Industrial machinery fleets in Ghana face challenges related to maintenance scheduling and reliability assessment, leading to increased downtime and operational costs. The methodology involves collecting historical maintenance records of industrial machinery, applying Box-Jenkins ARIMA (AutoRegressive Integrated Moving Average) model for forecasting, and incorporating robust standard errors to quantify prediction uncertainty. A significant proportion (60%) of machine failures could be predicted with an accuracy rate within ±15% using the ARIMA model, indicating a reliable approach for system reliability measurement in Ghanaian industrial settings. The time-series forecasting model successfully forecasts future maintenance needs and failure occurrences, offering valuable insights for improving fleet operational efficiency and cost management. Adoption of this predictive model can help industrial machinery owners optimise their maintenance schedules, reduce downtime, and enhance overall system reliability in Ghanaian industries. Industrial Machinery, Fleets Systems, Time-Series Forecasting, System Reliability, ARIMA Model The maintenance outcome was modelled as Y₈ₓ=₀+₁X₈ₓ+uᵢ+₈ₓ, with robustness checked using heteroskedasticity-consistent errors.
Agyei et al. (Mon,) studied this question.