A Quasi-Experimental Evaluation of Power-Distribution System Reliability in Kenya: A Case Study on Equipment Diagnostics and Maintenance Optimisation

{ "background": "Chronic power outages in sub-Saharan Africa are often attributed to ageing infrastructure and inadequate maintenance regimes. This case study addresses a critical gap in the region's engineering literature: the lack of rigorous, data-driven evaluations of maintenance interventions on distribution network reliability.", "purpose and objectives": "This study aimed to quantify the causal impact of a targeted equipment diagnostics and maintenance programme on the reliability of a Kenyan power-distribution network. The primary objective was to determine if the intervention significantly reduced the frequency and duration of sustained interruptions.", "methodology": "A quasi-experimental design was employed, comparing reliability indices from 15 treatment substations against 15 matched control substations over a 24-month period. The intervention involved thermographic surveys, dissolved gas analysis for transformers, and predictive maintenance scheduling. The impact was estimated using a difference-in-differences model: Yit = \β0 + \β1 + \β2 + \δ ( \× ) + \εᵢₜ, with cluster-robust standard errors at the substation level.", "findings": "The maintenance programme significantly improved reliability. The estimated reduction in the System Average Interruption Duration Index (SAIDI) was 4.8 hours per customer per year (95% CI: 2.1 to 7.5). Furthermore, fault location and repair times in the treatment group were reduced by an average of 32% compared to the control group.", "conclusion": "The quasi-experimental analysis provides robust evidence that a diagnostics-led, predictive maintenance strategy can yield substantial improvements in the operational reliability of ageing distribution networks in this context.", "recommendations": "Distribution network operators should institutionalise regular equipment diagnostics and adopt data-driven, predictive maintenance models. Regulators should consider incorporating incentives for reliability improvements based on verified intervention studies into tariff-setting mechanisms.", "key words": "power distribution reliability