The adoption of centralised maintenance depot systems is critical for the operational efficiency and safety of transport infrastructure in developing economies. However, there is a paucity of quantitative models to forecast and evaluate the rate of this adoption, hindering strategic asset management planning. This study develops and validates a time-series forecasting model to measure and project the adoption rate of engineered maintenance depot systems within the national transport sector, providing a tool for infrastructure investment planning. A longitudinal dataset of depot establishment and upgrade records was analysed. The core forecasting model is an autoregressive integrated moving average (ARIMA) formulation: yₜ = ₁ yₓ-₁ + ₁ ₓ-₁ + ₜ, where yₜ is the cumulative adoption count. Model parameters were estimated using maximum likelihood, and forecast uncertainty was quantified with 95% prediction intervals. The model forecasts a significant acceleration in adoption, with the projected proportion of the transport network served by formal depot systems rising from an estimated baseline to over 60% by the end of the forecast horizon. Parameter estimates were statistically significant at the 5% level, and diagnostic checks confirmed model robustness. The developed model provides a statistically robust tool for forecasting infrastructure system adoption, revealing a strong positive trajectory for maintenance depot integration within the transport sector. Infrastructure planners should integrate such quantitative forecasting models into long-term strategic plans. Further research should incorporate economic and climatic covariates to enhance model specificity. infrastructure management, maintenance depots, time-series analysis, ARIMA modelling, forecasting, transport engineering, asset management This paper presents a novel application of ARIMA time-series modelling to forecast the adoption rate of engineered maintenance systems, providing the first quantitative tool of its kind for the region's transport sector planners.
Moses Kato (2002) studied this question.