African Computational Statistics (Technology/Maths) | 05 April 2012
Time-Series Forecasting Model Evaluation for System Reliability in Tanzanian Manufacturing Plants
S, i, m, b, a, M, w, a, k, a, b, a, k, a, ,, K, a, m, a, u, T, s, h, a, b, a, l, a, l, a
Abstract
Manufacturing plants in Tanzania face challenges in maintaining system reliability due to fluctuating operational conditions. A comparative analysis of various time-series models was conducted using historical data from multiple Tanzanian plants. Robust statistical techniques were applied to assess the predictive accuracy and reliability of these models. The ARIMA model demonstrated superior forecasting capabilities, with an average prediction error within ±5% for key system performance indicators (SPIs). The time-series forecasting model evaluated in this study significantly improved the reliability assessment of manufacturing systems in Tanzania. Manufacturers should integrate the recommended ARIMA-based forecast into their maintenance and operation plans to enhance overall plant efficiency. time-series analysis, system reliability, Tanzanian manufacturing, ARIMA model 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.