"background": "Municipal infrastructure asset management in developing nations is challenged by limited data and predictive tools for long-term risk assessment. Current practices often rely on static condition assessments, lacking robust, data-driven forecasting to inform maintenance and capital renewal strategies. ", "purpose and objectives": "This data descriptor presents and methodologically evaluates a novel time-series forecasting model designed to quantify risk reduction for municipal infrastructure assets. The objective is to provide a replicable analytical framework for predicting asset deterioration and evaluating intervention scenarios. ", "methodology": "The methodology employs an autoregressive integrated moving average (ARIMA) model, specified as \ᵈ yt = c + =1^{p\ \ᵈ yt-i + =1^q\ -j + \, where \ᵈ is the differencing operator. Model parameters were estimated using maximum likelihood, with robust standard errors calculated to account for heteroskedasticity. The framework integrates asset condition, failure consequence, and intervention cost data. ", "findings": "The methodological evaluation indicates the model's utility in projecting asset condition trajectories and quantifying risk reduction from planned investments. A key finding is that targeted rehabilitation of a specific asset class is projected to reduce its associated critical failure risk by approximately 40% over the forecast horizon, with a 95% confidence interval of 35%, 45%. ", "conclusion": "The proposed time-series model provides a statistically grounded, practical tool for infrastructure risk forecasting. It represents a significant advancement from descriptive condition reporting towards predictive, scenario-based asset management. ", "recommendations": "Adoption of this forecasting approach is recommended for municipal engineers and planners to prioritise capital works. Future work should focus on integrating the model with geospatial information systems and refining the cost-risk parameters with localised data. ", "key words": "asset management, infrastructure risk, time-series analysis, forecasting, ARIMA, municipal engineering, predictive maintenance
Amina Mwinyi (Fri,) studied this question.