"background": "Reliable power distribution is critical for economic development, yet many African nations face persistent challenges in grid reliability. There is a pressing need for robust, data-driven methodologies to evaluate infrastructure and forecast performance to inform strategic investment. ", "purpose and objectives": "This study aims to methodologically evaluate power-distribution equipment systems and develop a time-series forecasting model to measure and predict system reliability, providing a tool for proactive infrastructure management. ", "methodology": "A longitudinal dataset of system interruptions was analysed. The core forecasting model employs an Autoregressive Integrated Moving Average (ARIMA) framework, specified as \ᵈ yt = c + =1^{p\ \ᵈ yt-i + =1^q\ -j + \, where parameters were estimated via maximum likelihood. Model diagnostics included checks for residual autocorrelation and stationarity. ", "findings": "The ARIMA (1, 1, 1) model provided the best fit, forecasting a 22% reduction in the System Average Interruption Duration Index (SAIDI) over the next five-year period, with a 95% confidence interval of 18%, 26%. The methodological evaluation identified transformer failures and line faults as the predominant causes of unreliability. ", "conclusion": "The developed model offers a statistically sound tool for forecasting reliability trends, demonstrating that targeted interventions in specific equipment categories can yield significant improvements in overall system performance. ", "recommendations": "Utilities should adopt similar time-series forecasting for capital planning. Immediate focus should be on enhancing maintenance protocols for transformers and distribution lines, informed by the model's identified failure modes. ", "key words": "Power distribution, reliability forecasting, time-series analysis, infrastructure evaluation, ARIMA modelling", "contribution statement": "This paper presents a novel application of ARIMA modelling to forecast long-term power system reliability in a sub-Saharan context, providing a validated tool for infrastructure planning
Uwimana et al. (Fri,) studied this question.