"background": "Transport maintenance depots are critical infrastructure for national economies, yet systematic evaluation of their long-term operational efficiency remains underdeveloped, particularly in sub-Saharan Africa. Existing assessments often lack robust, forward-looking analytical frameworks. ", "purpose and objectives": "This study aims to develop and methodologically evaluate a novel time-series forecasting model designed to quantify projected efficiency gains within transport maintenance depot systems. The objective is to provide a replicable analytical tool for infrastructure performance management. ", "methodology": "A hybrid forecasting model integrating Seasonal AutoRegressive Integrated Moving Average (SARIMA) components with exogenous infrastructural investment variables was constructed. The model, formalised as yt = \ + =1^{p\ yt-i + =1^q\ -j + =1^m\ Xt, k + \ₜ, was calibrated and validated using longitudinal operational data from multiple depots. Forecast robustness was assessed via rolling-window backtesting. ", "findings": "The model forecasts a significant upward trajectory in aggregate depot system efficiency, with a mean projected gain of 18. 7% over the forecast horizon (95% prediction interval: 14. 2% to 23. 1%). Diagnostic tests confirmed model stability, with all exogenous investment coefficients statistically significant at the 5% level. ", "conclusion": "The proposed forecasting model provides a statistically robust methodological framework for projecting efficiency improvements in transport maintenance systems. It successfully captures the dynamic relationship between strategic investment and long-term operational performance. ", "recommendations": "Infrastructure planners should adopt similar forecasting methodologies for strategic resource allocation and performance benchmarking. Future research should integrate real-time sensor data to enhance model granularity and predictive accuracy. ", "key words": "infrastructure management, predictive maintenance, SARIMA modelling, operational efficiency, forecasting, transport engineering", "contribution statement": "This paper presents a novel hybrid time-series model, uniquely tailored
Khamis et al. (Sat,) studied this question.