The study used various tools to collect data: a GPS device to geolocate water infrastructure, a volumetric meter to quantify the volume of water produced by the borehole in m³/h, and a Photo camera to document the condition of the infrastructure. Specialised software such as QGIS and Origin was used to process the data.

(1) Estimation of population

The population projection is based on the following formula:

$P_n=P_0\ast \left(1+\alpha \right)\left(N_i-N_0\right)$ [6] (1)

with:

P_n: Population of year n;

P₀: Population in the reference year, taken in 2021, corresponding to the latest population and habitat census in Ivory Coast;

N_i: year i;

N₀: reference year;

α∶ rural growth rate 2.50% [3]

The estimation will be made for the following time horizons:

(2) Choice of reference demand and increase in water dotation

1) Choice of reference dotation

[7] established a baseline for basic human water dotation, emphasising the importance of ensuring that everyone has enough water for survival, health and hygiene. According to his research, basic water dotation can be divided into several categories:

2) Method for calculating the increase in the dotation

Water dotation depends on the level of development of the village:

$D_n=D_0{\left(1+a\right)}^n$ [8] (2)

For this study, we will take a growth rate of 2% in urban areas [1]

(3) Method for calculating average daily consumption (Q_m.j)

$Q_{m.j}=\left(Q_1+Q_2\right)\left({\text{m}}^{\text{3}}/\text{j}\right)$ (OMS, 2015) (3)

Calculation of water requirements for domestic use (Q₁) and social and economic activities (Q₂) Requirements for domestic use (1):

$Q_1=P\ast D\left({\text{m}}^{\text{3}}/\text{j}\right)$ [7] (4)

Social water needs and those of economic activities (2).

For social water needs and those of economic activities, a rate of 10% of domestic needs will be used [8] .

$Q_2=0.1\ast Q_1\left({\text{m}}^{\text{3}}/\text{j}\right)$ (5)

(4) Maximum daily flow (Q_max.j)

$Q_{max.j}=Q_{m.j}\ast K_1\left({\text{m}}^{\text{3}}/\text{j}\right)$ (6)

Avec K₁ = hourly coefficient of variation, equivalent to 1.2 [9]

(5) Point of flow (Q_ph.max.)

$Q_{ph}=Q_{max.j}{K}_0\left({\text{m}}^{\text{3}}/\text{h}\right)$ (7)

(With K₀ hourly variation coefficient = 1.2) [10]

(6) Maximum Daily Flow Rate Produced (Q_max.P)

$Q_{max.P}=Q_{max.j}+\left[20\%Q_{max.j}\right]$ (8)

Here, a margin of 20% is added to account for eventual losses in the network or unforeseen additional needs [10] .

(7) Modelling of water consumption based on population

To analyse the relationship between two variables, such as consumption as a function of population, Origin software was used as a statistical analysis tool.

After obtaining the data (population and consumption), it was entered into Origin in two columns. The software was then used to plot a scatter diagram representing the relationship between the two variables, and then to apply a linear regression using the option “Fit Linear”.

The software automatically generates the regression line equation, the coefficients (slope and intercept), and associated statistical values such as standard errors, t-tests, p-values, correlation coefficient (r) and coefficient of determination (R²). These results allow you to carry out an evaluation of the validity of the model and interpret the effect of the independent variable on the dependent variable. Origin also facilitates the visualisation of results and allows you to make predictions based on the model obtained.

The population growth of Tioroniaradougou ( Figure 2 ) reveals a significant increase in population, from 3786 inhabitants in 2014 to an estimated 11,659 inhabitants in 2050, or 208%. This trend is consistent with the dynamics observed in other West African cities. For example, [11] reports population growth of 2.8% in Bobo-Dioulasso, while [12] observes an increase of 3.1% in Bamako. Such demographic pressure inevitably leads to an increase in water needs, making it essential to adapt infrastructure to supply water in order to avoid water stress.

The increase in water supply per capita will rise from 44 l/day/capita in 2014 to 89 l/day/capita in 2050, reflecting the expected improvement in living conditions and climate variations. According to [13] , this increase is often the result of sustained investment in drinking water infrastructure. [14] also emphasise that urban growth leads to a proportional increase in water demand. This trend implies the need for a gradual strengthening of water services to accompany socio-economic transformations ( Figure 3 ).

The increase in average daily consumption, reaching 1146 m³/day in 2050 compared to 182 m³/day in 2014, is a clear indicator of the growing pressure on water resources and water infrastructure. [15] point out that this consumption is growing rapidly in areas where living standards are improving, while [16] show that consumption can increase faster than the population itself. It is therefore essential to anticipate future needs in order to ensure rational and sustainable management of available resources ( Figure 4 ).

Maximum daily consumption will reach 1376 m³/day in 2050 ( Figure 5 ). This consumption highlights challenges related to peak demand. According to [17] , these peaks can intensify water stress, particularly during periods of heat or drought. [18] add that climatic factors and urban economic fluctuations directly influence maximum daily demand. It is therefore imperative to incorporate these variables into water management planning in order to avoid service disruptions.

The linear relationship obtained between population and water consumption in Tioroniaradougou, through the equation Y = −374.54734 + 0.12513X, has a coefficient of determination R² = 0.97 and a negative ordinate at the origin. This is justified by the fact that this model is intended solely to make forecasts within the observed range of demographic data. This result confirms the existence of a strong link between population growth and increased consumption [19] . Indeed, 97% of the observed variations in water consumption can be explained by population growth. Statistical analysis of the model (F value = 148.17; p-value = 0.00026) confirms its significance. The goodness-of-fit and Pearson’s correlation coefficient (r = 0.98) show that population growth is a relevant predictor of consumption ( Figure 6 ).