African Journal of Epistemology and Indigenous Knowledge Systems (IKS) | 27 January 2009
Time-Series Forecasting Model Evaluation of Municipal Infrastructure Assets Systems in Tanzania,
K, a, s, i, n, i, n, g, o, M, w, e, s, i, m, a, l, e
Abstract
The study examines municipal infrastructure assets systems in Tanzania, focusing on their reliability over time. A comparative study using time-series forecasting models to assess system reliability across different municipalities in Tanzania. The evaluation incorporates multiple regression analysis as the primary statistical model. The findings indicate that the ARIMA (AutoRegressive Integrated Moving Average) model outperforms other models, predicting asset condition changes with a confidence interval of ±5% over a five-year horizon. This study provides insights into enhancing municipal infrastructure management in Tanzania through robust forecasting methodologies. Municipalities should implement the ARIMA model for regular asset assessments to improve system reliability and resource allocation efficiency. Tanzania, municipal infrastructure, time-series forecasting, ARIMA, reliability The maintenance outcome was modelled as $Y<em>{it}=\beta</em>0+\beta<em>1X</em>{it}+u<em>i+\varepsilon</em>{it}$, with robustness checked using heteroskedasticity-consistent errors.