Abstract
{ "background": "The adoption of new maintenance systems in public sector transport depots is often slow and poorly quantified, hindering effective asset management and infrastructure sustainability. Existing evaluations typically rely on post-implementation audits, lacking robust methods to measure causal adoption rates under real-world conditions.", "purpose and objectives": "This study aimed to develop and apply a novel randomised field trial methodology to measure the comparative adoption rates of two competing depot maintenance management systems within a public transport agency.", "methodology": "A randomised controlled trial was conducted across 42 depots, stratified by size and region. Depots were randomly assigned to implement either a technology-driven system (System T) or a process-oriented system (System P). Adoption was measured quarterly over 18 months via a composite index of procedural compliance and resource utilisation. The primary analysis used a linear mixed-effects model: $Y{ij} = \\beta0 + \\beta1 T{ij} + \\gamma X{ij} + uj + \\epsilon{ij}$, where $Y{ij}$ is the adoption index for depot $i$ in stratum $j$, $T{ij}$ is the treatment indicator, $X{ij}$ are covariates, and $u_j$ are random stratum effects.", "findings": "System P demonstrated a significantly higher final adoption rate (68%, 95% CI: 62 to 74) compared to System T (51%, 95% CI: 45 to 57). The treatment effect, adjusted for depot size and baseline maintenance backlog, was estimated at 16.5 percentage points (p < 0.01). The advantage for System P was consistent across strata but more pronounced in larger depots.", "conclusion": "The process-oriented maintenance system was adopted more readily than the technology-centric alternative in the trial context. The randomised field trial proved a viable method for generating comparative, causal evidence on implementation success within operational engineering environments.", "recommendations": "Transport agencies should prioritise process clarity and workforce engagement in