Decision support tools for water reuse: a systematic review

AHP

analytical hierarchical process

CGT

cooperative game theory

CC

climate change

CML

Centrum voor Milieukunde Leiden

CP

compromise programming

DALY

disability-adjusted life year

DST4WR

decision support tools for water reuse

EP

eutrophication potential

FD

fossil fuel depletion

FEU

freshwater eutrophication

GHG

greenhouse gas

GP

goal programming

GWP

global warming potential

HT

human toxicity

ILCD

International Reference Life Cycle Data System

ISI

Institute for Scientific Information

LCA

life cycle analysis

LCC

life cycle cost

LCI

life cycle inventory

LCIA

life cycle impact assessment

LRV

log removal value

MAVT

multi-attribute value theory

MCDA

multi-criteria decision analysis

MEU

marine eutrophication

NPV

net present value

PDF

probability distribution function

QMRA

quantitative microbial risk assessment

SDG

Sustainable Development Goal

SqMRA

semi-quantitative microbiological risk assessment

TEC

terrestrial ecotoxicity

TODIM

Tomada de Decisão Iterativa Multicritério

TRACI

Tool for Reduction and Assessment of Chemicals and Other Environmental Impacts

WD

water depletion

WSM

weighted sum method

WSP

water safety plan

WTP

water treatment plant

WWTP

wastewater treatment plant

In the context of climate change, the state of water resources has become an issue of global importance (Maeseele & Roux 2021; Gómez-Monsalve et al. 2022; Contzen et al. 2023; Crovella et al. 2024). Climate changes affect coastal areas, leading to sea-level rise and the intensification of extreme precipitation events (Foglia et al. 2021; Crovella et al. 2024). Additionally, more frequent droughts contribute to water scarcity. Water supply in urban areas has become particularly vulnerable due to population growth, economic development, and increasing urbanisation (Goyal & Kumar 2020; Foglia et al. 2021; Maeseele & Roux 2021; Panagiotou et al. 2022; Silva 2023). High levels of urbanisation have contributed to the degradation of surface waters, especially those that flow through the centres of large cities (Isaac et al. 2022; Panagiotou et al. 2022; Contzen et al. 2023). According to the United Nations, water scarcity affects more than 40% of the world's population, and by 2050, at least one-fourth of the world's population is expected to live in countries suffering from chronic freshwater scarcity (Rodríguez et al. 2021; Gómez-Monsalve et al. 2022).

Water reuse is emerging as a viable option for addressing water shortages in a growing number of projects (Rebelo et al. 2020; Zhiteneva et al. 2020, 2021). This alternative can significantly reduce pressure on water resources by providing an additional source of water, aligned with circular economy principles, minimising water pollution by reducing the discharge of untreated wastewater, and supporting several Sustainable Development Goals (SDGs) (Kanchanapiya & Tantisattayakul 2023). First, it contributes to SDG 6 (Clean Water and Sanitation) by ensuring sustainable management of water, improving water quality, and reducing pollution. Second, it supports SDG 13 (Climate Action) by enhancing climate resilience through providing a reliable water source during droughts. Last, it promotes SDG 15 (Life on Land) by protecting terrestrial and freshwater ecosystems, thereby minimising wastewater pollution (Cabling et al. 2020; Dingemans et al. 2020; Foglia et al. 2021; Isaac et al. 2022; Lima et al. 2022; Riazi et al. 2023; Silva 2023; Crovella et al. 2024; Torre et al. 2024).

Several challenges hinder the widespread acceptance of water reuse. The general public often expresses concerns about the safety of water reuse, fearing the presence of contaminants (Contzen et al. 2023; Silva 2023). In addition, lack of knowledge about the benefits of reuse can lead to resistance and health concerns (due to microorganisms and chemicals in the water intended for reuse). Negative perceptions of water due to characteristics such as taste, odour, or colour are also a significant barrier (Contzen et al. 2023). The costs associated with reuse may be considered prohibitive by some stakeholders, and uncertain or confusing regulations may create uncertainty in the adoption process (Contzen et al. 2023; Riazi et al. 2023). Concerns about the environmental impact of reuse and lack of community involvement can also become additional challenges to overcome (Contzen et al. 2023; Silva 2023). Ultimately, public perception plays a key role in the general acceptance of water reuse (Contzen et al. 2023).

The potentially higher cost (especially compared with conventional water), human health risks, and public perception are the main challenges to widespread acceptance of water reuse. These factors are relevant to the sustainability of water reuse technologies, which is influenced by three interrelated dimensions: environmental, social, and economic (Chhipi-Shrestha et al. 2019). Health risks are particularly important because they directly impact public acceptance and trust, which are crucial for the social dimension of sustainability (Damaceno et al. 2022). Furthermore, managing health risks ensures the safety and well-being of communities, which is fundamental for long-term viability. Environmental criteria, such as the reduction of pollution, conservation of natural water resources, and mitigation of climate change impacts, are also essential for the sustainability of water reuse technologies (Szalkowska & Zubrowska-Sudol 2023; Crovella et al. 2024). Therefore, addressing health concerns and environmental impacts together with economic feasibility is vital for achieving sustainable water reuse.

There are several components to sound management of water reuse projects: water demand and availability, water quality (health and safety), treatment technology, policy and regulation, economics, stakeholder involvement, and public acceptance (Dingemans et al. 2020). Successful management of water reuse projects requires a comprehensive approach from the outset (Ddiba et al. 2023). This includes a rigorous assessment of project feasibility, careful selection of treatment technology to achieve the desired water quality for the intended purpose, and ensuring compliance with local and national regulations. Effective cost management, stakeholder involvement, and information sharing with the public are key (Riazi et al. 2023). Continuous monitoring of water quality is essential to ensure its safety. In addition, long-term sustainability requires flexibility, risk identification and assessment, and technical expertise to ensure project success. This expertise is essential for managing complex water treatment technologies, complying with stringent regulations, and adopting best practices that reduce operational and safety risks, thus promoting efficient and safe project management (Contzen et al. 2023; Riazi et al. 2023; Silva 2023).

Concurrently, decision support tools for water reuse (DST4WR) play a critical role in overcoming the challenges associated with the widespread adoption of water reuse and the effective management of its projects (Ddiba et al. 2023). DST4WR are systematic approaches or frameworks that aid decision-makers in evaluating and comparing different options by incorporating various criteria and data. These tools provide a sound basis for assessing risk, identifying critical areas, and making informed decisions. Approaches such as microbiological risk assessment (MRA), life cycle analysis (LCA), and life cycle cost (LCC) were specifically chosen for this literature review due to their ability to offer comprehensive evaluations from different perspectives. MRA helps in assessing and managing microbiological and safety risks, ensuring public health protection (Damaceno et al. 2022). LCA quantifies environmental impacts, facilitating the identification of sustainable practices (Crovella et al. 2024). LCC analyses the economic feasibility by evaluating the total cost of a project over its entire lifecycle (Torre et al. 2024). Integrating these results into a multi-criteria decision analysis (MCDA) allows for a balanced consideration of multiple factors, including environmental, economic, social, and technical criteria, leading to well-rounded and informed decision-making (Kanchanapiya & Tantisattayakul 2023).

Microbiological risk is a major concern for many water reuse applications. It is recommended that a decision support tool called MRA be used (Bailey et al. 2020). At the European level, MRA has become mandatory for agricultural irrigation. In Portugal, for example, it is required for all uses (Lima et al. 2022). The new Portuguese policy focuses on adopting projects supported by a risk management framework and quality standards, which are defined according to a fit-for-purpose approach based on ISO 16075 standards to ensure the application of best practices. European Regulation 741/2020 was also developed, setting minimum quality requirements for use in agricultural irrigation and following the same principles. This regulation has been in force since 26 June 2023 (EU 2020/741 2020).

