Abstract
{ "background": "The efficacy of maintenance depot systems is critical for transport infrastructure reliability and safety. In the South African context, systematic evaluations of how depot-level interventions translate into network-wide risk reduction have been limited, with a paucity of quantitative, hierarchical models linking operational factors to safety outcomes.", "purpose and objectives": "This case study aims to methodologically evaluate the relationship between depot system characteristics and transport risk reduction. Its objective is to quantify the impact of key depot-level operational and resource variables on provincial-level incident rates, controlling for network and traffic covariates.", "methodology": "A multilevel regression analysis was employed, treating depots as nested within provinces. The model, $y{ij} = \\beta{0} + \\beta{1}X{ij} + u{j} + e{ij}$, where $i$ denotes depots and $j$ provinces, was fitted to a longitudinal dataset of depot performance and incident records. Estimation used restricted maximum likelihood with robust standard errors to account for heteroskedasticity.", "findings": "Analysis indicates that a one-standard-deviation increase in depot preventive maintenance compliance is associated with a 15.2% reduction in major incident rates at the provincial level (95% CI: 11.8% to 18.6%). The intra-class correlation coefficient of 0.31 confirms significant variation attributable to provincial-level clustering.", "conclusion": "The findings demonstrate that depot system performance is a statistically significant predictor of broader transport network risk. The multilevel approach effectively captures the hierarchical structure of the infrastructure system.", "recommendations": "Infrastructure policy should mandate the collection of standardised, depot-level performance indicators. Investment should prioritise enhancing preventive maintenance capacity at the depot level, as it yields disproportionate system-wide safety benefits.", "key words": "multilevel regression, infrastructure maintenance, transport risk, depot systems, engineering management", "contribution statement": "This study provides a novel hierarchical modelling framework for infrastructure safety analytics, demonstrating a quantifiable