{ "background": "Municipal infrastructure efficiency diagnostics are critical for asset management in developing economies. A previous study proposed a Bayesian hierarchical model to analyse efficiency gains, but its methodological robustness and practical applicability required independent verification.", "purpose and objectives": "This replication study aims to methodologically evaluate the proposed Bayesian hierarchical model for infrastructure efficiency diagnostics. The objectives are to verify the model's computational reproducibility, assess the sensitivity of its parameter estimates to prior specifications, and test its predictive performance on an updated dataset.", "methodology": "We executed a full computational replication and then a robustness analysis using an expanded longitudinal dataset of municipal water and road asset systems. The core model is $\\eta{it} \\sim \\text{Normal}(\\alphai + \\beta X{it}, \\sigma^2)$, with $\\alphai \\sim \\text{Normal}(\\mu{\\alpha}, \\tau^2)$, where $\\eta{it}$ is the efficiency metric for municipality $i$ at time $t$. Sensitivity was tested using alternative priors for $\\tau^2$.", "findings": "The replication confirmed the original model's core finding of a positive mean efficiency trend ($\\mu\\alpha = 0.18$). However, the robustness analysis revealed that estimates for municipality-level random effects $\\alphai$ were sensitive to the choice of hyperprior for $\\tau^2$, with posterior credible intervals for 15% of municipalities shifting across zero under alternative priors.", "conclusion": "While the hierarchical structure is sound for capturing municipal heterogeneity, the model's diagnostic outputs for specific units are not robust to prior uncertainty. This limits its utility for precise, municipality-level efficiency ranking without careful prior elicitation.", "recommendations": "Practitioners applying this model should conduct thorough prior sensitivity analyses. Future methodological work should focus on developing more robust hyperpriors for variance components in hierarchical infrastructure models.", "key words": "Bayesian replication, infrastructure asset management, hierarchical modelling, robustness analysis, computational