In addition to public health, an environmental assessment may include a broader perspective considering the overall impact of alternatives. In the area of alternative water sources, the methodological approach of LCA has been used to understand and quantify the environmental impacts of different water-saving strategies. LCA follows a standardised methodological approach that includes defining the goal and scope, inventory analysis, impact assessment, and interpretation. The results of LCA form the basis for decision support tools, as they evaluate the environmental impacts of various scenarios in water reuse projects, thereby supporting decision-making on water scarcity, mitigation strategies, and the design of more sustainable options (Gómez-Monsalve et al. 2022; Kalboussi et al. 2022; Starkl et al. 2022; Torre et al. 2024). LCA can quantify environmental impacts, materials, energy uses, and releases to the environment (Cabling et al. 2020). As a complementary approach of the LCA, the LCC helps to assess the total costs associated with the project over time, providing financial clarity (Chhipi-Shrestha et al. 2019; Arden et al. 2020; Foglia et al. 2021). Together, these methods offer a comprehensive evaluation framework that considers environmental and economic aspects, facilitating informed decision-making and sustainable development practices.

Furthermore, increasing awareness of natural resources and their fragility demands greater responsibility. The question, therefore, arises as to what are the key factors that can be expressed in a decision support tool based on MCDA, whether from an economic, environmental, social, or technical point of view (Chhipi-Shrestha et al. 2019; Dehaghi & Khoshfetrat 2020; Goyal & Kumar 2020; Diogo et al. 2021; Isaac et al. 2022). MCDA is a useful approach, incorporating a combination of quantitative and qualitative information to take into account the preferences of different stakeholders (Isaac et al. 2022; Kanchanapiya & Tantisattayakul 2023). This is an important analysis because it helps identify interests that may be in conflict at the local level, thus helping decision-makers find solutions acceptable to all stakeholders (Starkl et al. 2022).

The integration of methodologies such as MRA and LCA/LCC with MCDA is extremely important. As it comprehensively addresses the main elements of the challenges highlighted by various authors – such as risk, acceptance, cost, quality, and environmental sustainability – it fosters the involvement of all stakeholders. By creating a structured and transparent framework for evaluating water reuse projects, decision-makers can balance public health concerns with environmental and economic considerations. For example, using MCDA allows for the incorporation of local data and stakeholder preferences, ensuring that the selected water reuse strategies are tailored to the specific needs and conditions of the community (Kanchanapiya & Tantisattayakul 2023). This integration facilitates a more comprehensive assessment, ensuring that the benefits and trade-offs of water reuse projects are thoroughly evaluated (Kanchanapiya & Tantisattayakul 2023; Torre et al. 2024).

Ultimately, this review article aims to study and evaluate different DST4WR (MRA, LCA/LCC) when used simultaneously to support an MCDA for informed decisions related to water reuse. The motivation behind this review lies in the urgent need to address water scarcity and ensure sustainable water management. By examining the integration of these tools, this review seeks to provide a valuable framework for decision-makers, enabling them to implement water reuse projects that are safe, economically viable, and environmentally sustainable. This comprehensive evaluation will contribute to advancing the knowledge and practices in the field of water reuse, ultimately supporting the achievement of global water sustainability goals.

According to Cook et al. (1997), conducting a systematic review offers many advantages, namely:

• allowing the researcher to select data;

• developing hypotheses;

• estimating the sample size;

• better defining the method to be applied; and

• defining directions for future applications.

The literature review was conducted as a general systematic review, following these principles to ensure a comprehensive and rigorous analysis. This methodology was applied in two stages. In the first stage, a search was carried out in a database selected on the basis of the defined keywords. During the second stage, the articles were analysed by applying the exclusion criteria.

According to the review methodology, this article searches technical-scientific publications in a predefined time frame, using specific keywords that could relate to water reuse and the DST4WR used/studied.

The time scale was set between 2020 and 2024, with the primary objective of obtaining more recent data. In addition, the short time scale allowed for a more coherent and rigorous study. Initially, the DST4WR ‘life cycle analysis or life cycle assessment’, ‘multi-criteria analysis’, and ‘risk assessment’ were researched; however, it was found that not all articles were applied to water reuse. A ‘plus’ was added, including ‘water reuse’. It is worth noting that when searching for the keyword ‘life cycle analysis’ or ‘life cycle assessment’, different cost evaluation methods were also found. However, only articles related to LCC were included in this study, as it is a commonly used methodology within LCA and serves as a complementary approach. It was decided to select only scientific and English-language articles, and the Online Knowledge Library (b-On) was used. b-On is a government initiative launched in Portugal in 2004 that allows the consortium to acquire titles from leading international scientific publishers from the Institute for Scientific Information (ISI) Web of Knowledge. It provides unlimited and permanent access to research and higher education institutions and is a reference for access to indexed and peer-reviewed scientific information.

In this step, the documents selected in step 1 were analysed according to the keywords mentioned in the previous point. The exclusion criteria were as follows:

• articles that appeared more than once in the search engine;

• articles that cited the keywords only in the references or title, without any mention of DST4WR in the text; and

• articles that, after a thorough read, were found to be focused on something other than the intended topic (for example, in MRA, priority was given to articles that addressed the public health context rather than the environment).

Furthermore, as an inclusion criterion, one known article published outside the specified time scale that could fill the identified gaps/weaknesses was added. This article from 2019 applied three DST4WR simultaneously – MRA, LCC, and MCDA (Chhipi-Shrestha et al. 2019).

Once the studies had been selected, a schematic map was created to analyse their geographical distribution. Subsequently, a discussion was held according to each keyword and, finally, integration on the relevance of using the studied DST4WR for more assertive decision-making regarding water reuse projects was carried out.

The results will be presented in a way that is analogous to the two steps of the methodology.

Figure 2 shows the geographical analysis based on the location of the case studies in the selected articles. LCA is used in more countries and continents when compared with MRA. MCDA is not yet used in Europe, even though the American and Asian continents present the most studies on these tools. Moreover, despite being the largest and most populous continent, Asia presents few studies on these DST4WR. It is important to note that Africa, the second largest continent in the world, is yet to use the MRA tool, despite having severe pollution problems. As for LCC, it is the least used and is adopted in the context of other methodologies.

This section discusses MRA, LCA/LCC, and MCDA, by this order. A table is presented for each tool, with the main aim of relating them to each other and checking what they have in common, even when applied to different case studies. The three different tables can be found in the Supplementary material.

Risk assessment, especially concerning human health, is of the greatest importance. To ensure safety for all, the risk needs to be adequately assessed and managed, so that full transparency of actions and information is communicated to the public, ensuring greater acceptance of reuse projects (Lima et al. 2022). The World Health Organization (WHO) suggests several risk assessment approaches that consider the different management needs of each reuse project (Bailey et al. 2020; Lima et al. 2022). These can be qualitative and semi-quantitative or quantitative models, in addition to sanitary inspection, which requires a simple method and an effective approach for smaller water supplies (Lima et al. 2022). All approaches convey four main steps directed at minimising the risks to public health and ensuring an acceptable level of risk (Rebelo et al. 2020; Zhiteneva et al. 2020; Kongprajug et al. 2021; Damaceno et al. 2022; Lima et al. 2022):

• 1. Identification of hazards – The hazard is related to a chemical, biological, or physical agent that can cause harm to public health.

• 2. Identification of exposure routes for different receptors – The receptors are the people exposed to the hazard via the exposure routes (ingestion, inhalation, or adsorption).

• 3. Dose vs. exposure/exposure scenarios – Dose of disease incidence in humans with the possible use of dose-response models. The scenarios will have to consider the case study, which introduces a risk to each receptor.

• 4. Risk characterisation – Assessment of the probability of occurrence of damage through the exposed hazards.

The MRA identified in this study did not include an evaluation of micropollutants but rather focused on microbiological risks. This focus is supported by ISO (2020), which indicates that there is currently no evidence of adverse effects from emerging contaminants, such as pharmaceutical residues and personal care products, on human health or the environment when using water reuse for irrigation or consuming crops irrigated with such water. Therefore, the primary hazards associated with water reuse for irrigation are microbiological, particularly pathogenic microorganisms. Similarly, Zhiteneva et al. (2020) emphasised that the main risks related to water reuse stem from faecal contamination indicators, which are crucial for assessing microbiological human contamination.

