Bayesian Hierarchical Model for Measuring Cost-Effectiveness of Transport Maintenance Depots in Ethiopia

This study focuses on evaluating the cost-effectiveness of transport maintenance depots (TMDs) in Ethiopia's logistics system. A Bayesian hierarchical model was employed to analyse data from multiple depots across different regions of Ethiopia. The model accounts for spatial variability in cost-effectiveness and incorporates uncertainty through robust standard errors. The analysis revealed significant variation in the cost-effectiveness of TMDs, with some depots showing substantial cost savings compared to others. Bayesian hierarchical modelling provides a nuanced approach to understanding cost-effectiveness, enabling targeted improvements in depot operations and resource management. Based on findings, recommendations include prioritising the expansion of more cost-effective depots and enhancing maintenance practices for improved efficiency. The maintenance outcome was modelled as $Y_{it}=\beta_0+\beta_1X_{it}+u_i+\varepsilon_{it}$, with robustness checked using heteroskedasticity-consistent errors.