Urban water infrastructure (UWI) comprises three primary systems, i.e., water supply systems (WSS), urban drainage/stormwater systems (UDS) and wastewater systems (WWS). These components are essential for delivering clean water to customers and collecting surface runoff from urban areas and sanitary sewage from households in cities. While providing these essential services, these systems also offer opportunities to implement strategies that can enhance urban resilience. However, these systems face various challenges and stresses that can result in technical failures or performance losses, leading to exorbitant costs ^[1]. Some of these challenges include ageing infrastructure and insufficient investment in infrastructure rehabilitation, which can reduce system capacity, population growth and rapid urbanisation that increase system loads, inadequate water infrastructure maintenance, climate change and extreme weather events such as flooding and drought. Addressing these challenges is crucial for maintaining the functionality of UWI and avoiding costly disruptions ^[2].

To address the challenges and stresses faced by UWI, it is crucial to establish a robust and resilient system that can withstand significant disruptions, dynamically reorganise itself and continue to perform essential functions without any interruptions ^[3]. This resilience can be achieved through various measures such as incorporating redundant capacity in the infrastructure, adopting flexible design principles, implementing advanced monitoring and control systems and promoting community engagement and awareness. By constructing a more resilient UWI and ensuring that residents have access to clean water, cities can better withstand the challenges and stresses they face ^[4].

Resilience in urban infrastructure is defined by its ability to maintain essential functions while adapting to external changes, promoting sustainability. Extensive research has been conducted in recent years to understand, analyse and enhance resilience in urban infrastructure, including the development of resilient systems and the measurement of experimental resilience, and improvements to various resilience infrastructures to overcome uncertainty related to future drivers ^[5].

However, measuring global resilience is challenging and requires the use of multiple indicators and metrics. Some common metrics used to measure resilience in UWI include the availability of clean and safe drinking water, the efficiency and reliability of sewage collection systems and the implementation of United Nations safety management facilities ^[6]. There are significant differences in the level of investment given to the main UWI components across the world. Although access to drinking water has improved in the Middle East, water recycling in UWI requires more investment. African countries suffer greatly from a lack of investment in both access to drinking water and proper collection of stormwater and wastewater.

