Abstract
{ "background": "Reliable power distribution is critical for economic development, yet many low-income nations face persistent challenges with system performance. There is a paucity of longitudinal, methodologically robust evaluations of distribution network reliability in such contexts, limiting evidence-based infrastructure investment.", "purpose and objectives": "This short report aims to methodologically evaluate the reliability of power-distribution equipment systems using a quasi-experimental design. The primary objective is to quantify the causal impact of targeted equipment upgrades on system reliability metrics.", "methodology": "A quasi-experimental difference-in-differences design was employed, comparing reliability indices in treatment and control regions before and after a large-scale intervention. The core statistical model is $Y{it} = \\beta0 + \\beta1 (Treati \\times Postt) + \\gammai + \\deltat + \\epsilon{it}$, where $Y{it}$ is the System Average Interruption Duration Index (SAIDI) for region $i$ in period $t$. Robust standard errors were clustered at the regional level.", "findings": "The intervention significantly reduced SAIDI. The coefficient $\\beta1$ was estimated at -12.7 hours per customer per year (95% CI: -18.3, -7.1), indicating a substantial improvement in reliability. This represents a 34% reduction in outage duration compared to control regions.", "conclusion": "Targeted upgrades to distribution infrastructure can yield statistically significant and practically meaningful improvements in system reliability. The quasi-experimental approach provides a robust framework for evaluating engineering interventions in real-world settings.", "recommendations": "Utilities should adopt similar quasi-experimental methodologies for programme evaluation. Investment planning should prioritise regions with baseline reliability indices above the national median, where marginal returns are likely highest.", "key words": "power distribution, reliability engineering, quasi-experimental design, difference-in-differences, infrastructure evaluation", "contribution statement": "This paper provides a novel application of a quasi-exper