South African municipal water systems are critical for public health but often suffer from reliability issues due to aging infrastructure and inadequate maintenance. A Bayesian hierarchical model will be employed to estimate the reliability of municipal water systems. This model accounts for variations between municipalities by incorporating random effects into the model structure, allowing for more accurate predictions and inference. The analysis reveals a significant proportion (35%) of water supply interruptions in rural areas, highlighting the need for targeted interventions to improve system reliability. This study provides robust estimates of municipal water system reliability using Bayesian hierarchical modelling, which can inform policy decisions aimed at improving service delivery and public health outcomes. Policymakers should prioritise investments in maintenance and upgrading of infrastructure in areas with higher interruptions to enhance the overall reliability of South African municipal water systems. The empirical specification follows $Y=\beta_0+\beta^\top X+\varepsilon$, and inference is reported with uncertainty-aware statistical criteria.