Emergency water supply during disasters in Iran: a multi-criteria group decision analysis

The emergency events database (EM-DAT) highlights a concerning global trend: an increase in the frequency and severity of natural disasters (EMDAT 2009; Loo et al. 2012). Disasters arise when hazards result in significant physical losses, damages, and socio-economic disruptions, posing direct or indirect risks to lives (Davis & Lambert 2002; Loo et al. 2012). The sudden and unexpected nature of these events severely limits swift and effective responses from individuals and organizations, exacerbating their catastrophic consequences.

Since humans can only survive for 3–6 days without water, the importance of having a reliable source of water increases under emergency conditions. One of the most critical and prioritized tasks after the occurrence of a disaster is providing safe and sufficient drinking water (Loo et al. 2012). During emergency conditions, access to fresh water may be restricted due to the following reasons (Bakos et al. 2011): (1) potential damages, failures, or contamination of fresh water supply sources; (2) damage to the water treatment facilities or potential failure and disturbance in their regular operations; (3) occurring potential damages, failures, or pollutions to the water distribution systems; (4) buildings' destructions and failures, which could result in breaking in water pipelines; (5) potential damages to the sanitation services; (6) potential break in the wastewater systems and drinking water contamination; and (7) residential areas pollution by permeation of wastewater.

Given these critical emergencies, appropriate planning, management, and provision of reliable water supply are of utmost importance once a disaster occurs. However, in the last decade, there have been limited efforts to optimally manage the water supply during emergency conditions.

Toland et al. (2023) developed a framework for assessing community vulnerability to estimate emergency food and water needs in developed countries facing geophysical hazards. They applied this framework using a simplified model to evaluate resource needs during the ‘ShakeOut’ scenario, a major earthquake on California's San Andreas fault within the Los Angeles Basin. Szpak & Szczepanek (2023) proposed a solution for the crisis water supply by utilizing water already accumulated in water pipes. Their concept of a drain well presents a method to supply water to the population during emergencies by tapping into the water stored within the pipes. This innovative approach integrates water supply drainage with the utilization of water reserves in crises. Bozek et al. (2012) examined the classification of water resource alternatives based on potential risk levels, which is a crucial step in incorporating these resources into emergency action plans. Their methodology suggests that groundwater can serve as a viable emergency supply with appropriate protections. Bakos et al. (2011) addressed vulnerabilities in hydrological structures and technological equipment essential for emergency water supply (EWS). They used fault tree analysis and brainstorming techniques to assess risks to groundwater sources, defining a vulnerability index for each element to prioritize water sources based on their potential damage levels. Steele & Clarke (2008) proposed a framework for selecting water treatment systems during the acute phase of emergencies. Their results demonstrated that this framework is highly reliable in identifying the most suitable option.

Emergency decision-making is complex because it requires integrating multiple criteria, models, and data sources within a limited time frame (Qu et al. 2016; Pagano et al. 2021). Several competing criteria can influence the selection of the most appropriate EWS alternative (EWSA). Decision-makers often have to make quick decisions without complete knowledge of the situation and available alternatives (Zhou et al. 2018; Ghandi & Roozbahani 2020). Therefore, using and implementing multi-criteria decision-making (MCDM) methods and analysis would be an appropriate framework to compare and choose the best solution for water supply in emergencies. Additionally, the MCDM approach incorporates the perspectives of decision-makers in the selection process (Hashemi et al. 2013, 2021). MCDM is a well-established decision-making process that addresses decision problems that include a number of decision criteria for each individual decision (Ebrahimnejad et al. 2012; Hosseinzadeh et al. 2014; Mousavi et al. 2014; Zamani-Sabzi et al. 2016; Banihabib & Shabestari 2017; Birgani & Yazdandoost 2018; Ghoddousi et al. 2018).

