Introduction

The problem of the sustainability of rural water services is not new and has received international attention since the International Decade of Potable Water and Sanitation (World Bank 2017). However, new facilities, once damaged or broken down, are abandoned after a relatively short period of time. Some studies (RWSN 2010; Improve International 2012) have shown that 35–50% of drinking water systems become inoperable 5 years after their construction. More recently, Foster et al. (2019) have shown that one in four water points is non-functional at any time in Sub-Saharan Africa and Asia. The cumulative costs of water point outages in Sub-Saharan Africa over the past two decades are estimated to range from $1.2 billion to $1.5 billion (Baumann 2009; IRC 2009).

In Cameroon and in Mvila Division in particular, very little data are available on the failures of Rural Drinking Water Supply System (RDWSS). However, Deal and Furey (2019) estimate that 32% of handpump are inoperable at all times. In addition, isolated studies in some Cameroonian municipalities show that 10–60% RDWSS are broken or not functioning optimally (Mvongo and Defo 2021; Mvongo et al. 2019; Minlo and Defo 2018; Tedah and Defo 2018; Defo et al. 2016). It is estimated that over 736 RDWSS have been constructed in Mvila Division including 431 wells, 164 boreholes, 104 springs, and 37 water supply network. These RDWSS are managed by rural communities regrouped in Water Point Committee (ProADP Mvila 2018).

To deal with this situation, it is essential to analyze the sustainability of water services to help managers to identify priority actions and investments to be implemented to improve the sustainability of services. Indicators and indices provide frameworks for important public policy analysis of water services management and are important public communication tools. Several indices related to the management of water services have been developed in the recent decade (2010–2020). These include the Index of Drinking Water Adequacy (Lee Kuan Yew School of Public Policy 2010), the Sustainable Water Governance Index (Iribarnegaray and Seghezzo 2012), equity index in water and sanitation (Luh et al. 2013), Sustainability Index Tool (Schweitzer et al. 2014), Water Sanitation and Hygiene (WASH) performance Index (Cronk et al. 2015), Sustainable Water Management Index. (Maiolo and Pantusa 2019), and index of water service quality (Molinos-Senante et al. 2019).

Index of Drinking Water Adequacy (IDWA) allows cross-country comparisons and help in ascertaining which of the five components (resource, access, use, capacity, and quality) of access to drinking water is weak and requires priority attention. In contrast, the Sustainable Water Governance Index (SWGI) assesses the link between the concept of governance and sustainability and their role in water and sanitation management (OECD 2015). SWGI relies on four pillars, including access, planning, human resources, and participation, to assess governance.

The equity index in water and sanitation assesses progress in equity in achieving human rights to clean water and sanitation (Luh et al. 2013). The Sustainability Index Tool (SIT) is designed to assess the sustainability of drinking water and sanitation programs and projects (Schweitzer et al. 2014) while the WASH Performance Index assesses the performance of water, sanitation and hygiene services. The Water Management Index measures the sustainability of water management and assesses the effects of policies taken to achieve sustainability goals. The index of water service quality is based on Analytical Hierarchy Process (AHP) and Monte–Carlo simulations to evaluate water service quality provided by Rural Drinking Water Supply System (Molinos-Senante et al. 2019).

Although there are successful experiments in the implementation of these indices, they are not fully applicable in rural areas without modification. Indeed, the rural environment faces problems and challenges in the water supply that are different from those in urban areas including the high rate of non-functionality of water supply systems (RWSN 2010; Foster et al. 2019, 2018), the non-profitability of the sector (Kleemeier 2010), the limited capacity of the structures responsible for the supply of drinking water (Githae et al. 2018). In addition, these indices do not provide a comprehensive assessment of the sustainability of rural water services; which could lead to poor decision-making on the part of managers. Therefore, it is necessary to develop a new sustainability index based on the contextual parameters and specificities of the rural environment. This index will be able to obtain information on rural water services, and prioritize problems related to service management with priority actions to improve the management of water services. The aim of this study is to develop a sustainability index for rural water services in Mvila Division (Southern Region of Cameroon) that is applicable in Sub-Saharan Africa in similar affinity pattern.

