ASCE-ASME Journal of Risk and Uncertainty in Engineering Systems, Part A: Civil Engineering

Current design codes and standards in the United States focus on the design of individual facilities and do not account for other coupled physical and nonphysical infrastructure. This does not provide a direct mechanism to enable the design of resilient communities since one network may depend on another and social and economic institutions are then adversely affected. Therefore, in this study, the current approach for designing individual physical components within a community is reimagined such that it not only considers the performance of a component individually after a damaging event but also considers the consequences its design has on other physical and nonphysical parts of the community. In this regard, a methodology was developed that links the performance of components within the built environment to community-level resilience goals by considering the dependencies and cross-dependencies for networks. This methodology enables disaggregation of the community-level objectives into a set of performance targets for the components of the built environment. The methodology presented herein can be implemented in risk-informed decision-making tools to design master-planned resilient communities as well as upgrade a community’s buildings and/or networked infrastructure to make the community more resilient to future disasters. This could be used, for example, to understand how to better isolate or decouple two or more sectors, establish redundancies, or decide whether to focus on retrofitting schools in the public sector, businesses, and residences in the private sector, or (more likely) some combination thereof.

The distribution of the maximum load generated by stochastic and recurring hazards is of primary importance in structural reliability analysis. In the current literature, this distribution is estimated by either relying on the asymptotic extreme value theory or assuming that occurrences of a hazard follow the homogeneous Poisson process. However, assumptions underlying these approaches become questionable when the maximum load distribution is required for a short service life, such as in reliability assessment of temporary structures and aging infrastructure systems nearing the end of life (e.g., old nuclear plants). This paper fills this gap in the literature by presenting a more general and accurate solution for the probability distribution of the maximum load generated by stochastic hazards which can be modeled as shock, pulse, and alternating renewal processes. This work is a considerable advancement of the state of the art in probabilistic analysis of maximum value distribution.

The operation of modern societies relies on the functionality of complex infrastructure such as those for potable water, electric power, and transportation. Difficulty in accessing life-supporting resources due to the loss of the functionality of infrastructure in the aftermath of natural or anthropogenic hazards can result in widespread societal disruptions. To promote societal risk and resilience analysis, this paper makes the following novel contributions: (1) probabilistic models are developed to predict the broad societal impact of disruptive events over time in terms of their impact on the well-being of individuals; (2) a mathematical formulation for societal resilience analysis is developed that integrates the immediate impact on and the recovery of individuals’ well-being; (3) the developed probabilistic models are implemented with Dynamic Bayesian Networks; and (4) a formulation is proposed to evaluate the quantified risks. To estimate the immediate impact on individuals’ well-being and model the subsequent recovery, the information from the recovery modeling of infrastructure and variations in the socioeconomic characteristics were incorporated into a time-dependent reliability analysis. The probabilistic modeling of the immediate impact and recovery of well-being were used to quantify societal resilience. To facilitate the probabilistic modeling, the time-dependent reliability analysis was implemented with a Dynamic Bayesian Network. Finally, the quantified risk and resilience were evaluated to provide insights about the severity levels of disruptive events. The proposed approach is explained, through a real case study, to quantify the cascading impact of infrastructure disruptions.

Although variability in material properties and loading conditions is commonly accounted for in safety and reliability studies, this is not the case for model assumptions. In addition, many other uncertainties affect the design and evaluation stage of civil engineering structures. This paper describes how a consistent modeling methodology can be developed for a multicriteria evaluation and multiobjective optimal design of constructions. A series of tools and methods [probabilistic description of variables, desirability functions, multiobjective particle swarm optimization (MOPSO), analytic hierarchy process (AHP), and decision aid method, among others] has been applied in order to be able to reach an optimal design alternative. The methodology is illustrated on the real case study of an industrial structure. The focus is given to several key points showing how uncertainties and lack of knowledge influence the optimization process. It is shown that items such as the epistemic uncertainty on correlation between material properties, consideration of soil–structure interaction, and designer preferences can affect the optimal solution. A robust optimization scheme is proposed that considers in a consistent way uncertainties of various origins.

A real options (RO) formulation is proposed for decision making on the timing to upgrade the seismic performance of existing seaports with increasing throughput demand in earthquake-prone areas. The payoff for the seismic upgrade investment option is estimated based on projected net earnings, repair cost, and downtime for a damaging reference seismic event having a prespecified annual probability of occurrence. These projections inform a discrete-time RO binomial tree, following the American option valuation framework, which propagates the probability of the reference seismic event assuming Poisson temporal distribution of earthquake occurrence. The net present value of the expected annual payoff of the considered investment is used as an index supporting risk-informed decision making discounted by the weighted average cost of capital (WACC). Numerical examples pertaining to decision makers with different capital costs—namely, port authorities and terminal operators—operating in different economic environments typical of developed and developing countries are furnished to illustrate the applicability of the proposed RO formulation. It is found that high WACC and/or low throughput growth bring the optimal seismic upgrade timing forward, while earthquake consequences and upgrade cost have almost no influence on this timing.

Risk management programs and catastrophe models use fragility and vulnerability curves extensively. For the case of coastal flood events, the independent variable for these damage functions is usually the inundation depth, sometimes combined with some expression of water velocity or wave action. Postdisaster surveys often provide the basis for these damage functions, where the investigators classify the observed damage into broad categories based on qualitative descriptions. This paper describes a method to transform these qualitative evaluations into quantitative descriptions of damage states, which are then applied to develop fragility and vulnerability curves. The authors present this process within the context of the development of coastal flood fragility and vulnerability functions for the Florida Public Hurricane Loss Model. The model characterizes and quantifies the damage states specific to a set of fragility curves by using damage distributions and component cost analysis and transforms the fragility curves into a vulnerability curve. The paper analyzes the uncertainties in the model due to the number and quantification of the damage states and an adjustment function included in the discretization process. The analysis shows that the number of damage states governs the overall uncertainty. Model outputs are compared with USACE expert opinion depth-damage functions to validate the model and identify aspects for further refinement.

The multiple uncertainties of both natural and human-caused disasters have attracted increased attention to the topic of resilience engineering. In this paper, an indicator-based method for measuring urban community resilience is proposed. The method is based on the PEOPLES framework, which is a hierarchical framework for defining the disaster resilience of communities at various scales. It consists of seven dimensions, summarized by the acronym PEOPLES, which stands for population, environmental and ecosystem, organized governmental services, physical infrastructures, lifestyle, economic development, and social capital. Each of the dimensions is split into several components and indicators, which were derived by the authors or collected from a wide range of literature. Each indicator is represented using a performance function, which portrays the functionality of the indicator in time. A higher functionality of the indicator denotes a higher resilience of the community. These functions can be constructed in a systematic manner using damage and restoration parameters. The aggregation of the performance functions passing through the different hierarchical levels of PEOPLES framework leads to one function that represents the dynamic performance of the analyzed community. This paper also introduces a matrix-based interdependency technique that serves as a weighting scheme for the different indicators. As a case study, the proposed methodology is applied to the city of San Francisco for which a resilience curve and resilience metric have been computed.