{ "background": "Industrial machinery fleets are critical for national infrastructure development, yet systematic methodologies for evaluating their operational yield in developing economies are scarce. Existing approaches often lack the longitudinal rigour needed to isolate technical efficiency gains from other factors.", "purpose and objectives": "This study aims to develop and apply a robust panel-data econometric framework to measure and explain yield improvement within industrial machinery fleets. The objective is to identify key technical and operational determinants of efficiency gains.", "methodology": "A novel unbalanced panel dataset was constructed from maintenance logs, operational reports, and fuel consumption records for a fleet of heavy earth-moving equipment. The core analysis employs a fixed-effects estimation model: $Y{it} = \\alphai + \\beta1 X{1,it} + \\beta2 X{2,it} + \\epsilon{it}$, where $Y{it}$ is the yield metric for machine $i$ in period $t$, $\\alpha_i$ captures unobserved machine heterogeneity, and $X$ variables represent operational parameters. Inference is based on cluster-robust standard errors.", "findings": "The model indicates that scheduled predictive maintenance interventions are a significant driver of yield, associated with a mean increase of 17.3% in volumetric output per fuel unit (95% CI: 12.1% to 22.5%). In contrast, operator experience beyond a threshold showed diminishing returns. The fixed effects accounted for a substantial portion of total variation.", "conclusion": "The methodological framework successfully isolates specific technical interventions contributing to yield improvement, moving beyond descriptive fleet management. The findings confirm that data-structured maintenance regimes are more impactful for yield than blanket operator training programmes.", "recommendations": "Fleet managers should prioritise investment in condition-monitoring systems to enable predictive maintenance scheduling. Policy should support the development of standardised digital record-keeping protocols to facilitate similar longitudinal analyses across the sector.", "key words": "Panel data analysis, fixed effects model, industrial machinery, fleet management,