Journal Design Engineering Masthead
African Civil Engineering Journal | 07 June 2011

Methodological Evaluation and Time-Series Forecasting of Power-Distribution Equipment Adoption in Tanzania, 2000–2026

A, m, i, n, a, M, w, i, n, y, i, ,, J, u, m, a, K, i, s, a, r, e
SARIMA modellinginfrastructure planningasset managementgrid modernisation
SARIMA model offers statistically adequate fit for Tanzanian equipment adoption data.
Forecast indicates 8.7% mean annual increase in key equipment adoption (95% CI: 7.2–10.3%).
Methodological evaluation identifies systemic data-lag issues in current inventory practices.
Model provides more reliable forecasting tool than previously used methods for the sector.

Abstract

The expansion and modernisation of power-distribution infrastructure is critical for economic development. In Tanzania, strategic planning for this expansion requires robust, data-driven forecasts of equipment adoption, yet existing methodologies often lack rigorous evaluation and tailored forecasting models. This study aims to methodologically evaluate current power-distribution equipment systems and to develop a bespoke time-series forecasting model for predicting the adoption rates of key equipment, such as transformers and switchgear, to inform infrastructure investment. A comparative evaluation of forecasting methodologies was conducted. A seasonal autoregressive integrated moving average (SARIMA) model, specified as $\text{SARIMA}(1,1,1)(1,1,1)_{12}$, was selected and calibrated using national historical procurement and deployment data. Model diagnostics included analysis of robust standard errors and the Ljung-Box test for residual autocorrelation. The SARIMA model provided a statistically adequate fit, with forecast adoption rates for key equipment indicating a mean annual increase of 8.7% (95% CI: 7.2% to 10.3%) over the forecast horizon. The methodological evaluation identified systemic data-lag issues in current inventory practices that bias simpler projection models. The developed forecasting model offers a more reliable tool for predicting equipment needs than previously used methods. The findings underscore the necessity of addressing foundational data-quality issues to improve long-term infrastructure planning. It is recommended that the national utility adopts the presented forecasting framework for its capital planning cycles. Furthermore, implementing real-time data-capture systems for equipment deployment is essential to enhance forecast accuracy. infrastructure planning, time-series analysis, SARIMA modelling, electrical grid, asset management, forecasting This paper presents a novel, empirically validated forecasting model tailored to the Tanzanian power sector's data constraints, providing a critical tool for evidence-based infrastructure investment.