Abstract
{ "background": "The operational reliability of industrial machinery fleets is a critical determinant of productivity and economic output in developing economies. However, rigorous, quantitative methodologies for evaluating the systemic reliability of such fleets, particularly in West African contexts, are underdeveloped in the structural engineering literature.", "purpose and objectives": "This case study aims to develop and apply a robust econometric framework to methodologically evaluate the reliability of industrial machinery systems. The primary objective is to quantify the causal impact of a targeted maintenance intervention on fleet-wide operational uptime.", "methodology": "A quasi-experimental difference-in-differences (DiD) model is employed, analysing panel data from a major Senegalese industrial operator. The treatment group received a systematic predictive maintenance programme, while a comparable control group continued with routine scheduled maintenance. The core 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 operational availability for unit $i$ in period $t$. Inference is based on cluster-robust standard errors at the fleet subunit level.", "findings": "The intervention produced a statistically significant positive effect on machinery availability. The DiD estimator, $\\delta$, was 0.087 (95% CI: 0.051, 0.123), indicating an 8.7 percentage point increase in average operational uptime attributable to the new maintenance regime. This effect manifested after a three-month implementation lag.", "conclusion": "The applied DiD model provides a rigorous methodological framework for reliability assessment, demonstrating that data-driven predictive maintenance strategies can substantially enhance the operational performance of industrial fleets in the regional context.", "recommendations": "Industrial operators should integrate quasi-experimental evaluation designs into their asset management programmes. Further research should focus on calibrating the