Municipal infrastructure asset management in developing economies often lacks robust, data-driven methodologies for forecasting performance and yield. Existing approaches in the region have been critiqued for their reliance on static assessments, which fail to account for temporal dynamics and degradation patterns. This report aims to methodologically evaluate current municipal infrastructure asset systems and to develop a predictive time-series model for forecasting asset yield, with the objective of providing a tool for long-term infrastructure planning and investment prioritisation. A quantitative analysis was conducted using a longitudinal dataset of key municipal asset performance indicators. The core forecasting model employs an autoregressive integrated moving average (ARIMA) framework, specified as Yₜ = + ₁ Yₓ-₁ + ₁ ₓ-₁ + ₜ, where Yₜ is the asset yield at time t. Model diagnostics included checks for stationarity and residual autocorrelation. The model forecasts a significant upward trend in aggregate municipal infrastructure yield over the forecast horizon, with a predicted mean increase of approximately 18. 7% from the baseline period. Forecast uncertainty, represented by a 95% prediction interval, widens in later periods, indicating increased volatility in long-term projections. The developed time-series model provides a statistically sound and practical methodology for projecting infrastructure yield, offering a substantial improvement over static evaluation techniques commonly used. Municipal authorities should adopt dynamic, model-based forecasting for asset management planning. Further research should integrate granular asset condition data to refine predictive accuracy and account for climate resilience factors. Infrastructure asset management, time-series forecasting, ARIMA modelling, municipal engineering, predictive maintenance This report provides a novel application of ARIMA modelling to forecast long-term municipal infrastructure yield, generating a specific, evidence-based projection for strategic planning.
Asante et al. (Sun,) studied this question.