations for the open coast and within more complex regions like coastal bays. With improvements to the existing form of the SRFs, reliable extreme-value hurricane flooding estimates can be obtained. Hurricane forward speed and approach angle are important meteorological parameters that can induce variations in surge estimates. Recent studies suggest that in the future sea level rise (SLR) may accelerate and major hurricanes may intensify. Here we present a methodology applied to modify the scaling laws to incorporate the effects of forward speed; we also introduce considerations being made towards developing scaling laws for approach angle and sea level rise effects.

The North American electric power grid is considered to be the largest and most complex technical system in the world. Reliability of this heterogeneous and highly interconnected systems is paramount since our national security, digital economy, and transportation and water systems require robust operation of the electric power system. The focus of this paper is on comparing top-down statistical approaches with bottom-up engineering models that are used in estimating the reliability of power systems during hurricanes and high wind events. It then gives a synthesized overview of the advantages and disadvantages of each approach and highlights areas in which additional research is needed.

This article presents a survey which was carried out to identify the most prevalent residential building types, their characteristics and their distribution in the state of Florida. The databases provided by county property tax appraisers were used for the survey. Detailed statistics on different building components are presented along with an analysis of their correlation to the year built of the structures. The results of the survey provided a basis for the generation of different vulnerability matrices in different regions of Florida. The survey, the resulting statistical analysis, and the weighting process of the vulnerability matrices are discussed.

An efficient, computational framework is discussed for hurricane risk estimation with emphasis on online applications. Basis of this framework is the characterization of each hurricane scenario through five model parameters: the location and angle of landfall, the central pressure, the forward speed, and the radius of maximum winds. Description of the uncertainty in these parameters, through appropriate probability models, leads then to quantification of hurricane risk as a stochastic integral. For efficiently evaluating this integral, application of response surface methodologies as a surrogate modeling approach is discussed. An illustrative example is presented that considers applications of these tools for hurricane risk estimation for the Oahu Island in Hawaii.

Regional wind loss predictions depend on vulnerability curves. A state of the art approach for developing the vulnerability curves is presented in the paper. It is based on engineering models that estimate the building damage caused by wind pressures, debris impact, and water penetration. This approach is a substantial improvement over traditional approaches, which derive vulnerability curves for different kind of buildings through curve fitting of historical insurance loss data. This paper describes the engineering model used to develop vulnerability curves for commercial-residential buildings in the Florida Public Hurricane Loss Model.

This paper presents an overview of the Database-Assisted Design (DAD) methodology, and describes its motivation, the technological developments that enabled its development, and its main features. DAD is an integrated methodology that does not just calculate wind loadings, but rather performs the automatic calculation of internal forces and demand-to-capacity indexes, and the detailed verification of the adequacy of the structural design. The paper discusses interpolation issues, wind directionality effects, and the estimation of wind effects with specified mean recurrence intervals. It concludes by noting the potential of using DAD in conjunction with Computational Fluid Dynamics methods.

A probabilistic model is developed to predict the deformation demand on wind turbine support structures due to the operation of the turbine, and wind and wave loading. An existing deterministic model is corrected by adding a correction term to capture the inherent bias and model error arising from an inaccurate model form or missing variables. The correction term and the model error are assessed using data obtained from detailed three dimensional nonlinear finite element analyses of a set of wind turbine systems considering different design parameters. Fragility of the support structure is then defined as a conditional probability of not meeting a specified deformation performance level. Existing simplified methods for the analysis of the support structure and foundation of wind turbines have limitations in the modeling of the nonlinear behavior of the foundations. Three dimensional nonlinear finite element analyses provide a more rigorous and accurate modeling of the soil mass, pile and their interaction, but they are computationally expensive and time consuming. The proposed probabilistic demand model provides an accurate framework for predicting the deformation of support structure properly accounting for the underlying uncertainties, and for estimating the vulnerability of the wind turbine support structure without the need of conducting complicated nonlinear finite element analyses.