In the past two decades, various MCDM methodologies have been employed in water resource management. Hamidifar et al. (2023) utilized the analytic hierarchy process (AHP), Fuzzy-AHP, and TOPSIS (technique for order of preference by similarity to ideal solution) to assess project implementation methods. Their findings showed that diverting river water and building temporary storage dams were the highest priorities, while pipeline branching to nearby areas was ranked the lowest. Boukhari et al. (2024) evaluated the water supply service in Taoura, Algeria, using household survey data and ranking performance indicators through the AHP. The main objective was to assess water supply service (WSS) management performance in Algeria by employing a methodology based on household satisfaction surveys and classification of performance indicators. Ncube et al. (2024) investigated criteria for prioritizing areas for conversion to continuous water supply using questionnaires and semi-structured interviews. The research focused on the Lusaka Water Supply and Sanitation Company as a case study. The study identified financial sustainability as the primary criterion for selecting areas to prioritize.

Opricovic (2011) introduced a fuzzy VIKOR (VIekriterijumsko KOmpromisno Rangiranje) method for ranking reservoir systems. Hashemi et al. (2013) developed a fuzzy MCDM framework using VIKOR and Atanassov's intuitionistic fuzzy sets to evaluate large-scale water plans. Linhoss & Jeff Ballweber (2015) combined ambiguous data and unknown weights for water sustainability in the southern US. Kim & Chung (2013) used fuzzy VIKOR to assess water supply susceptibility to climate change. Ashbolt & Perera (2017) proposed a hybrid multi-objective optimization and multi-criteria analysis for water grid operations in southeast Queensland. Hashemi et al. (2014) extended VIKOR for reservoir flood control. Liu & Han (2018) applied the simple additive weighting (SAW) method to assess satisfaction levels in district-metered areas. Salehi et al. (2018) used fuzzy TOPSIS for water network rehabilitation in Iran. Jamali et al. (2018) suggested a spatial multi-criteria model for subsurface dam construction. Peters et al. (2019) used various multi-criteria decision analysis (MCDA) methods, including AHP and elimination and choice translating reality (ELECTRE), to compare potential drinking water sources.

Previous studies have shown that MCDM methods have potential applications in modern emergency planning and management. Recently, Zamani et al. (2022) integrated the decision-making trial and evaluation laboratory (DEMATEL) with the analytic network process (ANP) techniques to weigh criteria, along with the VIKOR technique to prioritize water resources. They recognized seawater and treated wastewater effluent from the treatment plant (following advanced treatment) as feasible water sources for flood disasters in Bandar Abbas. Their research introduced an efficient model aimed at identifying the optimal water resources suitable for diverse natural disasters occurring in various geographical regions. Ju et al. (2020) developed a decision support system (DSS) specifically designed to aid emergency responses in water treatment plants affected by various disasters, with modules for planning, disaster sensing, and response. Wang et al. (2021) collaborated on optimizing EWS strategies during water scarcity by integrating transboundary water resource supply with prospect theory (PT) and using a multi-stage, multi-objective approach. By using a two-stage NSGA-II and TOPSIS-based model, the study identifies optimal compromise solutions and conducts sensitivity analyses on water demand. Assessments of proportional fairness degrees aid in pinpointing key decision factors. Through a case investigation within the Lancang–Mekong River basin, involving China and the nations located downstream, the study demonstrates the methodology's practicality by comparing previous decisions with optimal outcomes across various scenarios.

Loo et al. (2012) assessed various water treatment technologies, particularly membrane-based ones, suitable for emergency conditions. They also devised a methodology employing compensatory multi-criteria analysis to choose the most suitable water treatment technology. In line with this, Hosseinzadeh & Sarpoolaky (2024) further contributed to this field by offering membrane-based technology aimed at delivering high-quality drinking water. Yu & Lai (2011) proposed a distance-based fuzzy multi-criteria group decision-making (FMCGDM) methodology as a DSS for emergencies. Their real-life case study demonstrated that this approach could enhance decision-making objectivity in providing a reliable source of fresh water during emergency action plans. Levy & Taji (2007) used group ANP to create a DSS for emergencies. This DSS, integrating quadratic programming and interval data, was utilized by emergency management and civil defense personnel in Brandon during exercises, aiding in decisions between evacuation and ‘shelter-in-place’ options. Tamura et al. (2000) introduced a mathematical framework to assess alternatives in decision-making concerning low-probability, high-consequence occurrences such as earthquakes. Their analysis contrasted value function models grounded in risk with those based on expected utility theory, highlighting the efficacy of risk-oriented value functions in evaluating public risks. They highlighted the significant impact such risks could have on a large population or the environment, despite their low probability of occurrence. Hosseini et al. (2016) developed an MCDM model for choosing optimal temporary housing locations. Meanwhile, Qu et al. (2016) devised a multi-criteria group decision-making (MCGDM) model, utilizing fuzzy TOPSIS to identify the most suitable water technologies across various emergency supply situations. Their primary aim is to ensure the provision of safe drinking water within crisis management frameworks.