Meffe et al. (2021) applied a pharmaceutical risk assessment. Although the MRA decision support tool is used to assess microbiological risk in water reuse, it is essential to note that other sources of contamination, such as pharmaceutical waste, can also pose a significant risk to public health, especially for potable reuse (Zhiteneva et al. 2020). Meffe et al. (2021) assessed pharmaceutical risk in water reuse systems, highlighting the importance of considering sources of contamination other than microbiological risk. Although the methodology used by Meffe et al. (2021) is not described in this discussion (because it does not include aspects of qualitative or semi-quantitative assessment), it was included in this systematic review.

To support the decision-making process based on the health risk resulting from exposure to pathogenic microorganisms, the quantitative microbial risk assessment (QMRA) was developed (Seis et al. 2022) for water supply (Panagiotou et al. 2022; Seis et al. 2022). However, it was adopted as a generic approach because no other approach could apply to water reuse. This approach was applied by Masciopinto et al. (2020), Seis et al. (2022), Kongprajug et al. (2021), Arden et al. (2020), Zhiteneva et al. (2021), Bailey et al. (2020), Chhipi-Shrestha et al. (2019), Panagiotou et al. (2022), and Kanchanapiya & Tantisattayakul (2023). It is a probabilistic assessment of the pathogens present in each treatment step, which, together with defined exposure scenarios and dose-response models, obtain a risk characterisation. QMRA can be used as part of a water safety plan (WSP) and takes into account information in quantifying the tolerable health burden in terms of disability-adjusted life years (DALYs) (Zhiteneva et al. 2021). DALY is a public health measure used in WHO guidelines to measure the Global Burden of Disease (Panagiotou et al. 2022; Kanchanapiya & Tantisattayakul 2023).

The semi-quantitative microbiological risk assessment (SqMRA) is based on the risk matrix approach, which makes it possible to assess the different risks associated with water quality, the probability of exposure of a receptor to a specific hazard, and its severity/consequences in case it happens. SqMRA can be used to input possible quality standards, indirectly representing the possible tolerable doses. Through this approach, it is possible to combine physical, chemical, or biological barriers aimed at minimising direct contact with water reuse. The concept of an equivalent barrier can also be used and appears as a control measure that produces a result such as microbiological reduction, thus eliminating a given hazard and reducing the risk to an acceptable level (Rebelo et al. 2020; Damaceno et al. 2022; Lima et al. 2022). This methodology was applied by Lima et al. (2022), Rebelo et al. (2020), and Damaceno et al. (2022).

Starting with defining the identification of the hazards, hazard, as a concept, refers to chemical, biological, or physical agents that may endanger human health. In QMRA, a pivotal aspect involves the concentration of pathogens in water at various points within the system, encompassing treatment, distribution, and exposure stages. This concentration data, along with the defined exposure scenarios, is crucial in shaping human health objectives, particularly in relation to log removal values (LRVs) (Seis et al. 2022). LRVs denote the base 10 logarithm of the ratio of pathogen concentrations between influent and effluent water, serving as a fundamental tool for characterising the effectiveness of treatment barriers. By routinely monitoring the relevant (reference) pathogens and examining how treatment processes impact pathogen concentrations, LRVs can be effectively determined to assess treatment efficacy (Zhiteneva et al. 2020). In QMRA, to provide an estimate of the risk value with greater resolution, it is suggested to use probability distribution functions (PDFs). PDF, in turn, is a mathematical function that models the probability of different values in a dataset. It is used to describe the concentrations of pathogens in incoming water. While LRVs help identify necessary treatment steps, stochastic or probabilistic methods such as PDFs recognise the variability and uncertainty of the system and provide variable risk estimates (Zhiteneva et al. 2020). Statistical distributions such as log-normal (adopted by Bailey et al. (2020), Kongprajug et al. (2021), and Seis et al. (2022)), gamma (adopted by Masciopinto et al. (2020) and Seis et al. (2022)), and uniform (employed by Chhipi-Shrestha et al. (2019), Kongprajug et al. (2021), Zhiteneva et al. (2021), and Panagiotou et al. (2022)) are often used in water reuse risk assessment studies to describe the concentration of pathogens in source water. The choice of each distribution depends on the characteristics of the available data and the specific needs of the study (Zhiteneva et al. 2020).

In the last decade, the presence of Escherichia coli (E. coli) in reused water has been considered the most important indicator of faecal contamination due to its significant public health impact. It is an indicator easily detectable even at high dilution and is very sensitive to disinfection processes compared with others (Dingemans et al. 2020; Masciopinto et al. 2020; Lima et al. 2022). Seven articles highlighted the faecal contamination indicator E. coli in their studies, and agriculture is the sector where risk assessment is most frequently applied. Agriculture is the most requested application because it comprises the most common use of increasing water reuse practices, and it poses a risk that involves the environment, the health of workers, consumers, and the population living near irrigated areas. Norovirus (NV) was also chosen as a reference pathogen for risk assessment in some articles, notably in the studies by Schoen et al. (2020), Seis et al. (2022), Arden et al. (2020), and Zhiteneva et al. (2021). In Germany, NV infections are the leading cause of gastrointestinal infections in all age groups. The use of NV as a reference pathogen in QMRA was included in WHO guidelines in 2017. Nevertheless, NV data is rarely available over more extended surveillance periods to assess its variability (Seis et al. 2022). However, only Bailey et al. (2020) justified excluding the analysis of NV concentrations because no gene copies were detected in the water samples and the risk was significantly reduced. Nevertheless, it is pointed out that other pathogenic references, such as protozoa and bacteria, would also have to be considered for risk management (Seis et al. 2022). Kongprajug et al. (2021) used human-specific markers such as crAssphage and HF183 in QMRA because they are more abundant than waterborne pathogens. Regarding the assessment of human health risks from water reuse, disinfection by-products generated in water have not been widely studied. It could be of great benefit to identify a wide range of emerging contaminants and determine toxicity data and exposure parameters (Kongprajug et al. 2021).

In SqMRA, the E. coli parameter and the hazard quantification are determined by a direct quantification scale applied to a range of predicted concentrations corresponding to an applicable treatment level. A more advanced treatment level (more demanding) corresponds to lower E. coli concentrations and, consequently, a lower perception of risk (Rebelo et al. 2020; Damaceno et al. 2022; Lima et al. 2022).

In both methodologies, exposure assessment considers ingestion, inhalation, and adsorption as exposure routes that may occur through direct or indirect contact with different types of receptors. The receptors are the agents most at risk and are categorised as humans, animals (domestic or livestock), landscape vegetation, or crops (food or non-food). The receptors and their respective vulnerabilities always vary from project to project, as they depend on the type of water application for reuse. QMRA requires information on how the receptors will come into contact with the water, including volume, frequency of exposure, and duration of exposure (Kongprajug et al. 2021). Exposure scenarios are input data in both methodologies, where the main objective is to describe the potential exposure situations of the receptors in as much detail as possible and to assess probabilities (Rebelo et al. 2020; Damaceno et al. 2022; Lima et al. 2022). Besides evaluating the risks, this phase identifies the most critical aspects where alternative treatment scenarios can be tested and, therefore, which treatment train is the most recommended for the case study in question.

QMRA uses dose-response models to estimate the probability of infection. The selected studies used various dose-response models to assess the risk associated with exposure to pathogens. These models include exact beta Poisson, fractional Poisson, approximate beta Poisson, beta Poisson, exponential, and hypergeometric such as Schoen et al. (2020), Masciopinto et al. (2020), Kongprajug et al. (2021), Zhiteneva et al. (2021), Bailey et al. (2020), Chhipi-Shrestha et al. (2019), Seis et al. (2022), and Panagiotou et al. (2022). These models are selected based on the characteristics of the available data and the specific needs of each study. The selection of the appropriate dose-response model plays a critical role in accurately assessing risk and obtaining relevant results (Zhiteneva et al. 2020). When multiple dose-response models are used, evaluators are advised to perform a sensitivity analysis to determine how much the dose-response affects the final risk (Schoen et al. 2020; Kongprajug et al. 2021; Zhiteneva et al. 2021).

