Assessing water service performances in rural sub-Saharan Africa environment: The case studies of two councils of the southern and eastern regions of the Republic of Cameroon (Central Africa) Victor Dang Mvongo and Célestin Defo

This study uses multi-criteria analysis to assess water services performance provided to rural communities. The approach is based on five indicators (water availability, water quality, accessibility, affordability, and reliability) and allows the assessment of service levels. The indicators used provided a solid baseline for water services to identify a strategy for the improvement and achievement of universal access to water. We empirically applied the approach to rural water services in Cameroon and particularly in Mvangan and Mandjou Councils. A total of 77 water services has been investigated through technical inspection (water point inspection, flow measurement, and water quality analysis), semi-structured interviews with the technical services of these councils and surveys with water point committees, including 25 in Mandjou Council and 52 in Mvangan Council. The main results show that 43.40% of villages have enhanced services in Mvangan Council while only 4.76% of services are enhanced in Mandjou Council. Results also suggests that monitoring rural water services can improve the levels of services provided to populations and, hence, universal access to water. These assessments represent only a current snapshot of potable water delivery system conditions and should be conducted at regular intervals to track changes in overall and local conditions.


INTRODUCTION
For more than three decades, significant progress has been made in improving access to safe drinking water around This is an Open Access article distributed under the terms of the Creative Commons Attribution Licence (CC BY 4.0), which permits copying, adaptation and redistribution, provided the original work is properly cited (http://creativecommons.org/licenses/by/4.0/). the world. It is estimated that approximately 50 billion USD was invested in the construction of water infrastructure between 1990 and 2008 to improve access to safe drinking water (Carter & Lockwood ). At the same time, the proportion of the population with access to drinking water from a water system has increased from 44 to 58% (World Bank ). In addition, the Millennium Development Goals With the adoption of the Sustainable Development Goals (SDGs), governments have committed to ensure universal and equitable access to safe drinking water for all by 2030, reducing inequalities in access to safe drinking water, and providing high levels of water services in terms of quality, accessibility, and reliability. This involved a reorganization of the sector adapted to the new institutional situation, an urgent increase in financial resources for water, and the development of sustainable water service management models. The SDGs posed a triple challenge: improving the levels of service provided, ensuring the sustainability of services, and promoting universal access to safe drinking water.
In addition, Agenda 2030 commits UN member states to 'leave no one behind.' This involved analyzing who is excluded from the water supply and finding out why; to determine what can be done and to take steps to ensure that people who are marginalized in the past are included now and in the future. The aim was to reconcile the sustainability objectives of water services with the objectives of achieving universal access to water. Then, the purpose of this article is to use a multi-criteria approach to assess water service levels of drinking water services received by rural communities at the village level. More specifically, the multi-criteria analysis will be used to weigh the relative importance of service level indicators in order to assess water service performance.

Study area
The area chosen to apply the methodological tool consists of Council, on the other hand, is situated in Mvila Division (Southern Region of Cameroon). It is located between 11 50 0 and 12 00 0 north latitude and 2 00 0 and 2 65 0 east longitude ( Figure 1).

Biophysical environment
The Mandjou and Mvangan Councils are subject to a fourseason Equatorial Guinean climate (Olivry ; Suchel ). Rainfall is abundant and varies from 1,500 to 2,000 mm per year with an average annual temperature of 25 C and an average amplitude of 2.4 degrees. Annual relative humidity is around 82% (Mvangan Council ). Their geography is that of the Southern Cameroonian plateau with ferralitic soils. These councils present two phytogeographical complexes: the dense and humid forest in Mvangan and Mandjou, and the savannah in the northern part of Mandjou Council (Mandjou Council ). The river system is quite dense with the presence of rivers such as Kom in Mvangan and Dja in Mandjou.

Socio-economic background
The population of Mvangan Council is estimated at 31,475 inhabitants. The Fangs and the Boulou are the two main

Data collection
Secondary information on water service management was obtained from the Mvangan and Mandjou Communal Development Plans, databases of the Participatory Development Assistance Package, and activity reports from technical services. Specifically, these were data on the management of financial flows of services, the frequency of outages and maintenance operations, and data on water quality monitoring. The primary data were obtained through a technical inspection of water supply systems, semi-structured interviews, and a survey.

