A Quasi-Experimental Design for the Cost-Effectiveness Evaluation of Railway Maintenance Depot Systems in Uganda

{ "background": "The cost-effectiveness of railway maintenance depot systems in developing economies is poorly understood, with a lack of robust methodological frameworks for evaluation under real-world constraints.", "purpose and objectives": "This study aimed to develop and apply a novel quasi-experimental design to rigorously evaluate the cost-effectiveness of different maintenance depot systems, focusing on operational efficiency and lifecycle costs.", "methodology": "A quasi-experimental design was implemented, comparing a treatment group of depots under a centralised, technology-augmented system against a control group using decentralised, conventional practices. Cost and performance data were collected over an operational period. The primary analysis employed a difference-in-differences model: Yit = \β0 + \β1 + \β2 + \β3 ( \× ) + \εit, where robust standard errors were clustered at the depot level.", "findings": "The centralised system demonstrated a statistically significant reduction in average maintenance cost per kilometre, with an estimated treatment effect of a 17.5% decrease (95% CI: 12.1% to 22.9%). This was achieved without a concurrent reduction in asset availability, which remained stable across both groups.", "conclusion": "The applied quasi-experimental design proved viable for engineering cost-effectiveness analysis in a resource-constrained setting. The centralised maintenance system was found to be more cost-effective than the decentralised model for the studied network.", "recommendations": "Railway operators should consider phased implementation of centralised, technology-integrated depot systems. Further research should apply this methodological framework to longitudinal data to assess long-term structural performance impacts.", "key words": "quasi-experimental design, cost-effectiveness, railway maintenance, depot systems, infrastructure management, difference-in-differences", "contribution statement": "This paper provides a novel methodological framework for causal