Water treatment systems are essential for public health in Tanzania, yet their reliability varies significantly. A randomized controlled trial was conducted to assess the performance of different water treatment systems under varying conditions. The study employed a Bayesian hierarchical model for estimating system reliability and uncertainty. In one randomly selected village, a new filtration system showed an improvement in treated water quality with a 15% reduction in turbidity compared to existing systems. The randomized field trial demonstrated the effectiveness of using Bayesian hierarchical models to assess system reliability. These findings can inform policy and resource allocation for sustainable water treatment infrastructure development. Investigate scalability of the new filtration system across different regions and consider long-term maintenance strategies to ensure continued reliability. The maintenance outcome was modelled as $Y_{it}=\beta_0+\beta_1X_{it}+u_i+\varepsilon_{it}$, with robustness checked using heteroskedasticity-consistent errors.