The choice of dose-response models varies depending on the specific pathogen under consideration. For example, the exponential model was more commonly used for Cryptosporidium (five studies) and Adenovirus (two studies), while the hypergeometric distribution was exclusively adopted for NV (two studies). When it comes to E. coli (three studies), which was frequently found in these studies, the beta Poisson model was most commonly used.

The risk characterisation in the SqMRA approach consists of quantifying and prioritising the risk to human health that results directly from the factors hazard, exposure pathways, exposure scenarios, and implemented multi-barriers. The risk for each receptor is calculated from the product between the hazard (Hz), the vulnerability of the receptors, and the associated damage. Once the exposure pathways and scenarios have been identified, it is essential to calculate the vulnerability, which is done by multiplying the importance factor of the exposure pathways by the importance factor of the exposure scenarios (both scored from 1 to 9) and then dividing by the normalisation factor (the maximum importance value applied). The following step is to determine the damage. For each exposure scenario, barrier types must be identified. Damage was determined based on the relationship between the probability of barrier failure and the severity of partial damage, as shown in the matrix presented in the Lima et al. (2022), Rebelo et al. (2020), and Damaceno et al. (2022) studies. The global risk is the ratio between the sum of each risk for each receptor and the number of receptors considered in the case study. Finally, taking the results into account, prioritisation is made by converting the global risk into a qualitative scale with three levels (despicable, acceptable, and unacceptable) (Rebelo et al. 2020; Damaceno et al. 2022; Lima et al. 2022).

On the other hand, in QMRA, during risk characterisation, all uncertainties and variables are taken into account. Uncertainties may be related to the model chosen, its parameters, the volume of water ingested, and the concentration of microorganisms, among others. In order to perform this analysis, simulation and modelling approaches have been used in most of the articles: Monte Carlo, Markov Chain Monte Carlo, and Bayesian network. Some of them were even used to combat the lack of data and uncertainties they presented, as in the case of Schoen et al. (2020), Arden et al. (2020), and Zhiteneva et al. (2021).

• Data scarcity and uncertainty: Challenges related to data scarcity and uncertainty persist.

• Stochastic approach with PDFs: The stochastic approach, utilising PDFs, has gained prominence in QMRA. This approach has been applied in studies by Seis et al. (2022), Kongprajug et al. (2021), Bailey et al. (2020), Masciopinto et al. (2020), Zhiteneva et al. (2021), Chhipi-Shrestha et al. (2019), and Panagiotou et al. (2022). It provides a more realistic consideration of uncertainties in water systems. However, the incorrect use of PDFs can lead to underestimation or overestimation of the final risk, with potential consequences in health, financial, and legal aspects.

• Use of multiple dose-response models: Incorporating various dose-response models can significantly enhance the resolution of final risk assessments, as emphasised by Schoen et al. (2020) and Zhiteneva et al. (2020).

• Tailored risk models: There is an emphasis on developing specific risk models tailored to different pathogens, including viruses, bacteria, and other contaminants. This approach leads to more precise risk assessments, as highlighted by Arden et al. (2020), Bailey et al. (2020), Schoen et al. (2020), Kongprajug et al. (2021), Zhiteneva et al. (2021), and Panagiotou et al. (2022).

• Integration of epidemiological data: The integration of epidemiological data, when available, allows for more accurate model calibration and informed risk assessments. Schoen et al. (2020) and Seis et al. (2022) have successfully integrated epidemiological data, resulting in improved overall assessment accuracy.

• Understanding pathogen concentrations: A comprehensive understanding of pathogen concentrations in source waters, including their variability (e.g., seasonality) and parameter uncertainty, forms the basis for tailoring risk reduction measures to local conditions.

• MCDA tool applications: Arden et al. (2020) have suggested potential applications of their findings within an MCDA tool. In addition to QMRA, Arden et al. (2020) also applied LCA and LCC in their study, highlighting the importance of integrating these methodologies for a comprehensive assessment of water reuse projects. Similarly, Kanchanapiya & Tantisattayakul (2023) applied MCDA in their case study alongside QMRA, which will be explained further in the text. They also emphasised the importance of integrating different evaluation criteria within an MCDA framework.

• Incorporating economic considerations: Dingemans et al. (2020) have stressed the importance of incorporating economic considerations into governance arrangements for water reuse cases. In addition to risk assessment, Arden et al. (2020) and Chhipi-Shrestha et al. (2019) conducted a comprehensive economic analysis.

• Pathogen decay rates and exposure factors: Kongprajug et al. (2021) have highlighted the need for future research to investigate pathogen decay rates under local conditions and to obtain exposure factor values, including volume and frequency of exposure.

• Risk calculation for various risk levels and exploration of equivalent barriers: Lima et al. (2022), Rebelo et al. (2020), and Damaceno et al. (2022) employed SqMRA to calculate risk values for different risk levels, considering various scenarios. They also explored how implementing equivalent barriers could further reduce risk. This approach effectively functions as a ‘reiteration’ within the risk assessment process, allowing for the exploration of different risk scenarios and the evaluation of risk mitigation measures.

• Easy-to-implement and transparent process: SqMRA is regarded as an easy-to-implement approach suitable for evaluating various water management options. It is a transparent process that incorporates concepts such as risk, receptor vulnerability, and potential damage, promoting social acceptance of water treatment measures.

• Project-specific application and adaptability: SqMRA should be applied to each project individually, recognising that exposure levels may vary based on local characteristics and the operational conditions of irrigation, capture, and storage. SqMRA's ability to calculate risk values for different risk levels and explore the impact of equivalent barriers demonstrates its adaptability and potential for ongoing risk management and refinement as projects progress and more information becomes available.

Methodologies of QMRA and SqMRA are examined in this paper as applied to water reuse. For more details on the strengths and limitations of each approach, see Table 1 in Section 4.

LCA is an analytical tool that allows quantification of the environmental impacts that occur throughout the life cycle of a product or service (Arden et al. 2020; Rodríguez et al. 2021; Gómez-Monsalve et al. 2022; Kalboussi et al. 2022; Tampubolon et al. 2022; Negi & Chandel 2024). Through careful modelling, each phase (production, use, and end-of-life) is examined, taking into account the extraction and transportation of raw materials, the manufacturing process, distribution, use and maintenance, reuse, and waste treatment (Azeb et al. 2020; Rodríguez et al. 2021; Gómez-Monsalve et al. 2022). According to Gómez-Monsalve et al. (2022) and Crovella et al. (2024), LCA is a tool for understanding and quantifying the environmental impacts of different water bodies, contributing to conservation strategies, assisting in decision-making to alleviate water scarcity, and supporting the development of more sustainable options. On the other hand, Maeseele & Roux (2021) defined the LCA as a holistic tool to evaluate the efficiency of water reuse, including the impact of water depletion (WD), type of treatment technology, and energy consumption. In most cases, wastewater treatment decisions are mainly influenced by direct capital and operating costs, assuming that the project meets the required standards, while LCCs and environmental impacts are rarely considered. As per Foglia et al. (2021), Rodríguez et al. (2021), and Torre et al. (2024), the life cycle perspective can help achieve sustainable wastewater treatment, as it is a method for evaluating the environmental sustainability of treatment processes.

Unlike the other studies, Chhipi-Shrestha et al. (2019) did not apply the LCA decision support tool. They only evaluated three environmental indicators, namely: fresh water saving, energy use, and carbon emissions.

LCA is described in the ISO standards 14040 and 14044 (Azeb et al. 2020). This methodology includes four steps (Arden et al. 2020; Azeb et al. 2020; Boysen et al. 2020; Yoonus & Al-Ghamdi 2020; Foglia et al. 2021; Maeseele & Roux 2021; Kalboussi et al. 2022; Tampubolon et al. 2022; Szalkowska & Zubrowska-Sudol 2023; Hargitai et al. 2024; Torre et al. 2024).

