{ "background": "Municipal infrastructure asset management in West Africa faces challenges in quantifying long-term performance and yield. Existing evaluation frameworks often lack robust, forward-looking methodologies tailored to the region's data-scarce and dynamic operational contexts. ", "purpose and objectives": "This study conducts a comparative evaluation of methodological approaches for asset yield assessment and develops a bespoke time-series forecasting model to project future infrastructure performance, with a focus on water and transport networks. ", "methodology": "A comparative analysis of deterministic and stochastic yield measurement techniques was performed. A seasonal autoregressive integrated moving average (SARIMA) model, specified as \ (B) \ (Bˢ) \\D Yt = \ (B) \ (Bˢ) \, was developed and calibrated using historical asset performance data. Model diagnostics included checks for residual autocorrelation and heteroskedasticity. ", "findings": "The SARIMA model outperformed conventional linear regression, with a mean absolute percentage error (MAPE) of 4. 7% on test data. Forecasts indicate a potential yield improvement of approximately 18% over the forecast horizon, with 95% prediction intervals showing a range of ±2. 3% around the central estimate. The comparative analysis revealed that stochastic methods provided more robust yield estimates under conditions of high operational variability. ", "conclusion": "The developed forecasting model provides a statistically sound tool for projecting infrastructure asset yield, offering superior accuracy to established deterministic methods. This enables more reliable long-term planning and resource allocation. ", "recommendations": "Municipal engineers and asset managers should adopt stochastic time-series forecasting for capital investment planning. Further research should integrate climate resilience factors into the model framework. ", "key words": "asset management, infrastructure yield, time-series forecasting, SARIMA, municipal engineering, West Africa", "contribution statement": "This paper presents a novel application of SARIMA modelling for infrastructure yield forecasting in a West African context, providing a validated tool that significantly reduces projection error compared to standard industry
Ndiaye et al. (Tue,) studied this question.