Materials and methods

Study area

The study area is the Mvila Division, located in the Southern Region of Cameroon. It covers an area of 8695 km2 and is located between latitudes 2° 36′ 06.94″ North and 3° 04′ 14.58″ North and longitudes 10° 34′ 26.66″ East and 12° 15′ 36.93″ East (Fondjak 2006). Administratively, the Mvila Division is divided into eight Councils including Ebolowa 1st, Ebolowa 2nd, Biwong-Bane, Biwong-Bulu, Ngoulemakong, Mvangan, Mengong, and Efoulan (Fig. 1). It is limited to the South by the Ntem Valley Division and the Republic of Gabon, to the East by the Dja and Lobo Division, to the northwest by the Ocean Division, and to the North by Nyong and So'o Division.

Fig. 1
figure 1

Location of Mvila Division

Full size image

Its position in the South Cameroonian plateau in the heart of the Gulf of Guinea gives it a four-season Guinean-style Equatorial climate that characterizes this part of Sub-Saharan Africa. The average annual rainfall is in the order of 1876 mm, while the average annual temperature is around 24 °C (Fondjak 2006). The soils are ferralitic and rest on a basic Precambrian complex originally covered by sediment deposits bordering the older nucleus of the Congolese craton (Olivry 1986). Plant formations belong to dense, moist forest (Letouzey 1985). The department is watered by three watersheds: the Nyong Basin, the Ntem Basin, and the Lokoundjé Basin. The characteristics of the above biophysical environment are found in most Sub-Saharan African countries located in dense forest areas. However, they differ from those of countries in areas of Sudanese–Sahelian climate because of their vegetation, as soils have an influence on groundwater recharge.

Mvila Division has about 179,429 habitants (BUCREP 2005). The indigenous people are made up of the Boulou, Ewondo, Bene, Fong, and Fang ethnic groups. The population is mainly rural and lives in villages and hamlets of varying sizes located along the roads. The social organization is that of the people of the forest, of the acephale or segmental type. The economy is based on agriculture with a strong practice of cocoa farming. Other activities are also carried out by the village community. These include small-cut farming, fishing, picking, handicrafts or small trade in basic manufactured goods.

Data collection

Secondary data were obtained from scientific papers and technical reports on rural water service failures and factors influencing the sustainability of water services. The primary data were obtained from semi-structured interviews, a survey and a meeting with the main actors in water service management in councils in Mvila Division in the Southern region of Cameroon.

Survey

A survey was carried out among water service managers in some randomly selected villages in Mvila Division. The purpose of this survey was to gather the views of water service managers on the factors that influence the sustainability of water services. The minimum sample size was calculated using Eq. 1 suggested by Martinez-Bencardino (2012):

$$ n = \frac{{NZ^{2} p\left( {1 - p} \right)}}{{NZ^{2} + Z^{2} p\left( {1 - p} \right)}}, $$
(1)

where n is the sample size, N is the total number of service providers (N = 400 in this case study), Z is the standardized normal deviation for a particular confidence unit (Z = 1.96 for a 95% confidence level), p is the proportion of the population estimated to maximize the sample size (p = 0.025), and d is the allowable error (d = 0.05).

Based on Eq. 1, a minimum of 35 water service managers should be investigated. A total of 80 water service managers were surveyed with a response rate of 100% according to the following distribution: 10 in Efoulan council, 10 in Biwong-Bane Council, 10 in Mvangan Council, 10 in Biwong-Bulu Council, 10 in Mengong Council, 10 in Ngoulemakong Council, 10 in Ebolowa 1st Council, and 10 in Ebolowa 2nd Council. Each water service managers surveyed completed a questionnaire containing a panel of 63 indicators identified in the literature and deemed relevant to the sustainability assessment of water services. The questionnaire consisted of Multiple-Choice Questions (MCQ) and was divided into seven sections (general information, economic, environmental, social, technical, institutional, and governance). The first section presented general information on the service and the identity of the respondent. The second section concerned economic indicators (self-financing economy, financial autonomy, total coast recovery, etc.) while the third section presented environmental indicators (water quality, water quantity, climate risk, etc.). The fourth section concerned social indicators (affordability, accessibility, non-discrimination, equity, etc.) and the fifth section presented technical indicators (quality of construction, frequency of maintenance operations, access to spare parts, reliability, etc.). The sixth section concerned institutional indicators (post-construction support to council, post-construction support to service managers, regulation, formalization of contract, organization of the service, etc.) while the seventh section presented governance indicators (skills of water service managers, financial flux management, participation, etc.). It was, therefore, a question for water service managers to select from the indicators in the questionnaire those that are relevant to the assessment of water services in their area of intervention.