Despite the critical importance of emergency planning, there remains a notable scarcity of studies that specifically apply MCDM models to water supply planning and management during disasters. This gap is particularly concerning given that water supply is a fundamental need in emergencies, yet it has received limited attention in MCDM-focused research. Furthermore, existing studies have largely neglected the crucial factor of uncertainty in the decision-making processes related to selecting the most optimal water supply options during emergencies. This oversight is significant because uncertainty is inherent in emergency situations, where variables and conditions can change rapidly and unpredictably. As a result, there is a critical gap in the literature regarding the development of reliable and adaptable strategies for EWS management, leaving emergency planners without robust tools to address this vital issue.

To directly address these critical gaps, this study introduces an innovative FMCGDM framework that leverages the VIKOR technique within a fuzzy environment. The primary objective of this research is to systematically determine the relative preference order for selecting the most suitable water supply options during emergency scenarios. By doing so, the study aims to develop a robust and reliable emergency action plan that ensures a dependable supply of fresh water in post-disaster situations. This framework not only considers the complexity and uncertainty inherent in emergency decision-making but also provides a methodologically sound approach to optimize the selection of water supply choices, thereby enhancing the effectiveness of emergency response efforts.

This study contributes significantly by (1) introducing a new group decision-making model specifically designed for securing reliable fresh water supplies during emergencies, (2) ranking water supply alternatives in an uncertain environment using fuzzy numbers, and (3) proposing a novel aggregated index based on relative preference relations, as well as ideal and anti-ideal separation concepts, to rank potential alternatives. The main aim of the proposed FMCGDM approach is to enhance decision-making accuracy, reliability, and transparency, thereby improving the effectiveness of emergency response decisions. Additionally, the paper includes a case study that demonstrates the application of the proposed FMCGDM model in selecting appropriate fresh water supplies during emergencies and provides sensitivity analyses on varying criteria weights.

Definition 3 (Lee 2005; Wang 2014, 2015). A fuzzy preference relation is a fuzzy subset of with a membership function indicating the preference degree of fuzzy numbers over is considered:

• 1. P is considered reciprocal if and only if holds true for all fuzzy numbers and ⁠.

• 2. P is transitive if and only if holds true for all fuzzy numbers ⁠, ⁠, and ⁠.

• 3. P is a fuzzy total ordering if and only if it satisfies both the reciprocal and transitive properties.

is preferred to if and only if ⁠, and is equal to if and only if ⁠.

The step-wise process of applying the introduced novel fuzzy compromise solution technique by a team of experts is outlined as follows:

Step 1. First step involves assembling a team of experts tasked with identifying potential alternatives that address all existing conflicting criteria.

A team comprising L experts (l= 1, 2, …, L) is designated to conduct a multi-criteria analysis on m potential alternatives, labelled as A_i (i= 1, 2, …, m), considering each individual of the n criteria, labelled as C_j (j= 1, 2, …, n).

Step 2. The significance of each expert is evaluated using a linguistic term and then converted into a TFN, denoted as ⁠, expressed as ⁠.

Step 3. The weight assigned to each chosen criterion, denoted by j and evaluated by lth expert, is subjectively expressed using a linguistic term and then converted into the TFN, denoted as ⁠.

