Bayesian Hierarchical Model for Cost-Effectiveness Assessment of Industrial Machinery Fleets in Senegal

This study addresses the need for a robust method to evaluate the cost-effectiveness of industrial machinery fleets in Senegal, considering the economic and operational complexities of such systems. A Bayesian hierarchical model was developed using Markov Chain Monte Carlo methods for analysing data on maintenance costs, operational efficiency, and productivity outcomes from a sample of industrial machinery fleets across Senegal. The BHM accounts for heterogeneity within and between fleets to estimate cost-effectiveness ratios with uncertainty quantification. The analysis revealed that fleet utilization rates significantly impact the overall cost-effectiveness (utilization rate above 70% resulted in an average cost-effectiveness ratio of $15,000 per year). The Bayesian hierarchical model demonstrated its utility for assessing and optimising industrial machinery fleets in Senegal, offering a more nuanced understanding of fleet performance compared to traditional methods. Industry practitioners should prioritise higher utilization rates and implement preventive maintenance strategies to enhance cost-effectiveness. Policy makers are encouraged to support the adoption of advanced fleet management techniques. The maintenance outcome was modelled as $Y_{it}=\beta_0+\beta_1X_{it}+u_i+\varepsilon_{it}$, with robustness checked using heteroskedasticity-consistent errors.