2. Research Design and Bibliometric Analysis

The UWI components consist of various components, ranging from water abstraction to wastewater treatment and more. These components are categorised into three groups: abstraction, water/wastewater treatment and water storage (parts 2, 3 and 7) and distribution of water supply or stormwater/wastewater collection (parts 4, 5, 6 and 8). Water abstraction is often evaluated at the watershed or basin level [11]^[7], while resilience assessment in treatment and storage sections focuses on failure events [12]^[8]. This study concentrates on the distribution of water supply systems or stormwater/wastewater collection systems (i.e., resilience assessment of WSS, WWS and UDS), which is referred to as UWI hereafter. These components are grouped together as they share similarities and, in some cases, complement each other. The holistic approach involves integrating resilience as a fundamental design feature of the system, considering socio-ecological-technical factors to address chronic stressors. It evaluates the system’s capacity to withstand, adapt to and recover from shocks and stresses. This approach can be used to identify the potential risks and vulnerabilities in the system, prioritise investments in resilience-enhancing measures and evaluate the effectiveness of interventions [13]^[9]. The holistic approach involves examining the physical, social and environmental dimensions of UWI and their interactions. The physical infrastructure, which includes the design, construction, and maintenance of water distribution, urban drainage systems, and wastewater collection, as well as the condition and durability of pipelines, pumps, and other components, is a critical aspect of resilience assessment since it forms the backbone of water infrastructure systems [14]^[10]. The assessment should take into account the physical infrastructure’s ability to withstand various hazards. It is important to understand the vulnerability of the infrastructure to these hazards and how they might impact the system’s functionality [15]^[11]. Note that any type of stress caused by a hazard can have an impact on the physical infrastructure’s functionality in UWI. In addition to physical infrastructure, the assessment should also focus on social and institutional systems, which involves examining policies, regulations and governance structures that govern water management, as well as the roles and responsibilities of stakeholders, including water utilities, government agencies and community organisations [16]^[12]. Social and institutional systems are an essential aspect of resilience assessment since they influence the system’s ability to respond to and recover from shocks and stresses. The assessment should evaluate the effectiveness of the social and institutional systems in terms of coordination, communication and collaboration among stakeholders. It should also examine the system’s capacity to mobilise resources and implement interventions to enhance resilience [17]^[13]. Additionally, the natural environment’s focus should be on the ecological processes that support water infrastructure systems, such as the availability of required urban water and the impacts of climate change on these systems. The natural environment influences the system’s ability to adapt and respond to changing conditions, especially droughts, floods, and sea-level rise [18]^[14]. The technical approach focuses on improving the engineering and technical aspects of the system to increase its capacity and resilience to acute stressors or shocks [26]^[15]. Technical resilience approaches typically aim to increase the capacity and robustness of specific system components or infrastructure to withstand and recover quickly from acute stressors such as natural disasters or system failures through targeted engineering solutions such as reinforcing pipes and are more concerned with the physical aspects of the system rather than the social or ecological components [8]^[16]. A technical framework for resilience assessment typically includes several elements [27,28]^[17][18]. The first step is to identify critical infrastructure, mapping out the infrastructure and assets to understand how they are interconnected and dependent on one another. Next, risk assessment is conducted using scenario planning and modelling to better understand the potential impacts of different hazards. A vulnerability assessment is then conducted at different levels of the system, such as the individual assets, the subsystems and the overall system. Capacity assessment is conducted under different scenarios and stressors to better understand the system’s capacity for resilience. Performance evaluation involves using performance indicators to measure the effectiveness of different resilience strategies. Based on the results of the risk, vulnerability and capacity assessments, risk management strategies are prioritised. Finally, a process for continual improvement is established, involving regular monitoring, evaluation and review of the system’s resilience to identify areas for improvement and implement changes accordingly. The “S-FRESI” framework employs indices to assess urban flood resilience before, during and after a flood. The framework is composed of three main components: exposure, sensitivity and adaptive capacity. Exposure refers to the likelihood and severity of flooding, sensitivity measures the degree of susceptibility of urban systems and populations to flooding, and adaptive capacity assesses the ability of cities to recover from floods and build resilience for the future [29]^[19]. The sensitivity component is focused on identifying areas of risk and vulnerability in urban systems and populations to flooding. It helps to identify populations and infrastructure that may be particularly susceptible to flooding. The adaptive capacity component is used to assess the ability of cities to prepare for and recover from floods. It helps to identify areas where improvements could lead to increased resilience in the face of flooding [30]^[20]. The accuracy and reliability of the S-FRESI depend on the availability and quality of data used. The index requires detailed information on the physical, social, and environmental characteristics of urban areas, as well as historical flood events and their impacts. However, the availability and reliability of data required for the S-FRESI framework may be limited, especially in low- and middle-income countries where data collection and management systems may be weak or non-existent [31]^[21]. In addition, the accuracy and relevance of the assessment results can be impacted by the spatial scale and resolution of the assessment, which may not provide enough detail for local-level decision-making [32]^[22]. Moreover, the subjective judgments and weighting of indicators based on expert opinions and stakeholder inputs can lead to inconsistencies in results across different locations and times. This weighting of indicators may also vary depending on the context and objectives of the assessment. Furthermore, stakeholder