Semi-structured interview

The semi-structured interviews (SSI) were conducted with the aim of collecting information on the criteria for assessing the sustainability of water services authorities responsible for regulating the service at the local level. More specifically, these interviews discussed refining and supplementing the indicators selected by water service managers. A total of 11 semi-structured interviews were conducted with different managers. Respondents were selected based on their experience in managing water services and the functions they perform. They were chosen to cover different profiles including:

  • 03 officials of ministry in charge of water and energy (01 head at the central level, 01 in charge at the regional level, and one official at the district level);

  • 08 municipal heads in charge of water services management (01 in Ebolowa 1st, 01 in Ebolowa 2nd, 01 in Efoulan, 01 in Biwong-Bane, 01 in Mengong, 01 in Ngoulemakong, 01 in Biwong-Bulu, and 01 in Mvangan).

Meeting

A meeting was held in the city of Ebolowa (regional capital of Southern Cameroon) with the heads of Councils in charge of the management of water services in Mvila Division. Its purpose was to present the criteria for assessing the sustainability of water services selected as a result of semi-structured interviews and surveys. The meeting also made pairwise comparisons of the criteria used to calculate the weight of each of them in determining the sustainability of water services. A total of 99 people attended the meeting, including 11 water regulatory authorities who participated in the ISS, 80 water service managers who participated in the survey, and 08 researchers (02 professors and 06 PhD students).

Data processing

The development of the Water Service Sustainability Index (WSSI) was based on indicators to measure sustainability. An indicator is a qualitative or quantitative measure of specific facts or conditions (Juwana et al. 2012). If indicators are observed regularly, they can be used to analyze changes during the observation period (Nardo et al. 2005). Some indicators were grouped together to form a particular dimension or indicator and other indicators were broken down into sub-indicators. A group of indicators and/or dimensions put together in a grid provides an Index or composite indicator. An Index measures multidimensional criteria that cannot be explained by indicator (Nardo et al. 2005). The development stages of the WSSI that have been used are inspired by those proposed by several authors (Nardo et al. 2005; Juwana et al. 2012; OECD 2008, 2015) including the selection of dimensions and indicators, the calculation of the weight of dimensions and indicators, aggregation, and the interpretation of the index.

Selection of dimensions and indicators

Dimensions and indicators are the main constituents of the WSSI. They were selected through a literature review on existing evaluation frameworks (Chaves and Alipaz 2007; Juwana et al. 2010; Boukhari et al. 2018; Dominguez et al. 2019). This review of the literature provided a panel of dimensions and indicators that were then refined through discussions with the various stakeholders in the management of rural water services. The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) approach by Moher et al. (2009) was used to select publications to be used to identify dimensions and indicators. A total of 203 publications were identified, 04 were eliminated as duplicates, and 09 were eliminated on the basis of abstract. Of the remaining 190 studies, 167 were selected for the review of the literature after full readings. Of these, 34.1% related to water point’s failures, 23.9% were for the sustainability assessment of services, 40.2% related to water service management models, and 1.8% to the approach leaving no one behind in rural drinking water supply. The distribution of publications in Sub-Saharan Africa shows only 17 out of 48 countries have been studied, as shown in Fig. 2. The selection criteria for the dimensions and indicators used are those proposed by Liverman et al. (1988) including sensitivity to change over time, sensitivity to change in space or within groups, prediction and anticipation, availability of reference values or threshold values, impartiality, appropriate data transformation and integration.