Goal and scope definition: This is the first phase of the LCA methodology application. Here, the objectives, their application, and all hypotheses considered in this study are determined (Yoonus & Al-Ghamdi 2020). The system boundaries and the functional unit also are defined in this phase. The functional unit is the product or service unit whose environmental impact is to be evaluated or compared. It is usually expressed by the quantity of a particular product. The main purpose of the functional unit is to provide a reference to which inputs and outputs are linked, and it should be selected on the basis of the study (Ahmed 2010). In water reuse projects, the functional unit varies depending on the specific treatment and application scenario, such as the volume of treated water reused, irrigated area with treated water, or quantity of treated water for each specific use. For example, Tarpani & Azapagic (2023) defined the functional unit as the treatment of 1,000 m³ of secondary effluent using advanced wastewater treatment techniques and the treatment of 1,000 kg of dry matter for sludge treatment options. This approach ensures that the environmental impacts associated with both the water treatment and sludge handling processes are adequately accounted for in the assessment. Foglia et al. (2021) considered 1 m³ of treated wastewater, and Kalboussi et al. (2022) considered 1 ha of vineyard. The process of defining system boundaries should be linked to the main objective of the study. System boundaries are defined boundaries between a system and its environment or between two different systems and are determined by resource consumption and emissions to air, water, and solid waste. If data is lacking, it can be obtained from literature, estimates, or mathematical models. It is recommended that system boundaries be defined by graphical elements, such as a process tree or diagram, to provide a better understanding of the system overview, as was the case for all selected articles (Ahmed 2010). It included not only the line of water treatment processes but also their demand and nutrients from the environment of the irrigation area. For example, Tarpani & Azapagic (2023) considered advanced wastewater treatment techniques and sludge management processes. Foglia et al. (2021) included pre-primary anaerobic treatments, disinfection, and filtration in their boundary considerations. Kalboussi et al. (2022) defined their boundaries as removing large floating particles and coarse solids, encompassing the entire process from wastewater treatment to crop irrigation. Furthermore, Gómez-Monsalve et al. (2022) incorporated the operational phase of water treatment plants (WTPs) and wastewater treatment plants (WWTPs), emphasising the ongoing environmental impacts associated with these facilities' daily operations. In contrast, Boysen et al. (2020) assessed the environmental impacts of both the construction and operational phases, recognising the significance of each phase in the overall life cycle assessment.

Life cycle inventory (LCI): Inventory analysis is a technical process that collects data to quantify the inputs and outputs of the system (Azeb et al. 2020; Yoonus & Al-Ghamdi 2020). Inputs are the materials, chemicals, and energy that go into wastewater treatment technologies, including construction, operation, and maintenance. The outputs are the environmental emissions, particularly towards air, water, soil, and solid waste (Cabling et al. 2020; Rodríguez et al. 2021). To facilitate this analysis, the system under study is divided into several process subsystems, which are classified into different categories. For example, Arden et al. (2020) developed LCIs for each treatment configuration. Each system includes pre-treatment, biological treatment, and disinfection unit processes. Gómez-Monsalve et al. (2022) considered initial environmental impacts related to materials used in manufacturing and transportation components, operational impacts related to energy consumed during the processes' operation, and maintenance impacts related to materials used to maintain the systems, including manufacturing and transportation. Not only are all process inputs and outputs identified, but they are also quantified. According to Azeb et al. (2020) and Ahmed (2010), this is a fundamental process for the subsequent phase.

Life cycle impact assessment (LCIA): In the context of water reuse projects, LCIA involves translating the results obtained from LCI into potential environmental impacts, tailored to the specific objectives and methodologies employed (Ahmed 2010; Azeb et al. 2020; Yoonus & Al-Ghamdi 2020). For instance, Maeseele & Roux (2021), Foglia et al. (2021), Gómez-Monsalve et al. (2022), Tampubolon et al. (2022), Szalkowska & Zubrowska-Sudol (2023), and Hargitai et al. (2024) applied the ReCiPe method. ReCiPe was the most employed method among the reviewed studies, which is particularly suitable for assessing environmental impacts in Europe, utilising data from countries such as Switzerland, the Netherlands, and Germany. This method categorises impact into human health, ecosystem quality, and resources (Tampubolon et al. 2022). Conversely, Szalkowska & Zubrowska-Sudol (2023), Negi & Chandel (2024), and Torre et al. (2024) applied the CML (Centrum voor Milieukunde Leiden) method, which focuses on specific impact categories such as acidification, eutrophication, and human toxicity (HT). Rodríguez et al. (2021) and Cabling et al. (2020) used the TRACI (tool for reduction and assessment of chemicals and other environmental impacts) method, developed by the U.S. Environmental Protection Agency, which covers categories like climate change, acidification, and human health impacts. Pinelli et al. (2020) and Kalboussi et al. (2022) applied the International Reference Life Cycle Data System (ILCD) method, which provides a set of recommended impact categories for LCA studies in Europe.

In practical application, software tools have been crucial in facilitating transparency and traceability in impact assessment methodologies. OpenLCA was the most applied software, utilised in five studies, namely by Cabling et al. (2020), Tampubolon et al. (2022), Rodríguez et al. (2021), Szalkowska & Zubrowska-Sudol (2023), and Negi & Chandel (2024). OpenLCA is valued for its open-source platform, which offers flexibility and extensive databases for comprehensive environmental assessments. This was followed by SimaPro, used in four studies, as highlighted by Maeseele & Roux (2021), Pinelli et al. (2020), Kalboussi et al. (2022), and Torre et al. (2024). According to Kalboussi et al. (2022) and Gómez-Monsalve et al. (2022), SimaPro enhances the integration of inventory data with environmental impact assessments, using midpoint and endpoint indicators to quantify and characterise environmental impacts effectively.

Freshwater eutrophication (FEU) and HT were the most requested indicators, each being applied in 11 studies. According to Suresh et al. (2023), FEU is a major global concern, caused by excessive nutrient loadings (nitrogen and phosphorus) from human activities and likely exacerbated by climate change. HT is critical due to the potential exposure of humans to harmful chemicals and pathogens present in treated wastewater, making it essential to assess the safety and health risks associated with water reuse. Climate change (CC) was also a popular indicator, being applied in nine studies. It has become very popular due to the current state of the world, and it is, therefore, widely used in LCA's impact assessment methodology. Additionally, the GWP (global warming potential) indicator was specifically mentioned in eight studies, likely due to the increasing focus on mitigating climate change impacts through sustainable practices. Fossil fuel depletion (FD) was also heavily used and applied in seven studies, having been identified as one of the biggest challenges in recent years due to being limited and non-renewable on a human scale. The problem of pollution is also a recurrent challenge, which justifies the fact that the indicator of marine eutrophication (MEU) appeared in eight studies (Rodríguez et al. 2021). Although the energy indicator was not regularly requested, it significantly impacted environmental performance in LCA and was applied in only one study. Environmental impacts are often higher in countries where fossil fuel combustion is the main electricity generator.

Interpretation of results: This is the final phase of the LCA methodology, where the obtained results are presented in a concise manner, presenting the most critical impacts and options to reduce them (Ahmed 2010).

• Inconsistencies in data presentation: Crovella et al. (2024) conducted a systematic literature review and concluded that there is significant inconsistency in the presentation of data in the reviewed papers. Normalised data, percentage values, and point data often cannot be easily compared, leading to substantial uncertainty. The authors encountered difficulties in extrapolating information and results due to the varied presentation formats, such as percentages, mean values, or histograms without specific point indications.

• Environmental impact reduction: Using the LCA methodology, it is possible to determine the impacts of a water system and take measures to minimise them. According to Foglia et al. (2021), over 96% of impacts on FEU, WD, and HT are caused by direct emissions impacts. Conversely, the results from Boysen et al. (2020) show that overall impacts in the CH decrease when reuse is implemented.

• Selection of appropriate treatment technologies: More efficient and environmentally friendly technologies can be identified through LCA. For example, according to Foglia et al. (2021), the most affected impacts are related to primary treatment and biological processes, specifically in CH, FD, mineral resource depletion, and terrestrial ecotoxicity (TEC). Disinfection particularly affects the FD and CH categories. The choice of water reuse treatment technologies can affect LCA results. Some technologies may be more energy and resource-efficient than others, and LCA can help identify the best options. Arden et al. (2020) concluded that scenario 3 (described in Supplementary Table S2) performs well on many of the indicators, including cost, while other scenarios face challenges due to the need for additional pre-treatment (scenario 1), post-treatment (scenario 2), and disinfection processes (scenario 1). Tarpani & Azapagic (2023) concluded that nanofiltration (scenario 2) emerged as the most sustainable option across various operational parameters and sustainability criteria. In contrast, solar photo-Fenton (scenario 3) treatment was considered the least sustainable option.