Structural response due to directional wind must be taken into account rigorously and transparently in wind resistant design, particularly for tall buildings. A database-assisted design (DAD) technique makes it possible to account for directional effects on the structural design of tall buildings. To this end, DAD employs a climatological database consisting of the directional wind speeds at a meteorological station relevant to the building site, an aerodynamic database of the pressure coefficient time histories at large numbers of taps on the exterior building surface, and a micrometeorological data set consisting of the ratio of wind speeds at the standard elevation at that site to the mean hourly wind speeds at the top of the building. This study describes the application of the DAD methodology to the design of a 60-story reinforced concrete high-rise structure using the aerodynamic and wind climatological databases. The DAD procedure provides wind-induced responses with any mean recurrence interval, including demand-to-capacity indexes, inter-story drifts, and top-floor accelerations, and compares their compliance with design criteria.

For structures sensitive to wind directionality, methods for the estimation of wind effects require the use of time series of directional wind speeds with length exceeding the length of the MRIs of interest. This study proposes a procedure for generating such time series from relatively short synoptic wind data sets. The focus in this paper is on the estimation of the parameters of a probabilistic model of the wind speeds and on errors in the estimation. The wind speed data being generated can be used in the Database-Assisted Design approach to account for directional wind effects.

Northern Utah has a long history of landslides occurring on natural and man-made slopes. Earthquake activity in this region may exacerbate the incidence of landslides, thus increasing the potential of landslide related damage to residential structures and transportation corridors. In this context, the present study addresses the earthquake response in drained conditions of a typical, shallow landslide in northern Utah occurring in completely decomposed Norwood Tuff. The slide mass geometry was obtained using a 2-D seismic refraction profile and previously collected geotechnical borehole data. The Newmark sliding block analysis was employed with a translational failure mechanism to determine the permanent dynamic slope displacements under various input accelerograms. The yield coefficient was obtained from pseudo-static limit-equilibrium slope stability analyses using the Slope/W module of the GeoStudio engineering software package. Based on the computational results, a methodology to evaluate the peak ground acceleration threshold that would distinguish between insignificant ground movements and large, potentially damaging slope displacements during an earthquake was developed.

Ground motion attenuation relationships have traditionally been used to describe the spatial distribution of earthquake ground motions. Attenuation models provide the ground motion intensities at different locations of the distribution region by combining two key estimates: (i) median values of ground motion intensities and (ii) aleatory uncertainty parameters namely inter- and intra-event uncertainty. The present study simulates inter- and intra-event uncertainties as Gaussian random variable and two-dimensional stationary Gaussian random field, respectively, assuming they are independent events. Earthquake records from Next Generation Attenuation or, NGA database are used for this purpose. Intensity of ground motions are measured in terms of peak ground acceleration (PGA). To use in the simulation process, nonlinear site response model is represented with a suitable probability distribution. Result shows that intra-event uncertainty is more significant than inter-event uncertainty.

The purpose of this study is to understand the mechanism of the effective shear strength for a soil mass in the presence of spatial variability. Random field finite element analyses are used to simulate spatial average shear strengths and effective (or overall) shear strengths that govern the failure of a soil mass. Based on the simulation results, it is found that the statistics of spatial average strengths can be estimated by Vanmarcke's theory as to be expected. However, the effective shear strength is found to be close to the average shear strength along the actual slip curve, rather than the spatial average shear strength over the entire soil mass.

The expected destructive earthquake of Istanbul in the near future urges the researches to determine the proper and effective measures in order to make sure that the Bosphorus related traffic is affected at the possible lowest level. By considering the fact that the traffic on this bridge corresponds to the busiest in Turkey, especially in the peak hours, the amount of lives to be lost when the earthquake hits the bridge must be kept at the minimum levels according to the different scenarios. In this research a probabilistic approach is employed to develop a risk analysis model by manipulating the speeds of the vehicles and the length of the danger.

This paper presents a probabilistic approach to evaluating the geotechnical stability problem by incorporating the stochastic spatial variability of soil property within the numerical limit analyses. The undrained shear strength and unit weight of soil are treated as a random field which is characterized by a log-normal distribution and a spatial correlation length. The current calculations use a Cholesky Decomposition technique to incorporate these random properties in numerical limit analyses. The Random Field Numerical Limit Analyses are applied to evaluate effects of spatial variability of soil property on the slope stability and the failure mechanism. Monte Carlo simulations are then used to interpret the failure probability of slope for selected ranges of the coefficient of variation in soil property and the ratio of correlation length to slope height. Finally, the conventional safety factor of slope stability is evaluated to obtain an objective probability of slope failure.

