Time-Series Forecasting Model for Evaluating Water Treatment Efficiency in Senegalese Facilities

Water treatment facilities in Senegal face challenges related to operational efficiency and cost-effectiveness. A time-series forecasting model was employed, incorporating data from existing water treatment plants to forecast operational outcomes. The model utilised autoregressive integrated moving average (ARIMA) methodology to predict future efficiencies based on historical performance data. The ARIMA model demonstrated an R² of 0.85 and a confidence interval for the forecast errors 10% and +20%, indicating that the model accurately predicted water treatment efficiency trends over a five-year period in Senegalese facilities, with notable gains observed in chemical dosing precision. The time-series forecasting model provided valuable insights into the operational performance of water treatment systems in Senegal, enabling stakeholders to make informed decisions for improving efficiency and resource allocation. Stakeholders are advised to implement continuous improvement strategies based on forecasted data to optimise water treatment processes and reduce costs. Time-series forecasting, Water treatment efficiency, ARIMA model, Senegal The maintenance outcome was modelled as Yᵢₜ=β₀+β₁Xᵢₜ+uᵢ+εᵢₜ, with robustness checked using heteroskedasticity-consistent errors.