• Greenhouse gas (GHG) emissions reduction: LCA contributes to reducing the energy required for water treatment and transportation. According to Foglia et al. (2021), CH, FD, and TEC are mainly affected by electricity consumption and transportation. Lower energy consumption leads to reduced GHG emissions. LCA also highlights the reduction of long-distance water transport, saving energy and resources.

• Water resource conservation: LCA has shown that water reuse is an effective strategy for conserving freshwater, especially in regions with water scarcity. As per Foglia et al. (2021), WD is strongly influenced by direct water withdrawals from the environment. The most significant benefit observed in this category results from substituting water reuse. Conversely, Kalboussi et al. (2022) concluded that, among the remaining scenarios, defined reuse is the worst option in terms of ozone depletion, terrestrial acidification, and ionising radiation. However, it reduces CH, eutrophication, and ecotoxicity since it also avoids the use of fertilisers. Therefore, according to Kalboussi et al. (2022), Foglia et al. (2021), and Azeb et al. (2020), water reuse reduces the need for synthetic N and P fertilisers. It minimises the environmental impact of the energy required to produce them since fertiliser use is the most significant contributor in most impact categories.

• Indirect positive effects: LCA results reveal positive indirect effects, such as reduced contamination of water sources, resulting in less wastewater discharged into the environment and potentially reducing pollution of rivers, lakes, and oceans. According to Boysen et al. (2020), avoiding wastewater discharge into rivers reduces the FEU impact, unlike MEU, which has a significant impact on river discharge in a non-reuse scenario.

• Integration with other sustainability approaches: LCA is often integrated with other tools such as LCC and MRA. For instance, as observed in the previous section, Arden et al. (2020) applied QMRA, LCA, and LCC approaches in their study of non-potable water reuse systems. This integration allows for a comprehensive assessment of sustainability factors beyond environmental impacts, including economic and decision-making aspects. Torre et al. (2024) concluded that integrating WWTP simulation tools, life cycle methodologies, and multi-criteria analysis represents a promising approach to develop a comprehensive decision-making framework for urban wastewater systems worldwide.

Cost analysis was performed in only seven articles, as identified in the studies of Chhipi-Shrestha et al. (2019), Foglia et al. (2021), Pinelli et al. (2020), Arden et al. (2020), Boysen et al. (2020), Tarpani & Azapagic (2023), and Torre et al. (2024). According to Chhipi-Shrestha et al. (2019) and Tarpani & Azapagic (2023), LCC is a method of economic analysis that considers all the costs of owning, operating, maintaining, and disposing of a project or product. It is a method estimated as the sum of the capital, operating, repair, and replacement costs of water supply infrastructure. LCC is commonly utilised to assess the economic viability of water supply infrastructure and has been applied by Chhipi-Shrestha et al. (2019), Arden et al. (2020), Foglia et al. (2021), Tarpani & Azapagic (2023), and Torre et al. (2024). Although Pinelli et al. (2020) and Boysen et al. (2020) conducted a cost analysis, they did not use the LCC methodology, so they are not discussed in this review.

The LCA and LCC methods can be used synergistically to assess the cost-effectiveness of a project or investment over its entire life cycle. Chhipi-Shrestha et al. (2019), Arden et al. (2020), and Torre et al. (2024) applied LCC using the net present value (NPV) method of the National Institute of Standards and Technology. The NPV equation calculates the present value of future cash flows, is used to assess economic feasibility, and is linked to the LCC estimation process. Tarpani & Azapagic (2023) did not apply NPV but conducted an analysis of economic indicators using LCC, which encompassed construction costs, infrastructure replacements, operating expenses, waste management costs, and transportation costs. Additionally, revenue generated from products derived from sludge treatment was also considered. On the other hand, Foglia et al. (2021) employed a simple formula to calculate the annual cost of an operation or project by adding the annual operating costs (OPEX) and the annual capital costs (CAPEX).

The total capital cost of each treatment process is the sum of the unit cost of each process's direct costs and indirect costs. Unit process costs include capital expenditures for equipment and installation costs. Direct costs represent the necessary costs of each major wastewater process, while indirect costs include additional expenses. Total annual costs are the sum of operations and maintenance. Equipment replacement costs are included in material costs and take into account the life of each component of the system (Arden et al. 2020; Foglia et al. 2021). In an economic context, regardless of the financial tool, it is crucial to consider the different reuse alternatives for each stakeholder. The mutual agreement acts as a way to find a sustainable solution for all stakeholders (Chhipi-Shrestha et al. 2019).

• Environmental considerations in LCC: LCC, primarily focusing on economic aspects, can also incorporate environmental impacts when combined with LCA, as demonstrated in the studies by Chhipi-Shrestha et al. (2019), Arden et al. (2020), Foglia et al. (2021), Tarpani & Azapagic (2023), and Torre et al. (2024).

• Scenario evaluation: Similar to LCA, LCC allows the evaluation of various project scenarios, including different treatment technologies, implementation scales, and operating strategies, as illustrated in Supplementary Table S2 (the same scenarios used in LCA). The objective is to identify the most cost-effective option for water reuse projects.

• Total cost assessment: LCC provides a comprehensive assessment of the total costs associated with implementing and operating water reuse systems. This includes expenses related to water treatment to meet the required quality standards. Maintenance costs (considered in the aforementioned studies) and replacement of equipment and system components (as studied by Foglia et al. (2021) and Tarpani & Azapagic (2023)) are essential components of this evaluation. The durability of these elements is also a crucial factor.

• Consideration of variables: In LCC, it is important to consider variations in energy costs and inflation rates. However, it is noteworthy that some studies, such as Chhipi-Shrestha et al. (2019) and Foglia et al. (2021), did not consider these factors in their analyses.

Arden et al. (2020) applied the DST4WR QMRA, LCA, and LCC, noting that their results could be used as input to MCDA (a decision support tool presented below). In contrast, Tarpani & Azapagic (2023) and Torre et al. (2024) integrated LCA and LCC into MCDA.

Methodologies of LCA and LCC are examined in this paper as applied to water reuse. For more details on the strengths and limitations of each approach, see Table 1 in Section 4.

MCDA is the least common method used in water reuse. In the subsequent discussion, research articles using this decision support tool for water resources management and water reuse implementation in predefined case studies are analysed.

MCDA is a useful approach that can incorporate a combination of quantitative and qualitative information, considering different stakeholders' preferences. It is a methodology to support the decision-making process for water reuse measures (Isaac et al. 2022; Torre et al. 2024). The design of a water reuse project is a complex process, and it is necessary to define the criteria and weigh them using a comparative scale. This underlines the need for an MCDA to help in the planning and efficient management of the specific project. In short, this is a complex tool often used to evaluate the performance of the criteria chosen by decision-makers and to select the best alternative (Chhipi-Shrestha et al. 2019; Goyal & Kumar 2020).

MCDA begins with the definition of criteria and objectives. This can include criteria such as cost, risk analysis, and environmental impact, among others, as mentioned in all the selected studies. It is essential to define the objectives and goals of the project in question, in order to prioritise them according to their importance. It is noteworthy that the studies that used MCDA to evaluate water reuse projects demonstrated an adaptive approach to the selection of criteria and sub-criteria. This flexibility reflects the inherent complexity of water reuse projects, as requirements and priorities can significantly vary depending on the specific context of each study.

