Abstract
{ "background": "The reliability of industrial machinery fleets is a critical determinant of productivity and economic output. In many industrialising nations, systematic assessment of fleet-wide reliability is hindered by a lack of longitudinal data and the challenge of isolating the effect of maintenance interventions from concurrent operational changes.", "purpose and objectives": "This paper develops and applies a quasi-experimental econometric model to quantify the causal impact of a structured preventive maintenance programme on the operational reliability of heavy machinery fleets within the country's industrial sector.", "methodology": "A difference-in-differences (DiD) model is specified as $Y{it} = \\beta0 + \\beta1 \\text{Treat}i + \\beta2 \\text{Post}t + \\delta (\\text{Treat}i \\times \\text{Post}t) + \\epsilon{it}$, where $Y{it}$ is the monthly uptime percentage for unit $i$ in period $t$. The parallel trends assumption is tested, and robust standard errors are clustered at the fleet level to account for serial correlation.", "findings": "The implementation of the preventive maintenance programme was associated with a statistically significant increase in mean fleet uptime. The DiD estimator, $\\delta$, was 7.3 percentage points (95% CI: 5.1, 9.5). This effect was robust to the inclusion of control variables for machine age and utilisation hours.", "conclusion": "The DiD framework provides a rigorous methodological approach for reliability assessment in settings with limited pre-intervention data, successfully isolating the causal effect of the maintenance programme from underlying temporal trends.", "recommendations": "Industrial asset managers should adopt quasi-experimental evaluation designs to validate the efficacy of reliability-centred maintenance strategies. Further research should apply the model to different machinery types and sectors.", "key words": "reliability engineering, maintenance optimisation, causal inference, econometric modelling, industrial assets", "contribution statement": "This paper presents a