Abstract
{ "background": "The operational efficiency of industrial machinery fleets is a critical determinant of productivity and economic output in developing economies. In Uganda, a lack of robust, data-driven models for forecasting and evaluating efficiency gains has hindered strategic maintenance and capital investment planning within the industrial sector.", "purpose and objectives": "This study aimed to develop and methodologically evaluate a novel time-series forecasting model specifically designed to measure and predict efficiency gains within Uganda's industrial machinery fleets. The objective was to provide a validated tool for engineering asset management.", "methodology": "We developed an autoregressive integrated moving average with exogenous variables (ARIMAX) model, formalised as $yt = \\mu + \\sum{i=1}^{p}\\phii y{t-i} + \\epsilont + \\sum{i=1}^{q}\\thetai \\epsilon{t-i} + \\sum{i=1}^{r}\\betai x{t-i}$, where $yt$ represents fleet efficiency and $x_t$ represents maintenance expenditure. The model was trained and tested on a proprietary longitudinal dataset of fleet performance indicators, utilising robust standard errors for inference.", "findings": "The ARIMAX(2,1,1) model demonstrated strong predictive accuracy, with a mean absolute percentage error (MAPE) of 4.7% on the test set. A key finding was that a 10% increase in scheduled maintenance expenditure was associated with a 3.2% forecast improvement in fleet efficiency (95% CI: 2.1% to 4.3%).", "conclusion": "The proposed model provides a statistically reliable and technically sound methodological framework for forecasting machinery fleet efficiency, representing a significant advance for engineering management practices in the region.", "recommendations": "We recommend the adoption of this model by industrial operators and policymakers for proactive maintenance scheduling and budget allocation. Future work should integrate real-time sensor data to enhance model granularity.", "key words": "asset management, time-series analysis