Here, is associated with benefit criteria, where larger values of A_i signify stronger preference, and is associated with cost criteria, where lower values of A_i indicate stronger preference.

for cost criteria (⁠ and ⁠).

Step 13. The alternatives are ranked by sorting each value in increasing order.

Implementing the proposed model in real-world emergency scenarios may present practical challenges, such as data collection difficulties and limited expert availability. Resource constraints and restricted availability of domain experts can make obtaining accurate and timely data in emergency situations complex. However, the flexibility of the model allows it to adapt to varying levels of data quality and expert input.

Iran, located within the Alpine-Himalayan seismic belt, is prone to numerous natural disasters due to its geographical positioning. This region is considered one of the most active tectonic areas in the world, experiencing 31 out of 40 types of documented natural disasters. While these events are unpredictable, taking proactive measures can significantly mitigate their negative impacts. One crucial priority in dealing with such emergencies is ensuring immediate access to safe drinking water for affected populations.

This study addresses the pressing need for effective EWS management in Iran by proposing five EWSAs according to the World Health Organization (WHO 2003) guidelines, as detailed in Table 1. These alternatives include packaged water, water tankers, mobile water treatment units, emergency water tanks, and existing distribution systems. Each alternative offers distinct advantages and considerations, such as mobility, capacity, and cost-effectiveness, which ensure a comprehensive approach to providing emergency water. Additionally, four criteria, aligned with the standards set by the Environmental Protection Agency of the United States (USEPA 2011), were taken into account to select the optimal EWS strategies during disasters, as outlined in Table 2. These criteria encompass cost, delay in service, water quality, and the affected population, ensuring that the chosen strategies meet rigorous regulatory and operational requirements while addressing the diverse needs of affected communities.

Table 1

EWS alternatives

EWSAs .	Description .
Packaged water (⁠⁠)	Numerous organizations have employed collapsible models, characterized by their compact size when folded, enabling transportation at significantly lower expenses. Despite their cost-effectiveness, these models are not particularly durable. Alternatively, some agencies utilize stackable water containers equipped with snap-on lids featuring small openings (WHO 2003).
Water tankers (⁠⁠)	In the immediate term, water transportation can be facilitated by specialized water tank trucks, water tank trailers, or standard trucks equipped with tanks. However, this method entails considerable expenses (WHO 2003).
Mobile water treatment units (⁠⁠)	Mobile water purification systems, often installed on trailers or within shipping containers, typically integrate processes such as coagulation, filtration, and disinfection, or solely filtration and disinfection, and can deliver water at rates ranging from 4,000 to 50,000 litres per hour (WHO 2003).
Emergency water tanks (⁠⁠)	Water needs to be transported, potentially using pumps, to a storage reservoir of an appropriate capacity, contingent upon the population it will serve, the dependability of the water source, and the treatment infrastructure (WHO 2003).
Existing distribution systems (⁠⁠)	Vulnerable areas within distribution networks, like river crossings, exposed canals, areas affected by landslides, etc., or locations where pipelines intersect seismic faults, should undergo reinforcement measures (WHO 2003).

EWSAs .	Description .
Packaged water (⁠⁠)	Numerous organizations have employed collapsible models, characterized by their compact size when folded, enabling transportation at significantly lower expenses. Despite their cost-effectiveness, these models are not particularly durable. Alternatively, some agencies utilize stackable water containers equipped with snap-on lids featuring small openings (WHO 2003).
Water tankers (⁠⁠)	In the immediate term, water transportation can be facilitated by specialized water tank trucks, water tank trailers, or standard trucks equipped with tanks. However, this method entails considerable expenses (WHO 2003).
Mobile water treatment units (⁠⁠)	Mobile water purification systems, often installed on trailers or within shipping containers, typically integrate processes such as coagulation, filtration, and disinfection, or solely filtration and disinfection, and can deliver water at rates ranging from 4,000 to 50,000 litres per hour (WHO 2003).
Emergency water tanks (⁠⁠)	Water needs to be transported, potentially using pumps, to a storage reservoir of an appropriate capacity, contingent upon the population it will serve, the dependability of the water source, and the treatment infrastructure (WHO 2003).
Existing distribution systems (⁠⁠)	Vulnerable areas within distribution networks, like river crossings, exposed canals, areas affected by landslides, etc., or locations where pipelines intersect seismic faults, should undergo reinforcement measures (WHO 2003).