engagement (e.g., community members, local governments and other relevant actors) may not always be adequate, which can result in a limited understanding of local needs and priorities, as well as a lack of ownership and commitment to the assessment outcomes and recommendations [33]^[23]. RAF is an extension of the S-FRESI approach and concentrates on nature-based solutions for managing and controlling stormwater. It has several critical components, such as economic sustainability, environmental factors, spatial planning, involvedness, system robustness and level of service management. These components are then assessed and validated by external factors such as contributions and stakeholders [34]^[24]. However, similar to S-FRESI, RAF is subject to the subjective nature of the evaluation process, and the weighting of indicators may vary depending on the context and goals of the assessment. This subjectivity may lead to inconsistencies and variations in results across different locations and times. Furthermore, involving different stakeholders can make it challenging to reach a consensus on the indicators to be included and their relative importance [35]^[25]. Although designed to assess the resilience of nature-based solutions, it may not be suitable for evaluating other water infrastructure components [36]^[26]. The PESTEL framework takes a broader view in comparison to the other two frameworks by adding policy and law factors to the assessment factors. This helps identify potential risks and opportunities that may be missed by a narrower focus. PESTEL analysis is a flexible tool that can be adjusted to different contexts and applied at different scales, from individual projects to entire cities. The insights obtained from a PESTEL analysis can inform strategic planning for urban water management by identifying priorities and focusing resources where they are most needed [37]^[27]. However, PESTEL analysis focuses on external factors, such as political and economic conditions, which may limit its usefulness in identifying internal factors that may be contributing to resilience challenges. Furthermore, PESTEL analysis may result in inconsistencies and variations in results across different contexts and stakeholders. The external factors that impact urban water resilience are constantly changing, which can make it difficult to keep the analysis up-to-date and relevant over time [38]^[28]. The “Safe & Sure” framework measures the resilience of UWI using three risk-based parameters, i.e., risk assessment, risk management and recovery assessment. Risk assessment involves identifying potential hazards and assessing their likelihood and consequences. The risk management component focuses on developing and implementing strategies to reduce the likelihood and consequences of hazards, while the recovery assessment component evaluates the effectiveness of risk management strategies and measures the overall system resilience [19]^[29]. The Safe & Sure framework emphasises stakeholder engagement and collaboration in the resilience assessment process, which includes involving system operators, regulators, customers and other stakeholders in the development and implementation of risk management strategies. One of the strengths of the Safe & Sure framework is its flexibility and adaptability to different types of critical infrastructure systems and contexts. However, like other holistic resilience assessment frameworks, the Safe & Sure framework is subject to the involvement of stakeholders and the need for comprehensive and accurate data, which may be difficult to obtain, particularly for complex and interconnected systems [32]^[22]. “SAF” is a robust methodology designed to assess the resilience of urban areas to natural disasters. Its objective is to promote and facilitate interoperability by systematically identifying flood impact and flood source areas and identifying opportunities for the integration of different infrastructure systems to manage surface water [23]^[30]. SAF relies heavily on the data collected and prepared for analysis using geographic information systems (GIS). It uses spatial analysis to identify areas of high and low resilience based on the spatial distribution of various factors. The results of the analysis are then integrated and interpreted to identify the factors that contribute most strongly to resilience, as well as areas where interventions may be needed to improve resilience [39]^[31]. The smart city framework for resilience assessment is a robust methodology designed to evaluate the resilience of cities to various shocks and stresses, including critical UWI. The framework considers the complex and interconnected nature of urban systems and aims to provide a holistic approach to resilience assessment [25]^[32]. It incorporates a range of tools and techniques to facilitate the resilience assessment process, such as GIS mapping, stakeholder engagement, scenario planning and risk assessments. The framework emphasises the importance of collaboration and communication between stakeholders and the need for adaptive and flexible strategies to address the changing nature of urban risks and uncertainties. However, while the framework is comprehensive, it primarily focuses on the resilience assessment of the entire city rather than specifically on UWI [40]^[33]. Overall, technical approaches offer targeted solutions to identified problems, providing a clear focus for addressing specific issues. They often rely on data and quantitative analysis, which can lead to more objective and reliable decision-making [41]^[34]. Additionally, they can be efficient in terms of time and resources since they are narrowly focused on specific issues rather than the entire system [42]^[35]. However, technical approaches can be narrow in their scope, potentially overlooking important interconnections and interdependencies within the system. Their reductionist approach may break down complex systems into their constituent parts, missing the broader picture [43]^[36]. Additionally, they may not fully engage stakeholders or consider their perspectives and needs, leading to solutions that are not sustainable in the long run [44]^[37]. In contrast, holistic approaches consider the system as a whole, considering interconnections and interdependencies between its different parts. This can lead to more comprehensive solutions that address multiple issues and are more resilient to unexpected shocks and stresses [45]^[38]. Holistic approaches also prioritise stakeholder engagement, considering their perspectives and needs in the decision-making process [46]^[39]. However, holistic approaches can be time-consuming and resource-intensive, requiring a broad and detailed understanding of the system. They may also rely on subjective assessments and qualitative analysis, potentially leading to biased or incomplete decision-making [18]^[14]. Additionally, the complexity of holistic approaches may make it difficult to communicate findings to stakeholders who may not have a technical background [47]^[40].

