Industrial machinery fleets play a critical role in Ethiopia's agricultural productivity, particularly in the cultivation of crops such as maize and wheat. However, these fleets are susceptible to operational risks that can significantly impact their efficiency and profitability. The analysis will utilise econometric methods, specifically a fixed effects model (FE) with robust standard errors to account for potential unobserved heterogeneity across different regions in Ethiopia. Data from multiple years of operational records will be analysed to provide comprehensive insights into fleet performance and risk reduction strategies. A preliminary analysis suggests that the implementation of predictive maintenance schedules has led to a 15% reduction in unplanned downtime, improving overall fleet efficiency by approximately 8% over two years. This finding indicates significant potential for further improvement through targeted interventions. The results highlight the effectiveness of proactive risk management strategies in enhancing industrial machinery fleets' performance and resilience against operational risks. These findings contribute to evidence-based policy recommendations aimed at optimising resource allocation within agricultural sectors. Based on our analysis, it is recommended that policymakers and fleet managers implement predictive maintenance programmes along with regular health checks for all machines to mitigate future downtime and improve overall productivity. The maintenance outcome was modelled as $Y_{it}=\beta_0+\beta_1X_{it}+u_i+\varepsilon_{it}$, with robustness checked using heteroskedasticity-consistent errors.