View Large

Table 2

Criteria for selecting EWS

Criteria .	Description .
Cost (⁠⁠)	Expenses will vary based on several factors, including scale, duration, site characteristics, equipment accessibility, security concerns, and the extent of necessary infrastructure (USEPA 2011).
Delay in service (⁠⁠)	The duration for which residents could reasonably be expected to rely on their own water source (USEPA 2011).
Water quality (⁠⁠)	During brief durations (under 30, 60, or 90 days), focusing on meeting acute exposure standards might be more suitable than monitoring for substances linked to chronic, prolonged health hazards (USEPA 2011).
Affected population (⁠⁠)	In specific scenarios, the urgent need for water in urban settings may encompass not just the permanent residents but also the transient population of workers and visitors during the day. It is crucial to factor in the requirements and priorities of essential facilities such as hospitals and potential shelter sites when outlining emergency response plans (USEPA 2011).

Criteria .	Description .
Cost (⁠⁠)	Expenses will vary based on several factors, including scale, duration, site characteristics, equipment accessibility, security concerns, and the extent of necessary infrastructure (USEPA 2011).
Delay in service (⁠⁠)	The duration for which residents could reasonably be expected to rely on their own water source (USEPA 2011).
Water quality (⁠⁠)	During brief durations (under 30, 60, or 90 days), focusing on meeting acute exposure standards might be more suitable than monitoring for substances linked to chronic, prolonged health hazards (USEPA 2011).
Affected population (⁠⁠)	In specific scenarios, the urgent need for water in urban settings may encompass not just the permanent residents but also the transient population of workers and visitors during the day. It is crucial to factor in the requirements and priorities of essential facilities such as hospitals and potential shelter sites when outlining emergency response plans (USEPA 2011).

View Large

A panel of four skilled experts (⁠⁠) was established to evaluate and select the best water supply alternative during an emergency situation (Step 1). The importance of each expert was determined by the executive manager using Table 3 and incorporated into the corresponding triangular fuzzy numbers illustrated in Table 4 (Step 2). Five potential water supply alternatives (⁠⁠) and four criteria (⁠⁠) were selected for further assessments. Experts provided the importance weights for the four primary criteria using linguistic terms, as detailed in Table 3. These weights are shown in Table 5, corresponding to Step 3. The relative importance of the chosen criteria, as determined by all experts, was then aggregated, and the results are illustrated in Table 5, as part of Step 4.

In Step 5, four experts use linguistic terms outlined in Table 6 to represent their ratings of alternatives concerning criteria, as detailed in Table 7. The ratings, obtained by all four experts for each criterion, are then aggregated, with the results shown in Table 8 (Step 6).

In Step 7, the weighted aggregated decision matrix is assembled, and the results illustrated in Table 9. Following that, Step 8 identifies the PIS and NIS, and the respective outcomes are displayed in Table 9 as well. The ideal separation matrix and the anti-ideal separation matrix are computed using Equations (16)–(23), and the results are shown in Table 10, which represents the conclusion of Step 9.

Table 10

and matrices

EWSAs .	Relative preference degrees .	Criteria .
		C1 .	C2 .	C3 .	C4 .
WS1		0.534	0.500	0.500	0.699
		0.791	1.000	1.000	0.801
WS2		0.704	0.642	0.689	0.540
		0.622	0.858	0.811	0.960
WS3		0.500	0.930	0.551	0.500
		0.826	0.570	0.949	1.000
WS4		0.826	0.787	0.603	0.731
		0.500	0.713	0.897	0.769
WS5		0.519	1.000	1.000	1.000
		0.806	0.500	0.500	0.500

