A Time-Series Forecasting Model for Yield Improvement in Rwanda's Municipal Infrastructure Asset Systems: A Methodological Evaluation (2000–2026)

{ "background": "Municipal infrastructure asset systems in Rwanda face challenges in long-term financial sustainability and performance measurement. Existing asset management frameworks often lack robust, data-driven methodologies for forecasting operational yield, which is critical for capital planning and maintenance scheduling.", "purpose and objectives": "This article presents a methodological evaluation of a novel time-series forecasting model designed to measure and project yield improvement within these asset systems. The primary objective is to detail the model's construction, calibration, and validation protocol.", "methodology": "The methodology integrates an autoregressive integrated moving average with exogenous variables (ARIMAX) model, specified as Yt = \μ + \∑i=1ᵖ\ Yt-i + \∑j=1q\ \εt-j + \∑k=1ʳ\ Xt,k + \, where $Yt$ is the yield metric. Model parameters were estimated using maximum likelihood, with inference on coefficients based on robust standard errors to account for heteroskedasticity. A structured back-testing procedure on historical data was employed for validation.", "findings": "The methodological evaluation demonstrates that the model provides a statistically significant improvement in forecast accuracy over a naive benchmark, with a mean absolute percentage error reduction of approximately 18% in out-of-sample tests. Parameter estimates for key maintenance expenditure variables were positive and significant at the 95% confidence level, indicating their material impact on yield trajectories.", "conclusion": "The proposed ARIMAX-based methodology offers a rigorous, replicable framework for forecasting infrastructure yield, enhancing the evidence base for strategic asset management decisions.", "recommendations": "It is recommended that municipal engineers and planners adopt similar stochastic forecasting techniques, incorporating regular model updating with new data and sensitivity analysis around key exogenous inputs.", "key words": "asset management, infrastructure yield, time-series analysis, ARIMAX, forecasting, municipal engineering", "contribution statement": "This paper provides