Deep excavation is a complex engineering system. It is necessary to research the system reliability of deep excavation in order to reduce the number of accidents. Considering the interval features of mechanic parameters of retaining system and soils of deep excavation, the interval analysis method was applied to establish two performance functions against overturning and sliding for the gravity retaining system of deep excavation. In terms of the limitation and shortage of the interval calculating rules, the affine arithmetic approach was adopted to compute corresponding reliability index. Then, the interval reliability of some typical systems, such as series system, parallel system, series-to-parallel system, etc, was studied based on the extension principle and the system analysis. The analysis process was given by an engineering example.

A number of ground surface subsidence occurred within few months in a residential suburb in Kuwait due to sinkholes. Comprehensive studies were carried out to identify the reasons of the subsidence and underground deep cavities were detected. The cause of the sinkholes was attributed to the dissolution of the limestone bedrock and the subsequent raveling of the overburden soil cover. A pilot area was selected and treated by filling the underground deep cavities with cementatious grout materials. This paper develops a basis of a methodology for quantifying the risk of ground surface subsidence due to underground cavities for the above case. The hazard and risk profile development is limited to computing the probability of sinkhole formation over the selected pilot area based on field data and examining spatial characteristics of the cavities. The failure probability is based on a single limit state and failure mode. The spatial analysis of cavities demonstrated an analytical process and confirmed the random layout of cavities without any spatial characteristics at about the same depth. A regression model for predicting grout volume as a function of cavity height was developed.

The paper describes the methodology that was used to construct the dynamic risk map for a virtual large ‘Russian Gotham' city, which here and thereafter will be named as RG-City. The risk map is the main part of the safety passport (SP) for the RG-City. The SP is a document required by the Russian EMERCOM and the Russian State Agency for Industrial Safety. The SP contains extensive risk analysis of all potential dangerous objects that are located within the boundaries of the city. It analyzes the natural and technological threats/disasters that may occur in the city. While the city is virtual, the components of the methodology and examples are real.

Quick urban development coupled with technical failures and climate change have increased flood risk and corresponding challenges to urban flood risk management. In this context, there is a need to understand how networked systems are resilient because societal functions are highly dependent on networked systems and the operability of these systems can be vulnerable to disasters. That is why we develop a methodology for producing interdependent networks disturbance scenarios. In order to take advantage of this methodology software is needed.

The demands of deployments and home station security have severely overtasked Air Force Security Forces. Previous standards-based security doctrine was not flexible enough to allow commanders to manage limited resources to achieve acceptable, balanced risk tailored to the mission and priorities. Security Forces needed to change the paradigm of how they do business, to truly transform. USAF Security Forces are now moving to a risk-based approach to installation security, and are using a standardized model called ForcePRO to identify risks and develop risk management strategies. Developed by Air Force Research Laboratory (AFRL) contractors at Tyndall AFB, Florida, the ForcePRO model and software application were field tested at over 50 Air Force installations in the US, Europe and Southwest Asia. ForcePRO facilitates a thorough, repeatable, and understandable physical security analysis that documents the assets, threats, vulnerabilities and risks facing an installation. It aids in developing risk mitigation strategies to reduce risks exceeding the installation commander's risk tolerance.

Decision-makers want to perform risk-based cost-benefit prioritization of security investments. However, strong nonlinearities in the most common physical security performance metric make it difficult to use for cost-benefit analysis. This paper extends the definition of risk for security applications and embodies this definition in a new but related security risk metric based on the degree of difficulty an adversary will encounter to successfully execute the most advantageous attack scenario. This metric is compatible with traditional cost-benefit optimization algorithms, and can lead to an objective risk-based cost-benefit method for security investment option prioritization. It also enables decision-makers to more effectively communicate the justification for their investment decisions with stakeholders and funding authorities.

The EU is very concerned by flood risk. The effects of floods on urban areas are substantial in terms of damage and economic loss. It is expected, taking into account climate change and urban growth, that this particular risk will be more frequent and will cause more and more damage; the built environment has become very vulnerable to floods. There is a need to develop methods and tools to assess urban vulnerability to floods. Indeed, some solutions are already available to assess vulnerability of buildings or infrastructure such as networks. But, the city is a complex system and not just a sum of its components. Our purpose is about developing techniques to assess the vulnerability of the city using a systematic approach and integrating existing tools which assess the vulnerability of particular components of the city system.

