The reliability of water treatment facilities in Rwanda has been a subject of concern due to their critical role in ensuring safe drinking water and sanitation. A time-series forecasting model was developed and applied to historical data from to assess the reliability of water treatment facilities. The model incorporates robust standard errors for uncertainty quantification. The analysis revealed significant variations in water quality metrics over time, with a notable fluctuation in turbidity levels by up to 25% during peak usage periods. The proposed time-series forecasting model effectively captured the dynamics of water treatment system performance and can be utilised for future maintenance planning. Implementing regular quality control checks and upgrading infrastructure based on predicted trends is recommended to enhance long-term reliability. The maintenance outcome was modelled as Y₈ₓ=₀+₁X₈ₓ+uᵢ+₈ₓ, with robustness checked using heteroskedasticity-consistent errors.
Mushombo et al. (Tue,) studied this question.