"background": "The reliability assessment of industrial machinery fleets in developing economies is often hampered by sparse, heterogeneous, and censored failure data, leading to imprecise maintenance planning and resource allocation. ", "purpose and objectives": "This study develops and validates a novel Bayesian hierarchical modelling framework to estimate the reliability of heterogeneous machinery fleets operating in Uganda, providing robust failure rate estimates and quantifying uncertainty for improved asset management. ", "methodology": "A Bayesian hierarchical Weibull model was constructed, integrating fleet-level and individual machine data. The core reliability model for the i^{th machine is Ri (t) = \ (- (\ t) ^\), where \ (\) = \ + \ Xi + ugi, with ug \ N (0, \²) representing random effects for machine group g. Posterior distributions were estimated using Hamiltonian Monte Carlo. ", "findings": "The model successfully pooled information across fleets, yielding more precise reliability estimates than non-hierarchical methods. For a critical class of haulage equipment, the posterior median time to failure was 1, 240 operating hours, with a 95% credible interval of 1, 050, 1, 460. Fleet heterogeneity was significant, with group variance \ estimated at 0. 78 (CrI: 0. 52, 1. 12). ", "conclusion": "The proposed Bayesian hierarchical model offers a statistically rigorous and practically useful tool for reliability analysis under data-scarce conditions, effectively characterising uncertainty and variability across different machinery groups. ", "recommendations": "Adoption of this modelling framework is recommended for asset-intensive industries and regulatory bodies to inform data-driven maintenance schedules, spares provisioning, and lifecycle cost analyses. ", "key words": "Reliability engineering, Bayesian statistics, hierarchical modelling, asset management, maintenance, developing economies", "contribution statement": "This paper presents the first application
Ssekandi et al. (2022) studied this question.