"background": "The cost-effectiveness of railway maintenance depot systems is a critical yet under-researched factor in the operational efficiency of transport infrastructure in developing economies. Existing models often lack the temporal granularity to inform long-term strategic planning and budget allocation. ", "purpose and objectives": "This study aims to develop and evaluate a novel time-series forecasting model to measure and predict the cost-effectiveness of railway maintenance depot systems, providing a methodological framework for infrastructure investment decisions. ", "methodology": "A quantitative methodology was employed, integrating historical operational and financial data from multiple national depots. The core forecasting model is an autoregressive integrated moving average with exogenous variables (ARIMAX), specified as yt = \0 + =1^{p\ yt-i + \ + =1^q\ -i + =1^r\ xt, j, where yₜ is the cost-effectiveness index. Model diagnostics included checks for stationarity and residual autocorrelation. ", "findings": "The ARIMAX (2, 1, 1) model demonstrated robust predictive accuracy, with a 95% confidence interval for the one-step-ahead forecast error of ±4. 7%. The analysis projects a significant downward trend in systemic cost-effectiveness, with a forecasted decline of approximately 18% over the medium term if current maintenance strategies remain unchanged. ", "conclusion": "The developed model provides a statistically reliable tool for forecasting depot system performance, revealing a pressing need for strategic intervention to avert a substantial decline in cost-effectiveness. ", "recommendations": "Infrastructure planners should adopt this forecasting methodology for proactive budget formulation. Immediate investment in predictive maintenance technologies and depot workforce upskilling is recommended to counteract the identified negative trend. ", "key words": "time-series forecasting, maintenance depots, cost-effectiveness, railway infrastructure, ARIMAX, predictive modelling", "contribution statement": "This paper
Abebe et al. (Thu,) studied this question.