Abstract
{ "background": "Municipal infrastructure asset management systems are critical for sustainable urban development, yet their adoption in developing nations is poorly quantified. Existing evaluations often lack a formal probabilistic framework to integrate sparse, multi-level data and account for regional heterogeneity.", "purpose and objectives": "This case study develops and applies a Bayesian hierarchical model to evaluate the adoption rates of structured municipal infrastructure asset management systems. It aims to quantify national and regional adoption trajectories and identify key influencing factors.", "methodology": "A case study methodology was employed, utilising a longitudinal dataset from municipal records. The core statistical model is a Bayesian hierarchical logistic regression: $\\text{logit}(p{ij}) = \\alphaj + \\beta X{ij}$, where $\\alphaj \\sim \\text{Normal}(\\mu\\alpha, \\sigma\\alpha^2)$ represents region-specific intercepts, $p{ij}$ is the adoption probability for municipality $i$ in region $j$, and $X{ij}$ are covariates. Inference was performed using Markov Chain Monte Carlo sampling.", "findings": "The model estimates a national adoption probability of 0.32 (95% credible interval: 0.28 to 0.37) by the study's end-point. A key finding is the significant regional variation, with the posterior distribution of $\\sigma_\\alpha$ indicating substantial heterogeneity (\(median = 0\).82, 95% CrI: 0.61 to 1.10). Technical capacity and institutional support were the most influential covariates.", "conclusion": "The Bayesian hierarchical model provides a robust, probabilistic framework for assessing infrastructure management adoption in data-scarce contexts. It confirms that adoption is moderate and highly variable across regions, driven primarily by institutional and technical factors.", "recommendations": "Policy should prioritise building technical capacity and standardising institutional frameworks to reduce regional disparities. Future asset management programmes should incorporate probabilistic monitoring and evaluation tools akin to the model presented.", "key words": "asset