Technical inspection
The technical inspection of water supply systems consisted of identifying and classifying water points such as well, borehole, and spring, measuring flows and taking water samples Water quality data were collected at source by sampling and laboratory analysis. The collection equipment consisted of plastic vials because of the facilities they present for transport. The method of sampling was made according to the origin of the water. In the case of groundwater, there were two very different cases. When it was a borehole or a well equipped with a hand pump, the samples were taken after an uninterrupted pumping test lasting a total of 5 minutes.
In the case of a collection at a fountain terminal, the maximum flow faucet was opened for 5-10 seconds and then reduced to an average flow for 2 minutes. The bottle was then put under the tap without closing it in order to take the sample. In this study, two parameters were used to measure water quality: the presence of Escherichia coli in water and turbidity (less than 2 NTU). Turbidity was measured in situ using a PCE-TUM 50-brand turbidimeter.
A total of 77 samples was taken and sent to the laboratory for microbiological analyses. The method used to detect E. coli is membrane filtration. This method can detect and quantify E. coli bacteria in water. In principle, a volume of water (100 mL) is filtered onto a filter membrane that is The second section is for technical data and service reliability while the third part collects data on service accessibility. The last part is devoted to collecting information on the affordability of service. A total of 77 questionnaires was administered to the water service manager. The questionnaire was administered through semi-structured interviews with water point managers. Verbal consent was sought prior to the interview.

Weighting indicators
After selecting water service indicators, pairwise comparisons were made between them to calculate the weight of each of these indicators in determining service levels. For this purpose, four researchers and four professionals specializing in the management of rural water services in sub-Saharan Africa expressed their preference for one indicator over the other. The indicators were therefore compared two to two to determine their priorities. The relative importance values of one indicator over the other used in this study are those proposed by Saaty () and presented in Table 1. These comparisons lead to the establishment of a matrix called the judgment matrix (Equation (2)).  (3).
with CI: Consistency Index and RI: Random Index.
If CR 0:1 the matrix is considered sufficiently consistent. Otherwise, evaluations require revision to reduce inconsistencies (Saaty ). 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.
Consistency Index is calculated using Equation (4).
with n: the number of rows or columns of the square matrix of judgment; and λ max : the principal eigenvalue. Saaty () suggested that the greatest value can be obtained using Equation (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.

Aggregation and interpretation
The aggregation method used is the arithmetic method. This method produces perfect substitutability and compensability between indicators (Nardo et al. ). The Water Service Score (WSC) was evaluated using Equation (6).

Water service indicators
Drinking water services refers to the accessibility, availability, and quality of the main source used by households for drinking, cooking, personal hygiene, and other domestic uses. In implementing water service policy, COHRE et al.
() suggest four criteria (availability, water quality, physical accessibility, and affordability) to assess service levels, to which, reliability has been added (

Water availability
The water available for each person must be sufficient and constant for personal and domestic uses including consumption, individual sanitation, food preparation, and personal and domestic hygiene. The amount of water is the simplest and most used indicator for tracking and comparing services. It is measured by the number of liters per capita per day (l/h/d).

Water quality
The water required for every personal and domestic use must be safe, and therefore free of microbes, chemicals, and radiological risks that pose a threat to health. Water must have a color, smell, and taste acceptable to users. In general, the quality does not vary with the level of service.

Accessibility
Water, as well as adequate facilities and services, must be accessible safely for all segments of the population. Every human being has the right to a water service that is phys- This indicator brings together other parameters such as distance to water point and waiting time. Accessibility can be measured in minutes per capita per day (mn/h/d).

Affordability
Water services must be at a price that each person can afford without reducing their opportunities to acquire other essential goods and services such as food, housing, healthcare, and education. In the UK, the Department of Environment,

Food and Rural Affairs has determined that consumers
should spend no more than 3% of their income on water and sanitation, and that a rate between 1.5 and 3% of income should be considered (COHRE et al. ).

Reliability
Reliability is expressed as a percentage of the time the service is (or is not) fully functional. It indicates the extent to which the service is in line with the expected level. The concept of reliability or safety is based on the assumption that all services will eventually fail and, therefore, total safety can only be achieved if multiple water sources and/or systems are accessed (Moriarty et al. ). Reliability does not mean that a service is provided 24/7, but that it is predictable.  The calculation of the weights of the above indicators resulted in the establishment of the analysis grid presented in Table 5. This grid allows evaluation of service levels on the basis of scoring. These scores vary between 0 and 2 depending on the criteria met by the service and described in Table 5.