The next step is to proceed with data collection. Any analysis is usually preceded by a characterisation of the area to be studied, a review of the effluent quality, a definition of the potential end users or stakeholders, and a definition of the criteria as stated above (Chhipi-Shrestha et al. 2019; Dehaghi & Khoshfetrat 2020; Goyal & Kumar 2020; Diogo et al. 2021; Isaac et al. 2022). In addition, it is important to identify the different alternatives or scenarios considered in each project. These may include different technologies, treatment strategies, and water sources, among others. It is important to choose the MCDA approach, and each study selected applied the one that was most appropriate for each project, as it will be discussed later. To implement the MCDA approach, the following steps are required (Chhipi-Shrestha et al. 2019; Dehaghi & Khoshfetrat 2020; Goyal & Kumar 2020; Diogo et al. 2021; Isaac et al. 2022):

• Normalise the data so that they are all on the same scale.

• Weight the criteria based on their importance, with weights assigned by decision-makers.

• Assign scores to each alternative/scenario with respect to each criterion.

• Aggregate the scores to create an overall score for each alternative.

• Select the alternative that best meets the established criteria.

Isaac et al. (2022) applied cooperative game theory (CGT) and compromise programming (CP) methods simultaneously. They are methods based on multi-objective mathematical programming and use the concept of metric distance. Isaac et al. (2022) surveyed some industries that were in the proximity of the WWTP under study, which could be possible users. These industries served as alternatives for the application of the adopted methodologies. The CP method aims to determine the ideal solution associated with each decision-making unit. The best solution is defined as the point closest to the ideal point relative to the other alternatives by a distance measure. This method aims to minimise the distance of feasible points from an objective point, chosen by the decision maker and called the ‘ideal solution’ (Isaac et al. 2022). CGT is also a distance-based approach: unlike the previous method, the optimal solution is the one that maximises the distance.

Chhipi-Shrestha et al. (2019) applied the game theory to a selection of water reuse applications in a municipal context, targeting three stakeholders: the municipality, citizens, and farm operators. Game theory problems are determined by identifying specific subsets of outcomes called solution concepts. Chhipi-Shrestha et al. (2019) used the solution concept of Pareto optimisation. Pareto optimisation is the best solution in a cooperative game where improving one particular alternative without making another worse is impossible. This method can be applied to solve a plethora of challenges, including socioeconomic conflicts, design and planning, and political problems, even using qualitative information.

Goyal & Kumar (2020) and Torre et al. (2024) used the analytical hierarchical process (AHP) method to select criteria and valuable sub-criteria for planning and implementing water reuse systems. AHP is a method proposed by Saaty that uses hierarchical structures, matrices, and linear algebra to formalise the decision-making process. It forces decision-makers to carefully evaluate the importance of each criterion in relation to the others in a hierarchical manner.

Dehaghi & Khoshfetrat (2020) combined the TODIM (Tomada de Decisão Iterativa Multicritério) method (an acronym for interactive multi-criteria decision making) with the goal programming (GP) method considering the Leopold matrix. These methods were applied to different water application alternatives to optimise them considering the risk criteria set by the decision-makers in a given case study. The TODIM method allows to consider the subjectivity of the decision-makers according to the perspective theory. Sensitivity analyses can be performed to apply different attenuation factors, and the effects on the weighs of the alternatives can be formulated with particular operations. GP is a prevalent mathematical method for dealing with multi-objective programming. It has been applied considering the Leopold matrix to perform an environmental impact assessment. The objective is to eliminate alternatives until a satisfactory level of performance is achieved for each criterion. The constraints and decision variables are formulated in conventional linear programming (Dehaghi & Khoshfetrat 2020).

Diogo et al. (2021) implemented a scoring on a conventional scale, a comparison between criteria and, ultimately, a weighted sum method (WSM), written and programmed in a spreadsheet. This method was applied to all options of alternative water sources in a large tourist complex. For the sake of simplicity, Diogo et al. (2021) decided to use a WSM.

Tarpani & Azapagic (2023) applied the multi-attribute value theory (MAVT), which is a decision-making approach that allows decision-makers to evaluate and compare alternatives based on multiple criteria. It involves structuring a set of relevant decision criteria, assigning weights to these criteria based on their perceived importance, and evaluating alternatives against these criteria and weights. MAVT helps formalise the decision-making process by systematically considering the decision-maker's preferences. MAVT was used to assess and compare different options for advanced wastewater and sludge treatment.

Kanchanapiya & Tantisattayakul (2023) applied the PAPRIKA method, which is a comparative analysis designed to facilitate decision-making based on preferences and priorities. It enables the comparison of all possible pairs of alternatives using ranking and preferences to determine the best option among them. PAPRIKA is commonly used in MCDA, where several alternatives need to be evaluated against multiple criteria. They explored various water reuse options across three main activity groups: domestic reuse, agricultural reuse (specifically for planting coconut, Phuket's economic crops), and using effluent water as raw water for existing WTP.

• As mentioned earlier, preparing a reuse project is a complex process for which these eight articles have resorted to, conducting multi-criteria analysis to achieve an efficient management of such projects. It is a process that requires the consideration of multiple criteria, where more than a simple analysis of costs is required. This is a difficult stage, as it requires significant clarity about the scope under study and the assessment context to easily understand which criteria should be included and excluded in the methodology. Therefore, reviewing the sustainability criteria used in different case studies within the same application domain is essential.

• MCDA identifies the alternative that best meets the criteria and objectives established for a water reuse project. This helps make an informed decision about which approach or technology to implement. Torre et al. (2024) concluded that examining diverse decision-making scenarios broadens the perspective and enhances the understanding of potential wastewater treatment options. Chhipi-Shrestha et al. (2019) show that alternative 5 (described in Supplementary Table S3) is the optimal solution, with benefit sharing between the municipality and the citizens. In addition, MCDA shows how changes in criteria weights or input data affect the final decisions, as was the case in the study by Chhipi-Shrestha et al. (2019), which influenced Pareto optimisation.

• MCDA reveals which criteria are considered most important by decision-makers. This is useful for understanding priorities and focusing project efforts. The results of the questionnaire conducted by Isaac et al. (2022) showed that effluent quality stands out as one of the most relevant sub-criteria due to using it to avoid damage to health and to the production process. Another sub-criteria, also related to the previous one, is risk. It was also one of the most crucial sub-criteria in the context of human exposure and the care required in handling the water. Reliability was also considered important by the evaluators, given that ensuring the quality standard is essential. The sub-criteria of the need for post-treatment was also considered vital as it may impact the costs required to carry out this practice, depending on the purpose of the reuse (Isaac et al. 2022). Kanchanapiya & Tantisattayakul (2023) concluded that in their study, the investment cost indicator was identified as the most significant, followed by electricity costs and willingness to pay, while the nutrient load to demand ratio indicator held the least weight in their assessment.

• MCDA can reveal that an alternative is more cost-efficient but has a higher environmental impact. This helps make informed decisions about how to balance these trade-offs. This information is directly related to the importance that decision-makers assign to each criterion. For example, the results of the questionnaire conducted by Goyal & Kumar (2020) showed that, according to academia, technical criteria are the most important and economic criteria are the least important. This contrasts with the industrial perspective, which considers the economic criteria to be the most important and the environmental criteria to be the least important. Finally, for public utilities, social and legal criteria are the most and the least important, respectively. Public health and risk are top priority sub-criteria and are consistently mentioned by the three identities (Goyal & Kumar 2020). The results of these questionnaires determine the importance of the different criteria and sub-criteria in a reuse project, according to the different observations that contribute to effective planning and management at many different levels.

As can be concluded, the MCDA decision support tool can incorporate multiple criteria related to risk assessment, social acceptability, costs and benefits, environmental impact, as well as governmental and local regulations, ensuring compliance with applicable laws. More importantly, it assesses the degree of sustainability of the alternatives, taking into account the above criteria and other equally important criteria identified in the selected studies.

Water reuse is a key strategy for addressing the growing challenges of water scarcity and the need to ensure safe and sustainable water supplies for communities. However, the effective evaluation and implementation of water reuse projects requires careful analysis, taking into account several complex factors ranging from microbiological risks (MRA), environmental impacts (LCA), and financial costs (LCC). Proper planning plays a key role in this process, ensuring that the project runs smoothly and that operations remain less complex. DST4WR play a key role in the planning phase (Ddiba et al. 2023) so that stakeholders can consider and evaluate the variables of interest in an MCDA, providing a complete and global understanding.

