Hierarchical Bayesian Modelling for Cost-Effectiveness Analysis of Power-Distribution Equipment in Kenya

The adoption of power-distribution equipment (PDE) in Kenya has been hampered by inconsistencies in system performance and cost-effectiveness across different regions. A hierarchical Bayesian model was employed to analyse data from various regions in Kenya, accounting for both systematic differences between regions and within-region variability. The analysis revealed significant cost savings potential through the optimization of PDE placement, with identified clusters showing up to 30% reduction in total system costs compared to current practices. This study provides a novel framework for assessing and optimising PDE systems across Kenya, offering substantial benefits in terms of both efficiency and sustainability. Based on the findings, it is recommended that future policy initiatives prioritise targeted investments in identified cost-savings clusters to maximise overall system performance. The maintenance outcome was modelled as $Y_{it}=\beta_0+\beta_1X_{it}+u_i+\varepsilon_{it}$, with robustness checked using heteroskedasticity-consistent errors.