3. Resilience-Enhancing Strategies

System upgrade is a strategy for improving the robustness and redundancy of water infrastructure to increase its resilience. While investing in physical structures, this strategy is still widely regarded as a primary solution for increasing resilience, owing to the high investment and long-term performance that water infrastructure is expected to provide [28]^[18]. For example, long-term optimal rehabilitation strategies in WSS can be obtained by using sequential multi-objective optimisation models [48]^[41]. The majority of system upgrades involve centralised systems, which are criticised for their high energy requirements, changes to the natural hydrological system and long-term costs associated with their maintenance and operation [49]^[42]. Decentralisation is another strategy proposed to improve UWI’s resilience. The system becomes more flexible and reduces water loss due to cutting leakage in long-distance piping networks by distributing the water and wastewater distribution network across the city [50]^[43]. Furthermore, decentralised water systems can facilitate the circular economy of water and resources by allowing treated wastewater to be reused [51]^[44]. Nutrients in wastewater, for example, can be recycled and used as fertiliser in agriculture, and biogas produced from organic matter can be used as a renewable energy source [38]^[28]. This can help to reduce freshwater and energy demand, as well as waste and pollution in the environment. Decentralised systems, on the other hand, may necessitate more complex management and maintenance because they involve a greater number of smaller systems distributed throughout a city rather than a single centralised system ^[4]. This may necessitate additional resources for operation and maintenance, as well as specialised expertise. NBS are yet another type of resilience that combines natural elements and ecosystem services to provide cost-effective solutions such as increased infiltration, evapotranspiration, and stormwater runoff storage [52]^[45]. They have gained traction as an alternative or supplement to traditional UWI [53]^[46]. NBS is also known by the terms Low Impact Development (LID) and Sustainable Urban Drainage Systems (SuDS). This strategy can be combined with other traditional approaches, particularly flood management, to form a comprehensive strategy for long-term sustainable urban drainage [8]^[16]. Multi-criteria optimal planning of a variety of SuDS options can also be highly beneficial for enhancing UDS resilience and hence urban flood management [54]^[47]. While each of these strategies provides valuable information on its own, these strategies can be interrelated. Integrating various strategies can result in more effective and long-lasting UWI. Also, this would improve their ability to provide excellent customer service in the event of unanticipated system failures [64]^[48]. Within this context, creating an emergency response plan and setting up backup water distribution systems can help with swift action in the event of water-related disasters, thus increasing UWI resilience-based resistance. Alternatively, upgraded systems can combine digital technology like smart sensors and monitoring systems to aid in the detection of leaks and faults, resulting in a smarter regime and faster system repairs and maintenance [25]^[32]. Moreover, neighbourhood water recycling facilities and rainwater harvesting systems can be integrated with digital technology, and decentralised systems can become flexible and adaptable to changing water demands, assisting in water rationing during disruptions and thus increasing resilience in situations where UWI systems resistance fail [28]^[18]. Digitalisation and nature-based solutions can also be coupled to increase UWI resilience. For instance, digital technologies and nature-based solutions can be integrated to strengthen UWI resilience through smart green roofs [65]^[49]. These systems use sensors to track weather conditions, soil moisture levels and plant water requirements to optimise irrigation schedules. Flood monitoring and warning systems can also be combined with green infrastructure to serve as a preparedness and emergency response mechanism, increasing UWI’s resilience. These alert systems or device sensors keep track of rainfall patterns and water levels to provide early flood warnings [66]^[50]. Additionally, decision-makers can gain a better understanding of the water system by combining data from various sources such as sensors, weather forecasts and water quality monitoring systems. It can also reduce unnecessary infrastructure and costs by providing real-time data that allow for more efficient and targeted system maintenance and operation [24]^[51]. Another integrated strategy is to advance the upgraded systems by NBS and decentral systems. This strategy has experienced development constraints due to a lack of tools and collated information to determine or uncover its long-term value. However, the method of combining NBS solutions with system upgrades has the potential to increase and contribute to UWI resilience in urbanised areas from the perspectives of resource efficiency and societal, economic and environmental gains [22]^[52]. In UWI settings, the integration of system upgrades and infiltration trenches, vegetative swales and rain gardens would help to regulate stormwater runoff, alleviating demand on UWI and, as a result, pressure on urban drainage assets in urbanised areas. Replacing old drainage network pipes with newer ones can be an expensive upgrade work that many communities cannot afford, so integrating NBS techniques will assist communities that cannot afford such expensive UWI improvement or upgrade works to have a more affordable and resilient UWI. Another efficient strategy to enhance system resilience is to integrate combined sewer networks or UDS with detention ponds to relieve stress on the piped network in the case of failure [71]^[53]. The location and size of these detention ponds can be optimised by using multi-criteria decision-making frameworks [72]^[54]. The advantages of combining a decentralised distribution network and a system upgrade also include increased system efficacy, persistency, adaptability, transformability and sustainability of service provision, demonstrating UWI resilience by proactively providing new infrastructure to the decentralised system at a lower cost [73]^[55]. For example, replacing old pipes in a decentralised system may be less expensive than making the same improvement works to modernise a centralised distribution system. Additionally, such improvement works in a decentralised system will necessitate shorter-length pipes. In this context, system upgrades and decentralisation would more effectively manage water loss owing to leaks in long pipe networks and other wastage, strengthening the efficacy and resilience of UWI [74]^[56]. In addition to resilience enhancement through developing decentralised water reuse strategies, several other performance indicators, such as water conservation and environmental aspects, such as greenhouse gas emissions, can also be improved in UWI [75]^[57].

