Replication and Methodological Evaluation of Power-Distribution System Cost-Effectiveness in Tanzania: A Quasi-Experimental Design

{ "background": "Previous research on power-distribution system cost-effectiveness in sub-Saharan Africa has been largely theoretical or reliant on simulation models, with limited empirical validation in real-world, low-resource grid environments. This creates a significant evidence gap for infrastructure investment decisions.", "purpose and objectives": "This study aimed to rigorously replicate and methodologically evaluate a quasi-experimental design for measuring the cost-effectiveness of different medium-voltage equipment systems in a Tanzanian context. The core objective was to test the robustness of the original methodological framework under field conditions.", "methodology": "A quasi-experimental design was implemented, comparing operational costs and failure rates of conventional pole-mounted transformer systems against compact substation systems across matched distribution feeders. The primary analysis employed a generalised linear model: Cit = \β0 + \β1 Ti + \β2 Xit + \εᵢₜ, where C is cost, T is treatment, and X are covariates. Inference was based on robust standard errors clustered at the feeder level.", "findings": "The replication confirmed the direction of the original findings, indicating superior cost-effectiveness for compact substations, but the effect magnitude was substantially smaller. The estimated cost reduction was 18% (95% CI: 12% to 24%), compared to the originally reported 31%. Methodological evaluation revealed that maintenance labour cost variability was a key factor not fully accounted for in the prior model.", "conclusion": "The quasi-experimental design is a viable method for field-based cost-effectiveness analysis, but its application requires careful contextual calibration of cost parameters. The core finding of compact substation advantage holds, albeit with a moderated effect size.", "recommendations": "Future applied research should integrate more granular, locally sourced operational cost data into the modelling framework. Utilities should conduct pilot deployments informed by these calibrated results before large-scale technology adoption.", "key words": "Power distribution, cost-effectiveness, quasi-experimental design, replication study, sub