Journal Design Engineering Masthead
African Civil Engineering Journal | 24 September 2000

A Time-Series Forecasting Model for the Cost-Effectiveness of Power-Distribution Equipment in Tanzania

A Methodological Evaluation (2000–2026)
J, u, m, a, K, i, s, i, m, b, a, ,, N, e, e, m, a, M, w, a, m, b, e, n, e, ,, G, r, a, c, e, M, u, s, h, i
Asset ManagementForecasting ModelInfrastructure EconomicsUtility Planning
ARIMAX model achieves 8.7% MAPE in forecasting cost-effectiveness ratios.
Forecast uncertainty widens significantly beyond a five-year planning horizon.
Model provides a dynamic improvement over static historical analysis for asset management.
Framework is technically sound and operationally relevant for utility planning.

Abstract

{ "background": "The economic sustainability of power-distribution networks in sub-Saharan Africa is constrained by high capital costs and operational inefficiencies. Existing models for evaluating equipment cost-effectiveness often lack a robust, forward-looking component, limiting long-term infrastructure planning.", "purpose and objectives": "This paper presents a methodological evaluation of a novel time-series forecasting model designed to measure the cost-effectiveness of power-distribution equipment. The objective is to assess the model's predictive accuracy and utility for long-term capital planning.", "methodology": "The methodology employs an autoregressive integrated moving average with exogenous variables (ARIMAX) model, specified as $\\Delta yt = \\alpha + \\sum{i=1}^{p}\\phii \\Delta y{t-i} + \\sum{i=1}^{q}\\thetai \\epsilon{t-i} + \\sum{i=1}^{r}\\betai X{t-i} + \\epsilont$, where $yt$ is the cost-effectiveness ratio. Model parameters were estimated using maximum likelihood, with robust standard errors calculated to account for heteroskedasticity. Historical technical and financial data from a national utility were used for calibration and validation.", "findings": "The model demonstrates a statistically significant forecasting capability, with a mean absolute percentage error (MAPE) of 8.7% on the validation set. A key directional finding is that operational expenditure, rather than initial capital cost, is the dominant driver of long-term cost-ineffectiveness for transformers in the studied network. Forecast uncertainty, expressed via 95% prediction intervals, widens notably beyond a five-year horizon.", "conclusion": "The proposed ARIMAX framework provides a technically sound and operationally relevant method for forecasting the cost-effectiveness of distribution assets. It offers a material improvement over static, historical analysis for strategic asset management.", "recommendations": "Utilities should integrate such dynamic forecasting models into their asset management cycles. Further research should focus on incorporating climate resilience metrics as exogenous variables to enhance model robustness.",