Civil engineering systems are vulnerable to natural and man-made threats. In recent years, agencies are paying increased attention to investments that are geared towards reducing system vulnerability. For purposes of analyzing the effectiveness of such investments and also for vulnerability monitoring, a procedure is needed to quantify the level of system vulnerability at any time. This paper identifies vulnerability types and presents a general procedure for establishing system vulnerability. The procedure quantitatively assesses the threat likelihood and consequence. Threat likelihood is expressed as a function of the system environment (threat type); and consequence, is expressed as a function of the system attributes. Specifically, the consequence of failure due to the threat is evaluated on the basis of possible failure modes and the extent of public exposure in the event of failure. Public exposure, a function of the occupancy (or traffic) and the class of the system, reflects the effect of system failure on users and the surrounding community.

Verbally, resilience may be defined as a multi-attribute measure that describes the ability of a system of interdependent CIs to withstand a disaster shock, and its ability to recover within an acceptable envelope of time and cost. As of now, there is no quantitative definition of resilience and strategic preparedness. In this paper an attempt is made to give a generalized quantitative definition of resilience and preparedness as functions of time, taking into account that most of the multiple parameters on which resilience/preparedness is dependent, are random variables or random functions. The overall flow chart for assessing the dynamic integral territorial risk as defined by the "from bottom up" approach is given. The paper describes a new concept in territorial risk analysis and management based on the entropy principle and the territorial life quality index (TLQ1). The entropy principle is used to convolute the risk analysis problems, which may contain tens of thousands of variables, into a problem with only one variable - information entropy. The TLQI, which is composed by the regional GDP, regional life expectancy at birth, and the time that a statistically average person living in this region spends during his/her lifetime to provide for her/his well being, is used to account for the human factor in territorial risk analysis.

Predictions for changes in future sea level rise vary on spatial scales from global perspectives to a few miles of coastline and also from temporal scales of a few decades to almost a century. This wide array of forecasts is challenging enough; however, when confronted by the various ways the predictions are derived (linear extrapolations of individual tide gauges to global numerical models), attempts to navigate through to what is appropriate for risk assessment purposes can be confusing. We will present how concepts of spatial and temporal variability in sea level forecasts, data detrending and assimilation, and other aspects can be understood within the framework of vulnerability and risk. We also discuss how tide gauges can provide a baseline trend for forecasting future sea level change, and how these data can provide insights into intra-annual to decadal oscillations that often are of greater magnitude than the global trends predicted.

The Homeland Infrastructure Threat and Risk Analysis Center (HITRAC) of the U.S. Department of Homeland Security (DHS) provides threat, vulnerability, and consequence assessments of terrorist attacks on infrastructure in support of security risk assessments. This paper will describe the components of vulnerability that HITRAC considers and how it integrates those components into an overall vulnerability assessment. The assessment relies on an approach to target selection and preliminary vulnerability identification drawn from observed terrorist methods of operation. The attack methods used as the basis for scenario generation are taken from the 2008 Joint Special Assessment on Potential Terrorist Attack Methods. These are filtered to identify the attack methods most applicable to the facility, asset, or system being assessed. The vulnerability assessment can be used as a stand-alone analysis or feed into a risk model for national or regional assessments.

More and more floods occurred over the last decade in the world, causing important damages. Moreover, levees are often not well maintained, so they hardly resist to major floods and can break easily. At French national scale, the length of levees, estimated to 7500 kilometers, and the lack of data all along these infrastructures complicates their management. In this frame, levee managers need approaches and tools in order to be helped in their maintenance decision. The goal of our research is to develop methods modeling levee failure mechanisms and allowing performance levee assessment. These methods integrated in an existing GIS dedicated to levee management will contribute to obtain a spatial decision support system aiding levee managers in their maintenance decision. Finally we made some tests in a specific area to integrate in the SDSS the level of vulnerability behind dikes to improve the levee manager decision process.

