{ "background": "Reliability engineering in manufacturing contexts often relies on classical frequentist methods, which can struggle with complex, multi-level system data and the incorporation of prior operational knowledge. This is particularly relevant in developing industrial economies where system failure data may be sparse or heterogeneous across different plants.", "purpose and objectives": "This study presents a methodological evaluation of a Bayesian hierarchical model for quantifying system reliability within the manufacturing sector. The objective is to demonstrate its superiority in handling plant-level variability and providing probabilistic inferences for maintenance decision-making compared to conventional approaches.", "methodology": "We developed a three-level hierarchical model where the failure rate $\lambda{ij}$ for component $i$ in plant $j$ is modelled as $\lambda{ij} \\sim \\text{Gamma}(\\alphaj, \\betaj)$, with plant-level parameters $\\alphaj, \\betaj$ drawn from a common hyper-distribution. The model was implemented using Hamiltonian Monte Carlo sampling. Its performance was evaluated against a standard pooled model using data on critical pump failures collected from multiple plants.", "findings": "The hierarchical model yielded more precise and plant-specific reliability estimates, with the 95% credible intervals for mean time between failures (MTBF) being, on average, 34% narrower than those from the pooled model. Crucially, it revealed substantial heterogeneity in underlying reliability parameters across different facilities, a key factor masked by aggregate analysis.", "conclusion": "The Bayesian hierarchical framework provides a robust methodological advancement for reliability analysis in manufacturing systems characterised by inherent operational diversity. It formally accounts for plant variation, leading to more accurate and locally relevant reliability assessments.", "recommendations": "Manufacturing engineers and reliability managers should adopt hierarchical modelling techniques for plant-wide asset performance analysis. Further research should integrate covariate information at the plant and component levels to enhance the model's explanatory power.", "key words": "Reliability engineering, Bayesian inference, hierarchical modelling, manufacturing systems, maintenance optimisation", "contribution statement": "This paper provides a novel