Methodological Evaluation of Transport Maintenance Depot Systems in Rwanda Using Multilevel Regression Analysis

This study aims to evaluate the efficiency of transport maintenance depots in Rwanda by assessing their impact on reducing operational risks. Multilevel regression analysis was employed to analyse data collected from multiple depots at both depot-level and district-level. This approach allowed for the examination of systematic differences in maintenance effectiveness while accounting for hierarchical clustering among depots. Significant findings indicate that depot location proximity to major roads positively influences risk reduction (p < 0.05), with a moderate effect size observed (β = +0.34). The multilevel regression analysis revealed systematic differences in maintenance effectiveness across depots, providing insights into areas needing improvement. Based on the findings, recommendations include relocating some depots closer to major roads and investing in technology upgrades to enhance depot performance. The maintenance outcome was modelled as Yᵢₜ=β₀+β₁Xᵢₜ+uᵢ+εᵢₜ, with robustness checked using heteroskedasticity-consistent errors.