Population growth, economic development, and rapid urbanization in many areas have led to increased exposure and vulnerability of structural and infrastructure systems to hazards. Thus, developing risk-based assessment and management tools is crucial for stakeholders and the general public to make informed decisions on prehazard planning and posthazard recovery. To this end, structural risk and resilience assessment has been an ongoing research topic in the past 20 years. Recently, machine learning (ML) techniques have been shown as promising tools for advancing the risk and resilience assessment of structure and infrastructure systems. To date, however, there is a lack of a holistic review on ML progress across various branches of structural engineering; an in-depth analysis of literature that can provide a timely evaluation of risk and resilience assessment methods of the built environment, where different types of structural and infrastructure facilities are interconnected. For this reason, this study conducted a comprehensive review on ML for risk and resilience assessment in four main branches of structural engineering (buildings, bridges, pipelines, and electric power systems). To cover the crucial modules in the prevailing risk and resilience assessment frameworks, existing literature is thoroughly examined and characterized in terms of six attributes of ML, including method, task type, data source, analysis scale, event type, and topic area. Moreover, limitations and challenges are identified, and future research needs are highlighted to move forward the frontiers of ML for structural risk and resilience assessment.

Water distribution systems (WDS) are among the most critical civil infrastructure systems. A reliable and safe water supply is essential for the prosperity and continued well-being of society. Hence, adequate maintenance, repair, and renewal of the vast physical assets in WDS are of paramount importance, especially in places with widespread deterioration of critical water infrastructure assets. However, due to limited resources, water utilities struggle for their asset management strategies to keep pace with the aging of components. Recently, utilities have made efforts to refine the traditional condition-based approach to pipeline asset management that focuses on maintaining the current condition of assets into a service-based approach in which system performance is explicitly considered. This paper reviews existing literature on the various aspects of asset management of WDS, focusing specifically on buried pipelines. Existing condition assessment (CA) techniques, failure prediction models, time-dependent vulnerability models, maintenance planning strategies, and renewal techniques are reviewed. Various techniques of modeling WDS performance, as well as its interdependence with other infrastructure systems, are also reviewed. As such, the paper presents a broad overview of the various aspects that constitute a service-based asset management strategy for WDS. The advantages and limitations of different methods and approaches to the various aspects are also compared to assist selection of an appropriate tool. An overview of future research needs related to asset management and interdependence is presented.

Water distribution systems (WDSs) are among the most essential infrastructure systems for the sustenance of societies. Therefore, their continued reliability is of utmost importance. However, the reliability of WDSs decreases over time due to aging and deterioration of components, especially pipelines. Metallic pipes are subject to external and internal corrosion overtime, which can significantly affect their mechanical strength and hydraulic performance and, consequently, WDS reliability. Existing research on system reliability is typically based on statistical failure prediction of pipelines, which, however, typically assumes a constant failure rate, thereby ignoring time-dependent strength deterioration due to corrosion. This paper proposes a framework for modeling the time-dependent reliability of WDSs with cast iron pipes considering both mechanical and hydraulic reliabilities. Time-dependent failure of pipelines is evaluated through a fragility analysis that considers the decrease of wall thickness over time due to corrosion. The impact of internal corrosion on hydraulic performance is also accounted for with a time-dependent roughness model. A notional WDS is used to demonstrate the proposed framework. The results of the case study show that pipe corrosion can significantly impact the reliability of a WDS as it ages. The proposed framework links the deterioration of pipeline strength and hydraulic performance with system reliability and can, therefore, form the basis of a reliability-based plan for maintenance and rehabilitation.

Failure of a buried water pipeline can have an adverse effect on neighboring infrastructure, especially road networks. The impact of the failure of water pipelines on road networks and water distribution systems (WDSs) significantly increases the economic and social consequences of such failure. This paper presents a risk-informed decision support framework for WDSs considering the risk and the criticality of components to aid maintenance prioritization. The probability of water pipe failure is estimated using a physical probabilistic approach. The economic, operational, environmental, and social consequences of the failure of the integrated water and road segments are evaluated using 14 factors. The economic, operational, environmental, and social consequences are combined using fuzzy hierarchical inference to determine the overall consequence of the failure of each integrated segment (road and water network sharing the same geographical space). The risk of assets is determined by utilizing two approaches: risk equation and risk matrix. A shortest path–based network efficiency metric is then used to identify the impact of the failure of water pipelines on both infrastructure systems. The final decision alternatives are prepared by combining the outputs from the risk analysis and the network efficiency metric to prioritize maintenance tasks. A geospatial model is used to identify dependent road and collocated water segments sharing the same geographical space. The water and road networks of the Rancho Solano Zone III area of the city of Fairfield, California, are used to illustrate the proposed framework. The results show that the failure of a critical segment can have a significant impact on the efficiency of both networks. In the considered case study, the failure of a critical segment can result in about 7.5% and 9.6% system efficiency loss in the water and road networks, respectively. The proposed model is expected to assist in integrated municipal asset management decision-making.

Buried water pipelines are subjected to corrosion deterioration and are vulnerable to extreme events (e.g., earthquakes). Corroded metallic buried pipelines have experienced severe damage during past earthquakes resulting in the service disruption of Water Distribution Systems (WDS). Also, continuous and adequate water supply is imperative for fighting fires caused by earthquakes. Therefore, evaluation of the functionality of WDS and its service restoration after an earthquake is very important for the water service provider as well as communities. This paper presents a framework to evaluate both component-level and system-level seismic risk and resilience of WDS considering time-variant corrosion of pipelines. The vulnerability of individual pipelines is modelled by modifying the American Lifelines Alliance (ALA) pipeline vulnerability functions to explicitly account for the strength deterioration due to corrosion. Resilience is modelled using a functionality curve that relates system failure probability and repair activities over time. Probabilistic functionality fragility surface (FFS) is developed for both components and system by integrating the fragility curves and the restoration functions. The use of the proposed framework is demonstrated on two hypothetical water networks, one small scale and one medium scale. The results of the two case studies show that the restoration process has a significant influence on the shape of the FFS. Time to repair a break, number of breaks, network topology, the level of corrosion, and the available resources of utility companies are key parameters that determine the shape of the FFS. It is also observed that larger and older WDS networks experience a significantly higher number of breaks, which compromises their resilience. The proposed framework provides a useful planning tool for asset management of WDS subjected to seismic hazards.