4. Resilience Indicators

Measuring and assessing system resiliency is critical for effective decision-making and sustainable management. While holistic approaches evaluate resilience using quantitative or descriptive indicators, technical approaches use quantitative metrics. S-FRESI, which focuses on measuring resilience in the face of flood occurrence, demonstrates resilience based on hazard level, population potentially exposed to flooding, density of residential building and duration of water exposure with the population [21]^[58]. The exposure component is determined by a range of factors, including the frequency and magnitude of flooding events, the spatial distribution of flood risk across the urban area and the potential consequences of flooding for people and infrastructure. The sensitivity component evaluates factors such as the density and demographics of the population, the quality and age of infrastructure and the availability of emergency response resources. Adaptive capacity, including susceptibility, material recovery and duration effect, involves evaluating factors such as the availability and quality of emergency management plans and resources, the effectiveness of disaster response systems and the capacity of local government and civil society to coordinate and respond effectively to flood events [29,30]^[19][20]. The framework includes two stages, where the first stage focuses on measuring the nature-based solution at the planning level, stakeholder awareness, public finance, economic opportunities, citizens’ engagement and accessibility, social co-benefits, freshwater provision, water treatment, erosion prevention and maintenance of soil fertility, and habitats for species promotion. The second stage evaluates the role of selected nature-based solutions at the city level by measuring hazard and exposure mapping, land use and inclusion, service management and planning, resource availability and adequacy, flexible service, scenarios relevance for disaster response, infrastructure assets criticality and protection, infrastructure assets robustness, infrastructure monitoring and maintenance, infrastructure preparedness for recovery and build back, infrastructure dependence, and infrastructure autonomy [34]^[24]. The RAF framework emphasises the involvement of stakeholders in UWI’s resilience, measuring their awareness and participation and social co-benefits. However, the subjectivity of the assessment process and the involvement of different stakeholders can lead to divergent views and objectives, making it challenging to reach a consensus on the indicators to be included and their relative importance [35]^[25]. The “Safe & Sure” approaches to assessing UWI’s resilience employs three key indicators: risk level, reliability degree and recovery rate from extreme events. Risk assessment involves analysing the physical, technological and operational vulnerabilities of the system, as well as its dependencies on other systems and stakeholders. The risk management component includes measures such as redundancy, diversity and robustness, as well as plans for emergency response and recovery. The recovery assessment involves evaluating the system’s ability to absorb and recover from disruptions, adapt to changing conditions and maintain essential services and functions. As a result, it appears that measuring and quantifying resilience can be difficult, with no single universally accepted method. There are numerous frameworks and models that attempt to capture the various dimensions of resilience, but each has limitations and potential biases. Furthermore, because resilience is a complex and dynamic concept, developing metrics and models that accurately capture all the relevant factors and interactions can be difficult. Nonetheless, efforts to measure and evaluate resilience are critical for better understanding the concept and guiding decision-making in areas such as urban water management. “FRI” is another technical approach that investigates resilience in two stages: response and recovery time. In the response phase, water depth and flood duration are measured, and in the recovery phase, flood severity, total water depth and total flood are measured. Furthermore, the rate of affected elderly population, women households, and children in collaboration with household income will also be measured.