Fig. 2
figure 2

Space distribution of publications in Sub-Saharan Africa

Full size image

Weighting dimensions and indicators

In this study, the participatory approach was used in light of the fact that it directly involves all stakeholders involved in water service management. The analytical hierarchical process (AHP) was used because of its simplicity. It facilitates the decomposition of the problem into a hierarchical structure and ensures that the qualitative and quantitative aspects of the problem are taken into account simultaneously in the evaluation process in which expert judgments are taken into account through an even comparison matrix (Sharma and Rawal 2021; Siekelova et al. 2021; Leal 2020; Dominguez et al. 2019; Boukhari et al. 2018; Saaty 1980) It, therefore, provides a non-linear framework of analysis that takes into account several factors simultaneously and allows dependence and feedback, making numerical compromises to arrive at a synthesis or conclusion (Saaty 1990). The weights of the various indicators were established by following the steps and procedures recommended by Saaty (1990), including the development of the hierarchy structure, pairwise comparisons, the establishment of the pairwise comparison matrix, the calculation of priority vectors, and the determination of the index and consistency ratio (Fig. 3).

Fig. 3
figure 3

Process for calculating the weight of indicators using the AHP

Full size image
Structure of the hierarchy and pairwise comparisons

The structure of the hierarchy has allowed us to reflect the problem to be solved. It clarified the problem and identified the contribution of each dimension and indicators in determining sustainability. The objective of the problem was at the highest level, while the dimensions and indicators were at the intermediate and lowest levels, respectively. After establishing the hierarchy structure, pairwise comparisons were made between dimension and indicators to calculate the weight in determining service sustainability. For this purpose this, 99 people involving in the management of water services were consulted in a meeting. These people expressed their judgments using the Saaty scale (Table 1). Judgments were first exercised among dimensions, then among indicators. These comparisons lead to the establishment of a matrix called the judgment matrix (Eq. 2).

$$ A = \left[ {a_{ij} } \right] = \left[ {\begin{array}{*{20}c} 1 & {a_{12} } & \ldots & {a_{1n} } \\ {\frac{1}{{a_{12} }}} & 1 & \ldots & {a_{2n} } \\ \vdots & \vdots & \ddots & \vdots \\ {\frac{1}{{a_{1n} }}} & {\frac{1}{{a_{2n} }}} & \ldots & 1 \\ \end{array} } \right] $$
(2)
Table 1 Saaty numerical scale for pairwise comparisons
Full size table

With

  • A the decision matrix;

  • \(a_{ij}\) are the values of the importance of comparisons between indicators i and j for all i, j ϵ \(\left\{ {1, 2, \ldots , n} \right\}\).

The weight of the indicators of each matrix was determined by solving the own vector problem (Fig. 4).

Fig. 4
figure 4

Steps in calculating the weight of dimensions and indicators

Full size image
Consistency of pairwise comparisons

Once the weight values of each indicator were obtained, the consistency of the results was verified. For this, the consistency ratio (CR) was calculated. It can detect defects in calculations. The CR was calculated using Eq. (3).

$$ {\text{RC}} = \frac{{{\text{CI}}}}{{{\text{RI}}}} $$
(3)

With:

  • CI: Consistency Index

  • RI: Random Index

If \({\text{CR}} \le 0.1\) the matrix is considered sufficiently consistent. Otherwise, evaluations require revision to reduce inconsistencies (Sharma and Rawal 2021; Siekelova et al. 2021; Leal 2020; Dominguez et al. 2019; Boukhari et al. 2018; Saaty 1980). Once consistency is verified, it can be used to assess the sustainability of rural water services. Random Index (RI) values are shown in Table 2.

Table 2 Some Random Index values
Full size table

Consistency Index is calculated using Eq. (4).

$$ {\text{CI}} = \frac{{\lambda_{{{\text{max}}}} - n}}{n - 1} $$
(4)

With n the number of rows or columns of the square matrix of judgment and \(\lambda_{{{\text{max}}}} \): the principal eigenvalue. Saaty (1990) suggested that the greatest eigenvalue value can be obtained using Eq. (5).

$$ \lambda_{{{\text{max}}}} = a_{ij} \frac{{W_{j} }}{{W_{i} }} $$
(5)

With

  • \(a_{ij}\): Judgment matrix of the value of the i line element and the j column element,

  • \(W_{i}\): contribution to the selection of the best choice and each of the criteria,

  • \(W_{j}\): the contribution of specific criteria to the main objective.