This paper proposes and demonstrates a new interactive framework for sensitivity-informed de Novo planning to confront the severe uncertainties and risks within water management problems. The framework couples global sensitivity analysis using Sobol' variance decomposition with multiobjective evolutionary algorithms (MOEAs) to generate planning alternatives and test their robustness to new modeling assumptions and scenarios. We explore these issues within the context of a risk-based water supply management problem, where a city seeks the most efficient use of a water market. The case study examines a single city's water supply in the Lower Rio Grande Valley (LRGV) in Texas, using both a 10-year planning horizon and an extreme single-year drought scenario.

Nowadays, in the areas that are encountered with drinking water shortage, contiguity of sea with land create a facility to change salty water to drinking water using desalination plant and reverse osmosis method. The first desalination project of Iran has been completed in south of Iran in Asalouye. The project contract was registered between Noor Vijeh Company (special joint stock) and Pars Special Economic Energy Zone (PSEEZ) dependent to Oil Ministry of Iran. By this contract, Noor Vijeh Company must deliver 10000 m³/day of drinking water to PSEEZ from the lines of water delivery pipe. Type of the project contract is Build, Operate and Transfer (BOT) scheme within 21 years concession period. Delivered water price is 0.48 euro/m³ in 7 years duration from the beginning of the project and 0.375 euro/m³ for next 13 years. In this paper, the risks of the project are studied and scrutiny and risk management are performed by offering a matrix model.

To provide forecasting of erosion and debris flow processes for the energy supply and transport corridors of Georgia using the theory of reliability and risk, we have developeda methodology on basis of the theoretical analysis and long-term field surveys (1980–2010) regarding the intensity of rains and solar radiation. At the following stage this methodology will allow us to design new safe measures against erosion and debris flow processes. To provide study of development dynamics concerning the geometrical dimensions of ditches and gullies — characterizing values of erosion processes in the power supply corridors of oil pipelines (Baku-Tbilisi-Supsa, Baku-Tbilisi-Erzurum) and gas pipelines (Baku-Tbilisi-Shakdeniz), we have determined density distribution law on the basis of the intensity of rains and slope gradients for the case of separate and simultaneous events. For the forecasting of erosion and debris flow processes in the transport corridors (Transport Corridor Europe-Caucasus —Asia (TRACECA) and Georgian Military Road), we have studied the types of density distribution function for the erosive and accumulating areas of the slopes and debris-flow riverbeds, which will provide safe operation of the energy supply and transport corridors.

The concept of probabilistic dominance is developed for making multi-objective decisions where there is uncertainty regarding the performance of alternatives. This approach combines traditional the ideas of Pareto optimality and stochastic dominance for comparison of alternatives with both multiple criteria and uncertainty. A simple application is made for the strategic decision of an approach for continuing or ending water exports from the Sacramento-San Joaquin Delta in California.

A typical problem facing engineers when designing infrastructure to serve future demands is that those future demands are unknown, or at best uncertain. With respect to water supply systems, not only are future demands uncertain, so are the supplies that will be available to meet those demands. Traditionally engineers have based their design capacity estimates based in part on the statistical characteristics of past hydrological records. Today we recognize we can no longer make such assumptions. The past does not necessarily reflect what will happen in the future given the impacts of changes in climate and land cover and use. These changes together with changes in human populations, their age and spatial distributions, their standards of living, and their activities alter the demand for water along with the spatial and temporal distributions and amounts of runoff. Hence we face both demand and supply uncertainty where even the probability distributions of those possible values are unknown. Whatever those distributions will be, increased variability of supply is almost certain, and this in turn strengthens the argument for increased infrastructure capacities in advance of when they are needed. Some of this infrastructure, such as reservoirs and major water supply diversions, will be designed to last for the next 100 years or longer. The capacities of such projects are not easily expanded or reduced as the changes in stochastic nature of demand requirements and available supplies become more certain. One cannot monitor and then alter capacities accordingly in some adaptive framework as might be possible with respect to operating policies. How do we decide what capacities to build now that will affect our ability to serve those living in the next 100 years or so? Do we just apply engineering ‘safety factors'? This paper attempts to address this question.