EWSAs .	Relative preference degrees .	Criteria .
		C1 .	C2 .	C3 .	C4 .
WS1		0.534	0.500	0.500	0.699
		0.791	1.000	1.000	0.801
WS2		0.704	0.642	0.689	0.540
		0.622	0.858	0.811	0.960
WS3		0.500	0.930	0.551	0.500
		0.826	0.570	0.949	1.000
WS4		0.826	0.787	0.603	0.731
		0.500	0.713	0.897	0.769
WS5		0.519	1.000	1.000	1.000
		0.806	0.500	0.500	0.500

View Large

In Step 10, the values of ⁠, ⁠, ⁠, and are computed using Equations (24)–(27), while in Step 11, the values of indices and are derived using Equations (28) and (29). The values of and are set at 0.5. Finally, the ranking indices of all potential emergency water supplies have been calculated and are illustrated in Table 11 (Step 12).

Table 11

⁠, ⁠, ⁠, and ⁠, indices values, RI, and final ranking by the proposed model

EWSAs .	.	.	.	.	Indices values .		RI .	Final ranking .
					.	.
WS1	0.469	0.753	0.806	1.000	0.684	1.012	1.673	2
WS2	0.522	0.644	0.765	0.819	0.661	0.895	1.778	4
WS3	0.504	0.500	0.784	0.961	0.575	0.978	1.598	1
WS4	0.572	0.500	0.711	0.960	0.618	0.931	1.692	3
WS5	0.767	0.968	0.501	0.500	0.982	0.568	2.744	5

EWSAs .	.	.	.	.	Indices values .		RI .	Final ranking .
					.	.
WS1	0.469	0.753	0.806	1.000	0.684	1.012	1.673	2
WS2	0.522	0.644	0.765	0.819	0.661	0.895	1.778	4
WS3	0.504	0.500	0.784	0.961	0.575	0.978	1.598	1
WS4	0.572	0.500	0.711	0.960	0.618	0.931	1.692	3
WS5	0.767	0.968	0.501	0.500	0.982	0.568	2.744	5

View Large

According to Table 11, the ranking order of the five alternatives for the EWS (Step 13) is as follows: ⁠. Hence, the best alternative according to the selected criteria is ⁠.

The final ranking, with Mobile Water Treatment Units (⁠⁠) as the top priority, followed by Packaged Water (⁠⁠) and Emergency Water Tanks (⁠⁠), underscores the importance of flexible and rapidly deployable solutions in emergencies. By using this ranking to prioritize the most efficient and accessible water supply alternatives, decision-makers can direct resources. The lower ranking of Water Tankers (⁠⁠) and Existing Distribution Systems (⁠⁠) suggests that these methods may be less effective under certain emergency conditions, highlighting the need for adaptable and resilient water supply strategies in crisis management.

The comprehensive analysis in this study systematically evaluates and ranks potential EWSAs, demonstrating the effectiveness of the proposed model. In addition, while the case study focuses on Iran, the proposed MCGDM model is adaptable to other regions and disaster contexts by adjusting the potential alternatives and criteria to fit local conditions. This flexibility allows for a broader application, enhancing its relevance and utility in diverse EWS scenarios.

To validate the robustness of findings, a sensitivity analysis explores the impact of varying weights and criteria on the final rankings, offering deeper insights into the model's reliability and applicability in different emergency scenarios. The sensitivity assessment specifically aimed to assess the influence of different degrees of fuzziness in the weights on the ultimate ranking of alternatives. Therefore, symmetrical values for and between 0 and 1 were assigned.

Table 12 presents the numerical results of the sensitivity analysis, which are derived from the obtained RI. The obtained ranking indices show that assigning and considering different values for and under different emergency conditions of water supply would lead to change in the sort of ranks among the potential alternatives.