The final step was to determine the relative weight of the indicators relative to the weight of the dimensions using Eq. (6).

$$ {\text{SCr}}_{i} = C_{i} \times {\text{SC}}_{i} $$
(6)

With

  • SCri: Relative weight of indicator i relative to size i;

  • Ci: Weight of dimension i;

  • SCi: Weight of the indicator i obtained from the par pair comparison matrix.

Aggregation and interpretation of final index

The aggregation resulted in the WSSI. The aggregation method used is the arithmetic method. This method produces perfect substitutability and compensability between dimensions, indicators and sub-indicators (Juwana et al. 2012; Nardo et al. 2005). The WSSI was evaluated using Eq. (7).

$$ {\text{WSSI}} = \mathop \sum \limits_{i = 1}^{N} W_{i} S_{i} $$
(7)

With

  • WSSI: Water Services Sustainability Index;

  • N: Number of indicators to be aggregated;

  • Si: The value of the indicator i or the indicator i sub-index;

  • Wi: Weight of indicator i.

The interpretation of the WSSI was made on the basis of the quartile scale. Performance here reflects the status of the indicator, sub-indicator or aggregate global index at a particular time in the evaluation and will be used as the basis for a relevant action priority to be improved (Juwana et al. 2012).

Results

Sustainability assessment criterion

In general, the factors that influence the sustainability of rural water services are social, economic, environmental, and technical in nature (Behailu et al. 2016). These factors have been defined by several authors (Bendahmane 1993; Carter et al. 1999; Eneas da Silva et al. 2013; Giné and Pérez-Fouguet 2008; Katko 1991; Montgomery et al. 2009; Spaling et al. 2014; UN 2007). In this study, six categories of factors or dimensions were identified as relevant to sustainability analysis by service providers and service authority. These are the four classical dimensions (economic, social, environmental, and technical) to which two other dimensions (institutional and sustainable governance) have been added (Table 3). These last two dimensions are cross-cutting to the others. However, in the context of this study, they deserve to be analyzed very carefully. Indeed, in developing countries where economies are insecure and organizational systems are poorly controlled, these factors are decisive for the sustainability of the service. In addition, several authors (Hellström et al. 2000; Ashley et al. 2003; Van Leeuwen et al. 2012; Marques et al. 2015; Boukhari et al. 2018; Dominguez et al. 2019) have shown that the four conventional dimensions are not sufficient to assess the sustainability of water services. Each of the selected dimensions is described below.

Table 3 Dimensions and indicators selected
Full size table

Economic dimension

The availability of financial resources is necessary to ensure continuity of water services (Carter et al. 2010; Gorton et al. 2009; Haysom 2006; Whittington et al. 2008). In this study, the economic dimension aims to determine whether the revenue collected can cover all the costs of the service. This component is analyzed through three indicators: self-financing capacity, financial autonomy, and total coast recovery.

Environmental dimension

The environmental dimension assesses whether the service is meeting health and environmental standards from a qualitative and quantitative point of view. It analyzes on the one hand the compliance of the water distributed, and on the other hand, the quantities of water available in the context of climate change. The analysis of this component is based on three indicators: the amount of water available, water quality and climate risk.

Social dimension

The social dimension analyses on the one hand the affordability of the service and on the other hand the accessibility of the service. In addition, the distance to water supply system and social inclusion are considered to determine whether all social strata have access to the service. Thus, the whole affordability, accessibility, non-discrimination, and equity determines whether the service is socially sustainable.

Technical dimension

The technical dimension of water services is a key element of sustainability. It is associated with infrastructure and encompasses system performance, durability, reliability, flexibility, and adaptability (Ashley et al. 2003, 2004). In this study, four indicators are mobilized to analyze this component including the quality of construction, the frequency of maintenance operations, access to spare parts, and the reliability of water systems.

Institutional dimension

Institutional factors influence how water services are regulated and managed at the local level. They also play a key role in determining the sustainability of services. Thus, the institutional dimension assesses whether local authorities perform their functions in the management of water services. Five indicators are mobilized to analyze this component: post-construction support to Council, post-construction support for service managers, regulation, and formalization of contracts, and organization of the service.