Performance analysis of a water distribution systems involves hydraulic analysis to ensure that designed nodal demand in each node is maintained at a desired level of pressure and velocity over a design period. Those calculated performance values are mostly used as important surrogates while optimizing a systems design. However, input parameters used for the performance analysis are uncertain and both quantitative and semi-quantitative types. This paper presents a hybrid approach of uncertainty analysis to capture the both types of uncertainty associated in a water distribution systems. The uncertainty associated with nodal demands has been described by probability distribution functions and pipes roughness coefficient by fuzzy membership functions. The technique has been illustrated by a simple water distribution networks. The results from the hybrid approaches are consistent with the deterministic and stochastic approach.

This paper presents a new hybrid hierarchical integrated risk assessment framework (H-HIRA) developed for analyzing integrated risks and uncertainties in urban water systems due to future change pressures, such as climate change, population growth, urbanization and deterioration of infrastructure systems. It is based on several key concepts that include application of a system of systems approach to decompose a complex system into a hierarchical order; DPSIR framework to analyze the likely impacts of future change pressures; a hybrid approach for description and propagation of uncertainties; integrated system modeling to analyze the risk and vulnerability in a system; and a fuzzy set based multi-criteria analysis to select a robust risk management strategy. The framework can be applied for analyzing risks in any complex infrastructure system, and most importantly for risk informed decision making during the strategic planning of infrastructure systems.

Careful study of the patterns and extent of structural damage in the vicinity of a suspected blast event can be useful for identifying characteristics about the nature of the blast insult such as charge weight and directional properties. However, the lack of focused research in this area, as well as the lack of evidence collections standards, prevents post blast investigators from fully utilizing this data. This paper will discuss some preliminary steps taken toward developing guidance for documenting structural damage at a blast scene. A variety of structural materials and configurations are considered in terms of the ways they can fail when subjected to a blast insult and the existing literature is leveraged to develop guidance on how to identify and measure the corresponding structural damage. Once collected, such information can be used to back-calculate the properties of the initiating blast event.

Expected building damage is proposed as a measure of building performance against structural collapse. The proposed novel concept is illustrated with a case study of a typical steel framed building subjected to a bomb explosion and the resulting column removal. Stratified approach to systematic sampling was used to assign appropriate weights to sampled damage scenarios. Finally, an overall expected building damage resulting from randomly located explosions was analytically derived. The analytical expected damage gives a mean building failure as a function of the explosion reach. The presented expected damage is a scalar performance measure and thus it lends itself to a comparison of alternative designs. Expected damage function is a collapse signature of a given building that takes into account copiousness of explosion locations and feasible detonation magnitudes.

In this work the probabilistically-robust design of supplemental dampers for multi-span bridge systems is investigated. The adopted bridge model explicitly addresses nonlinear characteristics of the isolators and the dampers, the dynamic behavior of the abutments and pounding effects. A probabilistic framework is used to address the various sources of structural and excitation uncertainties and characterize seismic risk. Stochastic simulation is used to evaluate this risk and perform the associated design optimization for the dampers.

In present paper, the capability of current nonlinear static pushover methods on estimating the torsion effects of existing buildings has been studied. One real high-rise building with dominant torsion mode has been performed pushover analysis using two classic nonlinear static pushover procedures. Time history analysis has also been performed using tremor monitoring ground acceleration records. According to the comparison between the results of pushover analysis and time history analysis, the efficiency of MPA procedures and N2 method on estimating torsion effect for asymmetric high-rise buildings have been evaluated.

The study evaluates the combined effect of earthquake and flood-induced scour on the performance of bridges located in regions having moderate to high seismic and flood hazards. For the analysis California is chosen as the bridge site where the probabilities of having these two natural hazards are reasonably high. Two example reinforced concrete (RC) bridges are considered. A 100-year flood event with discharge rate of 158.6 m³/s, velocity of 0.8 m/s and upstream depth of 11.9 m is chosen to represent regional flood hazard. Finite element analyses are preformed to evaluate performance of these bridges under regional seismic events in presence of scour resulted from the scenario flood event. Result shows that in presence of flood-induced scour bridges become more vulnerable under seismic excitations.

