{ "background": "Process-control systems are critical for infrastructure efficiency, yet evaluations in resource-constrained settings often lack robust statistical frameworks to quantify gains. Previous studies in the region have provided point estimates but have not fully accounted for site-specific heterogeneity and uncertainty.", "purpose and objectives": "This study replicates and methodologically extends a prior efficiency analysis by applying a Bayesian hierarchical model. The objective is to rigorously quantify efficiency gains while formally partitioning variance between site-level and system-level effects.", "methodology": "We replicated the data collection protocol from the original study across multiple sites. Efficiency was modelled using a Bayesian hierarchical linear model: $y{ij} \\sim \\mathcal{N}(\\mu + \\alphai + \\beta x{ij}, \\sigma^2)$, with $\\alphai \\sim \\mathcal{N}(0, \\tau^2)$, where $y{ij}$ is the efficiency metric for observation $j$ at site $i$, $\\alphai$ represents the random site effect, and $x_{ij}$ denotes the implemented control parameter. Posterior distributions were estimated using Markov Chain Monte Carlo sampling.", "findings": "The model estimated a mean system efficiency gain of 18.7% (95% credible interval: 15.2% to 22.1%). However, the site-level random effects variance ($\\tau^2$) was substantial, indicating that nearly 40% of the observed variation was attributable to differences between specific site conditions, not the control system alone.", "conclusion": "The replication confirms the potential for significant efficiency gains but highlights that a substantial portion of the variance is context-dependent. The hierarchical approach provides a more nuanced and generalisable interpretation than the original aggregate analysis.", "recommendations": "Future engineering assessments of process-control systems should adopt hierarchical modelling to distinguish between universal and localised effects. System deployment must be accompanied by site-specific calibration protocols.", "key words": "Bayesian hierarchical model, process control, replication study, efficiency analysis, infrastructure management", "cont