Governance

Governance is a key element in the sustainability of water services. It implies that existing service management structures have the capacity to persist over time and continue to perform their functions over the long term. In this study, four indicators are mobilized to analyze this component. These include the skills of the management structure, the organization of management, the management of financial flows, and participation.

Weight of dimensions and indicators

Hierarchy structure

The structure of the resulting hierarchy consists of three levels as shown in Fig. 7. Level 0 corresponds to the objective, Level 1 compares dimensions against the objective, while Level 2 compares indicators against dimensions.

Weighting dimensions and indicators

Dimension weighting

Table 4a presents the peer-to-peer matrix of the dimensions of sustainability of rural drinking water services. This matrix highlights expert judgments on the relative importance of one dimension of sustainability over another on the basis of the Saaty scale. Analysis of the peer comparison matrix shows that the degree of consistency of comparisons is acceptable since the consistency ration (RC) is less than 10%. Figure 5 presents the graphic synthesis of dimension weights in determining the durability of the service.

Table 4 Pairwise comparisons matrix of dimensions and indicators
Full size table
Fig. 5
figure 5

Dimensions weight in WSSI calculation

Full size image

Analysis of Fig. 5 and the par pair comparison matrix show that the economic dimension (34.01%) and the technical dimension (19.96%) are the most important. They represent 53.97% of the total sustainability weight of rural water services. They are followed in order by environmental (14.65%), governance (13.10%), social (12.65%), and institutional (5.64%) dimensions. These results are similar to those obtained by Boukhari et al. (2018) in terms of economic and social dimensions, but differ for other dimensions. These observed differences are due to the social and geographic context in which the studies are conducted and confirm that the results of studies conducted in one context are not directly transposed to another context due to the influence of socio-cultural variables (Charreaux 2004).

Weighting of indicators

Table 4 (b, c, d, e, f, and g) presents the results of the peer-to-peer comparison of water sustainability indicators for each dimension identified. These comparisons are acceptable since the consistency ratios (RC) are less than 10%. Figure 6 presents the graphic synthesis of the weights of indicators in determining the sustainability of the service.

Fig. 6
figure 6

Indicators weight in WSSI calculation

Full size image

Analysis of Fig. 6 and Table 4 (b, c, d, e, f, and g) shows that self-financing capacity (17.84%), financial autonomy (11.35%), and water quality (7.18%) are the most important indicators in determining the sustainability of water services. On the economic front, the weight of the indicators is 17.84% for self-financing capacity, 11.35% for financial autonomy, and 4.81% for the rate of recovery of contributions. The environmental component is dominated by water quality (7.18%) followed by the amount of water (4.57%) climate risk (2.89%).

On the social front, affordability (6.20%) is the most important indicator. The technical aspect, on the other hand, is dominated by access to spare parts (6.54%) maintenance of drinking water systems. The sustainable governance component is dominated by participation (6.42%), financial flow management (4.09%), and the manager's competence (2.59%). As for the institutional dimension, it is mainly dominated by post-construction follow-up (1.63%) and support for municipalities (1.63%) (Fig. 7).

Fig. 7
figure 7

Hierarchy structure

Full size image

Construction of Water Service Sustainability Index

The Water Services Sustainability Index (WSSI) is the weighted sum of sub-index scores as shown in Eq. 7. This equation obtained as a result of the pairwise comparisons shows that the determination of the WSSI is done sequentially. Indicator values are aggregated to obtain sub-index values and sub-index values are aggregated to obtain the value of the index. The WSSI provides a holistic profile of key water service issues that allow comparisons between different services and analyses.

$$ {\text{WSSI}} = \mathop \sum \limits_{i = 1}^{6} W_{i} S_{i} $$
(8)

With WSSI: Water Service Sustainability Index, Si: The value of the indicator i or the indicator i sub-index and Wi: Weight of indicator i.

The arithmetic method used creates perfect substitutability and a perfect compensability of all sub-indexes. Thus, some weak sub-index values are offset by the high values of some sub-indexes. This method has been used for the development of several indices such as the Canadian Water Sustainability Index (Policy Research Initiative 2007), the Water Poverty Index (Lawrence et al. 2003), and the Watershed Sustainability Index (Chaves and Alipaz 2007).