The present paper collects a series of tools, methods, and convenient solutions used to model and handle efficiently the huge amount of random variables associated with techniques for the life-cycle simulation and maintenance optimization of bridge networks. The randomness in the deteriorating reliability of individual bridges is modeled by means of time-dependent models with random parameters. The uncertain effects of the preventive maintenance actions are represented by numerical superposition of four contributions and characterized by up to six random parameters. The complete rehabilitation of a bridge is supposed not to restore perfectly the original reliability level, and also in this case the extent of the imperfection (i.e. difference between original reliability and reliability after the intervention) is treated as a random parameter. Finally, the in/out of service state of the individual bridges along the life cycle of the bridge network is modeled by means of either a multivariate (i.e. correlated) Bernoulli random variable or a random field.

Model updating is useful for improving structural performance assessments. This paper examines an important assumption of traditional model-updating approaches. This assumption requires the error independence between points where predictions and measurements are compared. Simulations performed on a full-scale bridge show that uncertainties are correlated for both static and dynamic predictions. Therefore, traditional model-updating techniques are not appropriate in such situations. Model updating limitations related to randomness and independence of uncertainties may be overcome by an interpretation strategy called Candidate Model Search for System Identification (CMS4SI). Instead of judging a model by its ability to fit measured data, the approach falsifies models using threshold values that are based upon uncertainties. Uncertainties may be correlated, systematic, independent or random.

A mooring dolphin structures (MDS), consisting of steel-pipe-piles and soil around them, are important elements of marine structures. The reliability analysis of such complicated systems is very challenging. The limit state functions for them are implicit in nature. Reliability evaluation using the classical Monte Carlo simulation technique may not be practical since each deterministic evaluation will require several hours of computer time. A new hybrid method is presented in this paper. The system is represented by finite elements and its reliability is estimated by intelligently integrating the stochastic finite element, response surface methods, and advanced factorial schemes. The procedure is verified with the help of an example.

Seismic fragility reflects the ability of a structure to withstand future seismic demands. To obtain an accurate assessment of seismic fragility, it is critical to incorporate information about the current structural properties. This paper describes a probabilistic framework to incorporate information from nondestructive testing (NDT) in the estimates of the seismic fragility of a reinforced concrete (RC) bridge. The proposed framework combines global and local damage detection methods to incorporate information from nondestructive testing (NDT) on the properties of an existing bridge. As an illustration, the proposed probabilistic framework is used to assess the seismic fragility of an example reinforced concrete bridge.

The concept of containment is often employed at existing and planned landfills, at sites contaminated with dense nonaqueous phase liquids (DNAPLs) and at sites where a physical barrier is required. A vertical cutoff wall is one of the critical elements of the containment system. This physical barrier can provide hydraulic control and prevent migration of contaminants from the area of impacts. Several types of vertical cutoff walls are currently available and include steel or plastic sheet pile walls, geomembrane walls, slurry walls, deep soil mixing type walls or grouted walls. The other components of the containment system include surface water control and capping or liners. Ideally, a properly designed, constructed and operated containment system would create a totally impervious site encapsulation. Realistically, since the design, construction and operation are performed under conditions of uncertainty, containment system performance should be measured in terms of probability. Whether the containment system is in the design phase or in operation, the reliability of the vertical barrier should be assessed. The reliability of the vertical cutoff wall is a function of several variables. These variables are associated with uncertainties and are evaluated through a probabilistic analysis. The performance function is intended as a means to evaluate a component's dependence on design and control variables. These variables are selected based on engineering design principles, historical field data collected by the authors and data published by others. Assessment of long-term reliability of steel sheet pile wall with sealant is presented and a design equation is derived based on probability and field data. Similarly, reliability of each containment system component can be evaluated. The proposed model provides a means for comparison between different technologies and at the same time allows assessment of the reliability of multi-component systems. Additionally, this model can provide important information that can be used in risk analyses.

Managing biological invasions often necessitates decision making by public officials facing severe uncertainty about many important factors influencing eventual outcomes. Use of relatively sparse decision models is typically necessary and significant lead time for analysis is unfortunately an uncommon luxury. Moreover, development of relevant information to reduce uncertainty and promote more informed decisions in a timely manner is only rarely feasible. Such an uncertain decision making environment can often engender strategies that are simplistic, highly inefficient, and hard to justify in a rigorous manner. In this paper, info-gap decision theory is leveraged to examine a management strategy for light brown apple moth (Epiphyas postvittana) which was discovered in California during 2007. The guiding principle of the analysis is that robustness to uncertainty is an appropriate management objective when facing the severe uncertainty associated with biological invasions. The info-gap strategy is contrasted with strategies following from other popular decision criteria under uncertainty.