Table 12

Various values of and along with the preference order ranking according to the proposed model

and values .	.	EWSAs .
		WS1 .	WS2 .	WS3 .	WS4 .	WS5 .
and	RI	1.755	1.695	1.521	1.626	2.574
	Preference order ranking	4	3	1	2	5
and	RI	1.722	1.726	1.552	1.652	2.637
	Preference order ranking	4	3	1	2	5
and	RI	1.689	1.760	1.582	1.679	2.707
	Preference order ranking	3	4	1	2	5
and	RI	1.673	1.778	1.598	1.692	2.744
	Preference order ranking	2	4	1	3	5
and	RI	1.656	1.796	1.613	1.706	2.784
	Preference order ranking	2	4	1	3	5
and	RI	1.623	1.835	1.644	1.734	2.869
	Preference order ranking	1	4	2	3	5
and	RI	1.591	1.877	1.675	1.761	2.965
	Preference order ranking	1	4	2	3	5

and values .	.	EWSAs .
		WS1 .	WS2 .	WS3 .	WS4 .	WS5 .
and	RI	1.755	1.695	1.521	1.626	2.574
	Preference order ranking	4	3	1	2	5
and	RI	1.722	1.726	1.552	1.652	2.637
	Preference order ranking	4	3	1	2	5
and	RI	1.689	1.760	1.582	1.679	2.707
	Preference order ranking	3	4	1	2	5
and	RI	1.673	1.778	1.598	1.692	2.744
	Preference order ranking	2	4	1	3	5
and	RI	1.656	1.796	1.613	1.706	2.784
	Preference order ranking	2	4	1	3	5
and	RI	1.623	1.835	1.644	1.734	2.869
	Preference order ranking	1	4	2	3	5
and	RI	1.591	1.877	1.675	1.761	2.965
	Preference order ranking	1	4	2	3	5

View Large

Furthermore, sensitivity analysis was performed on the considered criteria weights. During this process, two levels of increases and decreases were considered. By increasing the fuzziness levels, M was converted to H, and by decreasing the fuzziness levels, M was converted into L, while other criteria remained the same.

In conclusion, the planning and management of water supply in the aftermath of disasters represent critical challenges requiring effective solutions. This study addresses these challenges by introducing a novel FMCGDM approach, incorporating fuzzy VIKOR and relative preference relations to select optimal water supply alternatives during emergencies. The model leverages linguistic variables represented by fuzzy numbers to determine the relative significance of each possible alternative and expert and provide their performance ratings versus criteria. Furthermore, distances between alternatives and ideal/anti-ideal solutions are obtained through the relative preference relation. Finally, an extended collective index has been presented, leveraging the relative preference relation to simultaneously consider both the shortest distance of each alternative from the positive ideal point and the farthest distance from the negative ideal point.

A real-life case study validates the proposed model by assessing five potential EWS alternatives using four evaluation criteria. The mobile water treatment units, one of the five potential EWS alternatives, with an RI value of 1.598, ranks first in this assessment, making the existing distribution systems, with an RI value of 2.744, unsuitable and the last priority. The results of the case study demonstrate the proposed FMCGDM model's efficacy in assessing and ranking alternatives in crisis scenarios. In addition, a thorough sensitivity analysis was conducted on key parameters and criteria weights in the assessment procedure. The findings indicate that the final rankings have remained largely consistent despite variations in these factors.

The study is subject to some obstacles and limitations. A barrier to implementing the proposed model is the difficulty of establishing a team of experts, gathering data on the importance of criteria, and assessing the performance of prospective alternatives in emergency situations. Moreover, the experts must have a comprehensive comprehension of the evaluation criteria in order to reduce potential errors while providing input data for the proposed model. Additionally, the mathematical intricacy of the suggested approach may occasionally restrict its applicability. This constraint can be resolved by implementing the procedure using uncomplicated and easily accessible software, such as Excel, which is suitable for decision-makers. For future research endeavours, the proposed model holds promise for extension and application in decision-making processes following disaster emergencies. Further exploration could entail refining and enhancing the model's capabilities to accommodate evolving scenarios and address emerging challenges in EWS management. Additionally, investigating the model's applicability across diverse geographical contexts and disaster types could yield valuable insights into its robustness and adaptability, contributing to more resilient emergency response strategies.

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

The authors declare there is no conflict.