The interpretation of the WSSI is made on the basis of the quartile scale. This method provides for four levels of performance with a maximum value of 100 and a minimum value is 0 as shown in Table 5.

Table 5 WSSI interpretation
Full size table

Table 5 shown that WSSI is based on a quartile scale with four levels of performance (Poor, Poor–Medium, Medium–Good, and Good). The performance reflects the condition of the issue related to an indicator, a sub-indicator, or the overall aggregated index at a particular time of assessment and would be used as the basis for relevant “Priority of Action” to improve the water service sustainability at the village scale. The “Priority of Action” reflects the priority of action to improve the ‘Performance’ of indicators or sub-indicators, which is the opposite of its respective ‘Performance’ (Juwana et al. 2012). This method of interpretation is the same as that used in the development of the West Java Water Sustainability Index (WJWSI) (Juwana et al. 2011).

Discussion

Dimensions and indicators

The tool proposed in this study has similarities compared to others developed with multi-criterion analysis in particular for the assessment of sustainability (Boukhari et al. 2018; Dominguez et al. 2019; Dwivedi and Bhaduria 2014; Marques et al. 2015) or the quality of services (Molinos-Senante et al. 2019). The indicators were categorized into six categories (economic, social, environmental, technical, institutional, and governance). It consists of a large number of indicators to ensure that indicators with a lower impact are not underestimated (Saaty 1990). In addition, although some indicators (water quality, quantity of water, access to spare parts, maintenance operations, etc.) present in this tool, may be found in other tools (Boukhari et al. 2018; Dominguez et al. 2019; Murtaza et al. 2016), however, the weights assigned to them are different. This is due to the fact that weights are the result of subjective evaluations (Scholz and Tietje 2002) and are also influenced by contextual factors.

Despite the similarities between the tool proposed in this study and the tools developed in the literature, there are a number of differences. The first is to take climate change into account through the climate risk indicator. Climate change is having an impact on the availability of water resources, specific per capita consumption, quality of service, the state of water infrastructure, and the costs of investment and operation of the service (pS-Eau et al. 2013). In addition, the risks associated with climate change can comprehensively challenge the sustainability of services. For example, the extra costs generated by both the long-term and episodic impacts of climate change can lead to a change in budgetary balances, and thus changes in pricing (pS-Eau 2016).

The second is the introduction of a long-term approach that is not only retrospective but also forward-looking. For example, climate risk takes into account not only past and present climate variations, but also future climate variations. The same is true of the service's self-financing capacity, which, while analyzing the service's pricing policy, takes into account the price revision due to current and future inflation.

Third, most methods assessing the sustainability of rural drinking water services use a small number of indicators, presumably for pragmatism, the costs associated with data collection, or the use of indicators that are considered agents to assess dimensions or criteria. The dimensions and indicators selected for the evaluation tool were essential to measuring sustainability for decision-making. The tool developed provides a holistic way to assess the sustainability of water services in a local context.

Construction of WSSI

The arithmetic method used for WSSI construction has previously been used by several authors (Policy Research Initiative 2007; Lawrence et al. 2003; Juwana et al. 2011; Dwivedi and Bhaduria 2014; Molinos-Senante et al. 2019) in the construction of the construction of indices such as the Canadian Water Sustainability Index (Policy Research Initiative 2007), the Water Poverty Index (Luh et al. 2013), and Watershed the Sustainability Index (Chaves and Alipaz 2007). The arithmetic method creates perfect substitutability and a perfect compensability of all sub-indexes. Thus, some weak sub-index values are offset by the high values of some sub-indexes (Juwana et al. 2012). It is possible to have the same value of WSSI in dispute case. Consider two indicators with different values. In the first case, the values of the sub-indexes are 40 and 40 (maximum scale of 100) and in the second case, 10 and 70. If the equal weight arithmetic method is applied, both cases of figures will have the same WSSI value, 40.