5. Resilience-Simulating Tools

Physically based modelling is an effective tool that can predict how UWI behaves under different stressors and scenarios. It can simulate the impacts of natural disasters, such as floods, hurricanes and earthquakes, on UWI and assess the effectiveness of various resilience measures. For example, it can predict the behaviour of water distribution networks under different scenarios, such as power outages, pipe failures and extreme weather events, and identify areas that require resilience measures [82]^[59]. Furthermore, physically based modelling can evaluate the effectiveness of different adaptation strategies, such as green infrastructure, in reducing the vulnerability of UWI to natural disasters. This approach is linked to global resilience analysis (GRA), which assesses the resilience of systems and communities at a global level [83]^[60]. GRA involves identifying the key drivers and indicators of resilience, analysing their interconnections and assessing the resilience of systems and communities based on their ability to adapt and respond to shocks and stresses [26]^[15]. The software tools for modelling the UWI that are commonly used in the research work for simulating the resilience performance in UWI are (1) EPANET, (2) Stormwater Management Model (SWMM), (3) MIKE URBAN, (4) Urban Water Optioneering tool (UWOT) (5) WaterMet² and (6) System for Infrastructure Modelling and Assessment (SIMBA). The first three tools (EPANET, SWMM and MIKE URBAN) are physically based models that are typically data demanding and hence their applications are limited to those components that access to all physical data is available. EPANET is a simulation model for water distribution systems and is typically coupled with optimisation models to obtain optimal rehabilitation strategies and operation [84^[61][62],85], while it has also been applied for resilience assessment of system failure [26]^[15]. However, the other three (UWOT, WaterMet² and SIMBA) are conceptually based models that are less data demanding with simplified system components used for modelling purposes. Although MIKE URBAN allows the integration of real-time data, such as weather forecasts and sensor measurements, to provide more accurate modelling and predictions, SWMM is more popular due to its free availability and greater capabilities in simulating single flood events or long-term runoff [28,86]^[18][63]. WaterMet² is a software tool used for both technical and holistic approaches in an integrated UWI. It can also combine hydrological, hydraulic and water quality models to simulate and optimise UWI performance [87]^[64]. SIMBA is a comprehensive tool that models various UWI components and uses a simulation-based holistic approach to assess the performance of UWI under various scenarios, such as changing populations or climate conditions [88]^[65]. This tool supports decision-making in the planning, design and operation of UWI. UWOT is a conceptually based modelling tool for performance assessment of UWS and can be efficiently used to estimate resilience indicators of various water management options (e.g., household appliances and fittings or rainwater harvesting schemes) under various scenarios/stressors. UWOT also estimates the energy required by water appliances and evaluates water and wastewater reuse and other green technologies [89]^[66].