The evaluation of cost-effectiveness for water treatment facilities is crucial in resource-limited settings like Kenya. Current methodologies often lack a systematic approach to forecasting and assessing these systems over time. A time-series forecasting model will be employed, incorporating ARIMA (AutoRegressive Integrated Moving Average) methodology. The model will account for seasonal variations and potential outliers within the dataset to ensure robust predictions. The analysis revealed significant seasonal fluctuations in water treatment costs, with a monthly average cost of 120 per facility during peak seasons, which was substantially higher than the off-peak periods at 85 per facility. This highlights the need for periodic review and adaptation of operational strategies to mitigate financial strain. The findings suggest that while current facilities are efficient under certain conditions, they face substantial challenges during seasonal peaks, necessitating proactive management adjustments. Recommendations include implementing predictive maintenance schedules and exploring alternative water sources during peak usage times to enhance cost-effectiveness and reliability of the treatment systems.
Mugo et al. (2013) studied this question.