This method is different from the geometric approach. Indeed, the geometric method does not create a perfect substitutability and a perfect complacentness between the values of the sub-index of indicators (Juwana et al. 2012). Thus, if we consider the two previous cases of figures, the value of WSSI is 40 for the first case and 26.5 for the second case. It, therefore, takes into account the differences between the values of the sub-indexes in the aggregation process, while the arithmetic method does not take these differences into account (Nardo et al. 2005). Therefore, the geometric method is more appropriate when the differences between the values of the sub-indexes are large. Thus, based on the weightings obtained in this study, the differences between the values of the sub-indexes will not be very large. Hence the interest of the arithmetic method.

Limitation of WSSI

The consideration of uncertainties in the development of indices appears in several studies (Luh et al. 2013; Nardo et al., 2005; Molinos-Senante et al. 2019). In the WSSI development process, there are uncertainties that appear that affect the final index. Potential sources of uncertainty appear in the size and indicator selection process, the weighting method and the aggregation method. As it stands, the validity of the results obtained could be challenged in the absence of an analysis of the robustness of the index (OECD 2008). Thus, the tool could be improved by identifying all sources of uncertainty, assessing the impact of uncertainties on the final index, and conducting an index sensitivity analysis. In addition, the novelty of this study lies in the development of a sustainability index for rural water services. However, it does not assess the application of the index in other environmentally similar regions to the study area. Therefore, a future study could be conducted to validate the results obtained for a wider geographical area.

At this stage of the study, it is difficult to find mitigation results for water management in the area. Nevertheless, many authors have analyzed the situation of water service performance in Sub-Saharan Africa. Mitigation measures identified by this authors could be apply in Mvila Division. Some authors (Carter and Rwamwanja 2006; Harvey and Reed 2006; Moriarty et al. 2013; Schouten and Moriarty 2008) have worked on water service managed by rural communities and have shown that services are only sustainable if technical, financial and managerial support to Water Point Committee (WPC) is provided. Lockwood and Smits (2011) show that post-construction support to WPCs is an integral part of community management. However, some authors (World Bank 2017; Lockwood and Le Gouais 2015; Chowns 2015) suggest that community management is only viable for simple technologies such as wells, boreholes and springs.

In response to the challenges posed by community management, some authors (Whaley and Cleaver 2017; Lockwood and Le Gouais 2015; Adank et al. 2014; Lockwood and Smits 2011) proposed professional models. Professionalization in its simplest form means strengthening the state's ability to provide high-quality services, hiring professional staff or outsourcing to private companies that can provide services effectively at agreed service performance levels. A study by Van der Byl and Carter (2018) shows that under the right conditions, management models by private operators can be a viable option to contribute to the sustainability of rural water supply in Sub-Saharan Africa. However, the commitment of private operators is not a panacea for rural water supply and it is, therefore, necessary to have mixed management models to achieve universal access to safe drinking water.

Conclusion

Despite decades of research on the sustainability of rural water services, maintaining water supplies remains a development challenge in rural communities in developing countries. A sustainability index based on multi-criterion analysis has been developed to measure the sustainability of water services. It consists of six (06) dimensions and twenty-one (21) indicators. The process of establishing this tool consisted on the one hand of selecting the criteria and criteria to be used for sustainability, and on the other hand, to develop the index. The main results showed that the economic dimension (34.01%) and the technical dimension (19.96%) are the most important dimensions. They are followed in order by environmental (14.65%), governance (13.10%), social (12.65%), and institutional (5.64%) dimensions. In terms of indicators, self-financing capacity (17.84%), financial autonomy (11.35%) and water quality (7.18%) are the most important indicators in determining the sustainability of drinking water services in Sub-Saharan Africa and particularly in Mvila Division (Southern-Cameroon). This index provides a solid baseline on the sustainability of water services at the village level in rural area and identifies priority actions to be taken to move services towards sustainability. The application of WSSI will shed light on the state of play of the sustainability of water services and measures aimed at improving water management in Mvila Division, such as the carrying out of the work of maintenance of works under management and more generally the pooling of Mvila Division or several councils. However, the tool could be improved by establishing thresholds or benchmarks for indicators. In addition, members of rural communities could assign weights to the criteria (size) and criteria (indicators), which could improve the appropriation of results and the mobilization of activities that generate positive changes in the services assessed. The tool could also be improved by doing a sensitivity analysis of the final index.