"background": "Chronic unreliability in electrical power distribution remains a critical infrastructural constraint on socio-economic development. Existing assessments often lack rigorous causal frameworks for evaluating the performance of ageing equipment systems against defined reliability metrics. ", "purpose and objectives": "This study aimed to develop and apply a novel quasi-experimental methodological framework for the longitudinal evaluation of power-distribution system reliability. The primary objective was to isolate and quantify the causal effect of equipment interventions on key reliability indices. ", "methodology": "A difference-in-differences (DiD) design was employed, analysing panel data from intervention and control groups of distribution feeders. System reliability was measured using the System Average Interruption Duration Index (SAIDI). The core statistical model is SAIDI{it = \0 + \1 (Treati \ Postt) + \ Xit + \ + \ +, where robust standard errors were clustered at the feeder level to account for serial correlation. ", "findings": "The analysis indicates a statistically significant reduction in SAIDI for the treatment group following targeted equipment upgrades. The DiD estimator, \1, was -18. 7 hours/customer/year (95% CI: -25. 3, -12. 1), representing a reliability improvement of approximately 32% relative to the control group's mean. Transformer failure rates were identified as the predominant contributor to unreliability in the baseline period. ", "conclusion": "The proposed quasi-experimental framework provides a robust analytical tool for empirically evaluating the efficacy of capital investments in power-distribution infrastructure. It demonstrates a clear causal link between specific equipment interventions and enhanced system reliability. ", "recommendations": "Utility regulators and network operators should adopt quasi-experimental designs for the ex-post evaluation of major capital projects. Investment planning should prioritise interventions targeting the identified high-failure-rate components to maximise reliability gains. ", "key words": "power distribution, reliability evaluation, quasi-experimental
Okonkwo et al. (Thu,) studied this question.