Analysis grid
In Table 5, water quality is considered good when it does not contain E. coli and turbidity is lower than 2 NTU. If one of these two parameters is not verified, the water is classified as questionable. In the event that these two parameters are not checked, the water is considered bad. In terms of reliability, this indicator will be described as very reliable when users are sure to obtain good quality water at any time. When users have water only during the opening hours of the water point, this indicator will be classified as low. On the other hand, when there are periods of service disruption, outages and repairs that drag on, this indicator will be described as uncertain. This grid is similar to the IRC water service delivery ladder framework    service, and 4.76% receive no services. Figure 2 shows the graphic synthesis of service levels per council.
In reading Figure 2, there is a significant disparity in the water services received by communities in these two municipalities. These disparities are due, on the one hand, to the management system put in place in each of these municipalities and, on the other hand, to the local context. Indeed, prior to the implementation of water management policies in these two municipalities, they were facing many problems such as the irregularity of monitoring, the poor functionality of water point committees, the low structuring of the maintenance chain, the low financial flows generated by the sale of water, and the absence of the culture of water sales (Dang , ). These deficiencies have been identified as the result of the incomplete implementation of the ongoing decentralization process in Cameroon and most countries in West and Central Africa (Dang ). In response to this situation, the Mvangan Council has put in place a water management system that has been implemented. The Mandjou Council set up a water point management system that never worked. In addition, the massive influx of Central African refugees has led to a rapid deterioration of water service level provided to the population. Thus, the comparative analysis of water service levels received by the populations of these two councils ( Figure 2) suggests that monitoring water services can improve water service levels provided to the population. Service that the quantity and quality of water is insufficient, the water point is too far away or the system is unreliable, and the service is not affordable High 25% WSC < 50% Limited service The service does not meet the basic service criteria quantity and density at the water point High 50% WSC < 75% Basic service Service that the quantity and quality of water is sufficient, provided collection time is not more than 30 minutes for a roundtrip including queuing, the system is reliable, and the service affordable Medium 75% WSC 100% Enhanced service Service that the quantity and quality of water is sufficient, the water point is located on premises (less than 10 minutes), the system is very reliable, and the service affordable were absent in all water points.
In terms of accessibility, most households take between 10 and 30 minutes to collect water. They travel an average of 800 m to get drinking water. However, this parameter is difficult to measure at the local level (Flores et al. ).
Economically, services are accessible throughout the Mvangan Council. Indeed, the amount allocated by households for the contribution to the maintenance fund is less than or equal to the critical threshold of 1.5% of household income. This is due to the municipality taking part of the costs of the service. In Mandjou Council, the situation is rather mixed and it depends on the level of household wealth. This distinguishes between villages that can be described as 'rich' such as Bazzama, for which the costs of service are affordable, and so-called 'poor' households for whom the costs of service are too expensive. In terms of reliability, villages close to borough chiefs are more secure than remote villages. This is due to the difficulties for the repairing craftsman to move to ensure the technical monitoring of the works because of the chronic poor condition of the roads and especially in the rainy season.

DISCUSSION
Multi-criteria analysis is used to assess indicators and determine the most important factors in assessing service levels by assigning weights to them. As a result, it allows for a more accurate assessment of the performance of water services. Indeed, in unweighted approaches to This tends to produce results that do not reflect reality.
On the other hand, the use of the conducted analysis has better results since it takes into account the relative importance of the indicators. However, the use of multi-criteria analysis and, in particular, the analytical hierarchy process The results also suggest that the approach proposed for assessing service levels has similarities to the work of  presumably for pragmatism, the costs associated with data collection, or the use of indicators that are considered agents to assess service levels. The indicators selected for the assessment approach were essential for measuring water service level for decision-making. The use of multi-criteria analysis is used to assess indicators and determine the most important factors in assessing service levels by assigning weights to them. As a result, it allows for a more accurate assessment of the performance of water services comparatively to unweighted approaches.
The approach provides a holistic way to assess drinking water service levels in a local context. In analyzing water service levels, the scale of services and the empirical results of the assessment, when shared with community members, proved to be a communication tool for improving rural water services. In addition, the approach can be used to monitor the state of the right to water and as a reference point for decisions. The approach can also be used in similar contexts, providing a baseline for holistically assessing service levels, designing improvements, and monitoring behavior over time. The approach can help community managers prioritize their actions and investments, and selfmobilize for improvements for external support.
Politically and managerially, the results show the importance of post-construction monitoring of drinking water systems and community structure managers and the need for governments to provide grants to local governments to ensure that this monitoring is effective. The assessments provided represent only a current snapshot of potable water delivery system conditions. Therefore, a time interval for conducting such assessments to track changes in overall and local conditions should be considered. The approach developed could be improved by establishing thresholds or benchmarks for indicators. In addition, community members could assign weights to attributes, which can appropriate results and mobilize for actions that generate positive changes in the evaluated system.

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