Decision-making for conservation is conducted within the margins of limited funding. Furthermore, to allocate these scarce resources we make assumptions about the relationship between management impact and expenditure. The structure of these relationships, however, is rarely known with certainty. We present a summary of work investigating the impact of model uncertainty on robust decision-making in conservation and how this is affected by available conservation funding. We show that achieving robustness in conservation decisions can require a triage approach, and emphasize the need for managers to consider triage not as surrendering but as rational decision making to ensure species persistence in light of the urgency of the conservation problems, uncertainty, and the poor state of conservation funding. We illustrate this theory by a specific application to allocation of funding to reduce poaching impact on the Sumatran tiger Panthera tigris sumatrae in Kerinci Seblat National Park, Indonesia.

Planning is a crucial part of delivering policies on sustainable development of urban infrastructure systems. However, strategic planning has been plagued due to future uncertainties. This paper presents a new technique based on information gap theory to analyze the failure rate of underground pipes under uncertainty and situations of scarce data. It predicts pipe failure based on a simple regression based model and Monte Carlo simulation technique. Once the failure rate is calculated, it analyses the robustness of failure using an information gap robustness function. The information gap robustness function ensures the robustness of the decision under maximum uncertainty without violating the minimum acceptable failure rate of the pipes. The methodology has been demonstrated on a case of water pipes failure in Birmingham UK. The trade-off between uncertainty and failure rate will be helpful for decision making while preparing a sustainable water pipes rehabilitation plan.

Concepts of robustness are often employed when decisions under uncertainty are made without probabilistic information. We present a theorem which establishes necessary and sufficient conditions for non-probabilistic robustness to be equivalent to probability of success. When this "proxy property" holds, probability of success is enhanced (or maximized) by enhancing (or maximizing) robustness. Applications to investment in risky assets and forecasting are discussed.

Benefit cost ratios (BCR) have been applied to conservation decisions for two reasons: cost-efficiency and transparency in decision making. Because BCRs are ratios of benefits to costs, the uncertainties associated with the two components (benefits and costs) are compounded. Therefore, BCRs can potentially involve more uncertainty than allocation strategies based solely on maximizing benefits. The robustness of decisions, defined here as the inverse of the number of misallocations due to uncertainties in benefits and costs of projects, is an unexplored component of applying BCRs to conservation decision making. To investigate the robustness to uncertainty of conservation investment with BCRs, we developed an information-gap model (info-gap) for using BCRs in selecting "portfolios" of conservation projects. Our model allows us to explore how uncertain we can be in our estimates of benefit and cost parameters while still selecting a portfolio that performs better than a critical threshold of misallocations perceived to be unacceptable. We first give a full theoretical description of our info-gap model formulation and then explore applications of the model to several hypothetical data sets.

The problem of determining the coefficients of a linear model from experimental data has been the subject of scientific research since the beginning of the 19^th century. The standard technique, which minimizes the residual error, is sensitive to uncertainties and becomes ill-posed when the data is redundant. Regularization techniques replace the original ill-posed problem with a well-posed problem, but are sensitive to the proper selection of the regularization parameter, which controls the fidelity to the original problem. We have recently applied info-gap techniques to derive a robust-satisficing linear regression. The method is based on satisficing a level of residual error that is consistent with the data. Here we demonstrate the performance of this method and compare it to existing regularization methods including L-curve and generalized cross-validation. It is shown that the proposed method yields superior results in the presence of (i) significant noise, (ii) correlated noise, and (iii) model uncertainties.

A decision maker wishes to distribute a fixed amount of resources between several alternatives. Each alternative responds differently to allocation, as expressed by its responsiveness. The alternatives' different responsiveness to investment can be exploited for efficient allocation of resources. However, the responsiveness is highly uncertain, so prediction is difficult and uncertainty must be accounted for in designing an allocation. We use info-gap theory for satisficing (not minimizing) accumulated loss from several alternatives. We demonstrate the trade-off between robustness to uncertainty and total loss, and show the tight connection between the decision maker's notion of adequacy and the resulting allocation.