The use of DST4WR in water management is crucial to ensure the efficiency and sustainability of water reuse systems (Ddiba et al. 2023). According to Rupiper & Loge (2019) and Tortajada (2020), the main challenges in the application of water reuse (potable or non-potable) are lack of local regulation, high costs, social acceptance, environmental and health risk concerns, and lack of support for technical programme development, management, and financial viability of water reuse systems.

As discussed in previous sections of this article, various DST4WR have been used to address the aforementioned challenges in water reuse projects. However, each of these tools has its own strengths and limitations. Table 1 provides a comprehensive analysis of water reuse assessment tools: QMRA, SqMRA, LCA, and LCC, highlighting their strengths and limitations.

Integrating these methodologies into an MCDA not only provides a more comprehensive view of the challenges associated with water reuse but also offers a unique opportunity to overcome the limitations inherent in each approach. MCDA can integrate different criteria – such as public health, social, economic, environmental, and technical aspects – to provide a comprehensive assessment of water reuse systems. As shown in Table 1, the strengths and limitations of each methodology indicate the need for an integrated approach in each specific case. This ensures the specificity of the application of safe practices in relation to public health, environmental quality, availability of water resources, and efficient use of public financial resources, resulting in an integrated assessment of the viability of the water reuse system. In addition to the limitations already mentioned in Table 1, the study of decision support methods covered in this article (MRA, LCA/LCC, and MCDA) reveals several significant gaps that indicate needs for future research: (1) Generally, these methods often involve subjectivity that could be mitigated by more specific criteria and further scientific advancements. (2) In the case of MRA, the lack of specific databases for microbiological contamination represents a significant limitation in its practical application. (3) Regarding LCA and LCC, the scarcity of studies focused on water reuse limits their applicability and accuracy in this context. It is essential to expand research efforts to enhance their utility in environmental and economic assessments of water reuse projects. (4) Incomplete and unavailable inventory data often hinder accurate LCA analysis. (5) Insights into the life cycle impacts and potential trade-offs of emerging water reuse practices are limited by the lack of systematic evaluations across different implementations. Crovella et al. (2024) highlighted the paucity of systematic reviews in this field as a significant gap. This underscores the importance of our current article, which aims to address this gap by providing a comprehensive systematic review. We advocate for further research to compare the influence of different crops on wastewater quality and quantity, as well as the required treatments. Such investigations are essential for advancing research and improving inventory and impact datasets in the field. (6) The specific application of MCDA methods in water reuse projects is still not widely explored, highlighting the need to develop more tailored and adaptable approaches. (7) Torre et al. (2024) noted that while MCDA methods offer a comprehensive view, they can introduce biases when assigning criteria. Environmental and social indicators are complex and varied, requiring a more detailed breakdown to capture all relevant aspects accurately. Additionally, wastewater treatment parameters can change over time due to emerging contaminants, which necessitate new treatment approaches. This evolving nature of contaminants means that the study's findings and methodology need constant updating and reassessment to remain effective. These identified gaps underscore the importance of future research focused on improving the precision, robustness, and applicability of these decision support tools in water reuse contexts. This study aims to address these issues through the development of a more integrative and robust MCDA methodology, named DST4WR, thereby contributing to advancing the field of sustainable and scientifically grounded water reuse.

As mentioned by Ahmed (2010), although DST4WR offer significant benefits, none of them can solve all the challenges associated with managing a water reuse project in isolation. The integration of the tools mentioned in this review complements various criteria and information in decision-making in water reuse systems, promoting effective and comprehensive management.

It is recommended to use specific DST4WR based on the application, scenario, or case study:

• 1. Microbiological risk assessment (MRA): This tool should be prioritised in scenarios where public health risks are a primary concern, such as in potable water reuse or agricultural irrigation. MRA helps assess and manage the safety risks associated with pathogens and microorganisms, ensuring the protection of public health. It is often mandated in regulatory frameworks, as seen in European regulations for agricultural irrigation and in Portugal for other applications based on Decree-Law No. 119/2019 (Portugal 2019).

• 2. Life cycle analysis (LCA): LCA is crucial for evaluating the environmental impacts of water reuse projects. This tool should be used when the primary goal is to understand and mitigate the environmental footprint of water reuse systems. LCA helps quantify the impacts on resources, energy use, and emissions, supporting the identification of sustainable practices and the minimisation of negative environmental outcomes.

• 3. Life cycle cost (LCC): For projects where financial feasibility is a significant concern, LCC should be employed to assess the total costs over the project's lifespan. This tool is essential in scenarios where budget constraints or economic efficiency are critical factors, helping decision-makers understand the long-term financial implications and optimise cost management.

• 4. Multi-criteria decision analysis (MCDA): MCDA is recommended when a comprehensive evaluation is needed, considering multiple criteria simultaneously, including environmental, economic, social, and technical aspects. This tool integrates the results from MRA, LCA, and LCC, providing a balanced assessment and facilitating informed decision-making. MCDA is particularly useful in complex scenarios where trade-offs between different factors need to be carefully weighed.

In practice, the following sequence is suggested for a thorough evaluation of water reuse projects:

• Step 1: MRA to ensure public health safety.

• Step 2: LCA to assess and mitigate environmental impacts.

• Step 3: LCC to evaluate the economic feasibility and optimise costs.

• Step 4: MCDA to integrate the findings from MRA, LCA, and LCC, and provide a comprehensive and balanced decision-making framework.

By following these guidelines, stakeholders can ensure that water reuse projects are evaluated and implemented in a manner that addresses health, environmental, and economic concerns, leading to more sustainable and accepted water reuse solutions. In summary, combining these DST4WR is not only a wise choice, but a necessity to better address the complex challenges of water reuse. This integrated approach is an important step towards building more efficient, sustainable, and beneficial water reuse systems for society and the environment.

Based on the analysis of the DST4WR presented in this study, it is clear that water reuse requires a multifaceted and integrated approach. The selected articles mainly used QMRA and SqMRA, the former focusing on water supply while the latter is applied to water reuse. However, some studies have applied QMRA to water reuse, even if not for potable reuse. These are methodologies that play a crucial role in the assessment of water reuse, with a focus on public health and the management of associated risks. LCA allowed for the identification of the most relevant indicators for the assessment of water reuse: CH, FEU, and FD. These indicators were the most analysed due to being directly related to the most significant and urgent environmental impacts we currently face, such as climate change, resource scarcity, and water quality degradation. In addition, analysis of these indicators can provide vital information for decision-making on water management and other human activities. This allows for identifying more environmentally and economically sustainable options as we move forward with the inclusion of LCC. LCC allowed a comprehensive analysis of the financial costs associated with water reuse projects. This joint approach with LCA not only considers aspects of environmental impact but also incorporates essential economic considerations. The integration of these methodologies in an MCDA is a major step forward in water reuse management. MCDA allows multiple criteria to be assessed simultaneously, including public health, social, environmental, economic, and technical aspects. Each selected study defined its own evaluation criteria based on its specific case, emphasising the need to adapt the approach to each situation.

To effectively integrate MRA, LCA, and LCC into an MCDA, establishing a framework that allows for the simultaneous assessment of multiple criteria would be beneficial. This framework could encompass public health, social, environmental, economic, and technical aspects relevant to each specific case study. It would also be advantageous for each study to define its evaluation criteria based on its unique circumstances, highlighting the importance of adapting the approach to local conditions.

The combined use of these DST4WR has the potential to facilitate the identification of synergies and trade-offs among the assessed criteria. This approach could significantly enhance decision-making in water reuse management, enabling the selection of sustainable and beneficial options for society and the environment.

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.

This research was supported by the Doctoral Grant UI/BD/153586/2022, attributed to the first author, under the Collaboration Protocol for Funding the Multi-Annual Research Grant Plan for Doctoral Students, established between FCT - Fundação para a Ciência e a Tecnologia, IP, and the R&D Unit Centre for Territory, Environment and Construction (4047), and funded by FCT and NORTE2020 - North Regional Operational Programme.