{ "background": "Bayesian hierarchical models (BHMs) have been proposed for optimising the operational yield of industrial machinery fleets in resource-constrained settings. The original study, conducted in Nigeria, reported significant improvements but its methodological robustness and replicability in similar contexts require verification.", "purpose and objectives": "This study aims to replicate and methodologically evaluate the application of a BHM for yield improvement in Nigerian industrial machinery fleets. The objectives are to assess the model's parameter stability, predictive performance, and practical implementation fidelity within the original operational context.", "methodology": "We executed a direct replication using the original model specification $y{ij} \\sim \\text{Normal}(\\alphaj + \\beta X{ij}, \\sigma^2)$, $\\alphaj \\sim \\text{Normal}(\\mu_{\\alpha}, \\tau^2)$, on a newly collected dataset from a comparable Nigerian fleet. Methodological evaluation focused on prior sensitivity, Markov chain Monte Carlo convergence, and comparison of posterior estimates. Robust standard errors were computed to assess inference stability.", "findings": "The replication confirmed the model's core finding of a positive association between structured maintenance scheduling and yield, with a posterior mean coefficient of 0.32. However, the hierarchical variance component $\\tau^2$ was not statistically distinguishable from zero (95% credible interval [-0.14, 0.09]), indicating minimal fleet heterogeneity in the new data and challenging the original model's generalisability.", "conclusion": "While the BHM's directional finding on maintenance was corroborated, the assumed hierarchical structure lacked empirical support in this replication, suggesting the original model may be over-parameterised for this specific industrial context.", "recommendations": "Practitioners should consider simpler pooled or fixed-effects models for fleet yield analysis in settings with high operational homogeneity. Future research should pre-test hierarchical assumptions with variance component analysis before full BHM implementation.", "key words": "Bayesian hierarchical model, replication study, machinery fleet